Aren
/
NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
NetBSD/gnu/dist/gawk/doc/gawk.texi

20821 lines
739 KiB
Plaintext
Raw Blame History

\input texinfo @c -*-texinfo-*-
@c %**start of header (This is for running Texinfo on a region.)
@setfilename gawk.info
@settitle The GNU Awk User's Guide
@c %**end of header (This is for running Texinfo on a region.)
@c inside ifinfo for older versions of texinfo.tex
@ifinfo
@c I hope this is the right category
@dircategory Programming Languages
@direntry
* Gawk: (gawk.info). A Text Scanning and Processing Language.
@end direntry
@end ifinfo
@c @set xref-automatic-section-title
@c @set DRAFT
@c The following information should be updated here only!
@c This sets the edition of the document, the version of gawk it
@c applies to, and when the document was updated.
@set TITLE Effective AWK Programming
@set SUBTITLE A User's Guide for GNU Awk
@set PATCHLEVEL 3
@set EDITION 1.0.@value{PATCHLEVEL}
@set VERSION 3.0
@set UPDATE-MONTH February 1997
@iftex
@set DOCUMENT book
@end iftex
@ifinfo
@set DOCUMENT Info file
@end ifinfo
@ignore
Some comments on the layout for TeX.
1. Use at least texinfo.tex 2.159. It contains fixes that
are needed to get the footings for draft mode to not appear.
2. I have done A LOT of work to make this look good. There are `@page' commands
and use of `@group ... @end group' in a number of places. If you muck
with anything, it's your responsibility not to break the layout.
@end ignore
@c merge the function and variable indexes into the concept index
@ifinfo
@synindex fn cp
@synindex vr cp
@end ifinfo
@iftex
@syncodeindex fn cp
@syncodeindex vr cp
@end iftex
@c If "finalout" is commented out, the printed output will show
@c black boxes that mark lines that are too long. Thus, it is
@c unwise to comment it out when running a master in case there are
@c overfulls which are deemed okay.
@ifclear DRAFT
@iftex
@finalout
@end iftex
@end ifclear
@smallbook
@iftex
@c @cropmarks
@end iftex
@ifinfo
This file documents @code{awk}, a program that you can use to select
particular records in a file and perform operations upon them.
This is Edition @value{EDITION} of @cite{@value{TITLE}},
for the @value{VERSION}.@value{PATCHLEVEL} version of the GNU implementation of AWK.
Copyright (C) 1989, 1991, 92, 93, 96, 97 Free Software Foundation, Inc.
Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.
@ignore
Permission is granted to process this file through TeX and print the
results, provided the printed document carries copying permission
notice identical to this one except for the removal of this paragraph
(this paragraph not being relevant to the printed manual).
@end ignore
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the entire
resulting derived work is distributed under the terms of a permission
notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation approved
by the Foundation.
@end ifinfo
@setchapternewpage odd
@titlepage
@title @value{TITLE}
@subtitle @value{SUBTITLE}
@subtitle Edition @value{EDITION}
@subtitle @value{UPDATE-MONTH}
@author Arnold D. Robbins
@ignore
@sp 1
@author Based on @cite{The GAWK Manual},
@author by Robbins, Close, Rubin, and Stallman
@end ignore
@c Include the Distribution inside the titlepage environment so
@c that headings are turned off. Headings on and off do not work.
@page
@vskip 0pt plus 1filll
@ifset LEGALJUNK
The programs and applications presented in this book have been
included for their instructional value. They have been tested with care,
but are not guaranteed for any particular purpose. The publisher does not
offer any warranties or representations, nor does it accept any
liabilities with respect to the programs or applications.
So there.
@sp 2
UNIX is a registered trademark of X/Open, Ltd. @*
Microsoft, MS, and MS-DOS are registered trademarks, and Windows is a
trademark of Microsoft Corporation in the United States and other
countries. @*
Atari, 520ST, 1040ST, TT, STE, Mega, and Falcon are registered trademarks
or trademarks of Atari Corporation. @*
DEC, Digital, OpenVMS, ULTRIX, and VMS, are trademarks of Digital Equipment
Corporation. @*
@end ifset
``To boldly go where no man has gone before'' is a
Registered Trademark of Paramount Pictures Corporation. @*
@c sorry, i couldn't resist
@sp 3
Copyright @copyright{} 1989, 1991, 92, 93, 96, 97 Free Software Foundation, Inc.
@sp 2
This is Edition @value{EDITION} of @cite{@value{TITLE}}, @*
for the @value{VERSION}.@value{PATCHLEVEL} (or later) version of the GNU implementation of AWK.
@sp 2
@center Published jointly by:
@multitable {Specialized Systems Consultants, Inc. (SSC)} {Boston, MA 02111-1307 USA}
@item Specialized Systems Consultants, Inc. (SSC) @tab Free Software Foundation
@item PO Box 55549 @tab 59 Temple Place --- Suite 330
@item Seattle, WA 98155 USA @tab Boston, MA 02111-1307 USA
@item Phone: +1-206-782-7733 @tab Phone: +1-617-542-5942
@item Fax: +1-206-782-7191 @tab Fax: +1-617-542-2652
@item E-mail: @code{sales@@ssc.com} @tab E-mail: @code{gnu@@prep.ai.mit.edu}
@item URL: @code{http://www.ssc.com/} @tab URL: @code{http://www.fsf.org/}
@end multitable
@sp 1
@c this ISBN can change! Check with SSC
@c This one is correct for gawk 3.0 and edition 1.0 from the FSF
@c ISBN 1-882114-26-4 @*
@c This one is correct for gawk 3.0.3 and edition 1.0.3 from SSC
ISBN 1-57831-000-8 @*
Permission is granted to make and distribute verbatim copies of
this manual provided the copyright notice and this permission notice
are preserved on all copies.
Permission is granted to copy and distribute modified versions of this
manual under the conditions for verbatim copying, provided that the entire
resulting derived work is distributed under the terms of a permission
notice identical to this one.
Permission is granted to copy and distribute translations of this manual
into another language, under the above conditions for modified versions,
except that this permission notice may be stated in a translation approved
by the Foundation.
@sp 2
@c Cover art by Etienne Suvasa.
Cover art by Amy Wells Wood.
@end titlepage
@c Thanks to Bob Chassell for directions on doing dedications.
@iftex
@headings off
@page
@w{ }
@sp 9
@center @i{To Miriam, for making me complete.}
@sp 1
@center @i{To Chana, for the joy you bring us.}
@sp 1
@center @i{To Rivka, for the exponential increase.}
@sp 1
@center @i{To Nachum, for the added dimension.}
@page
@w{ }
@page
@headings on
@end iftex
@iftex
@headings off
@evenheading @thispage@ @ @ @strong{@value{TITLE}} @| @|
@oddheading @| @| @strong{@thischapter}@ @ @ @thispage
@ifset DRAFT
@evenfooting @today{} @| @emph{DRAFT!} @| Please Do Not Redistribute
@oddfooting Please Do Not Redistribute @| @emph{DRAFT!} @| @today{}
@end ifset
@end iftex
@ifinfo
@node Top, Preface, (dir), (dir)
@top General Introduction
@c Preface or Licensing nodes should come right after the Top
@c node, in `unnumbered' sections, then the chapter, `What is gawk'.
This file documents @code{awk}, a program that you can use to select
particular records in a file and perform operations upon them.
This is Edition @value{EDITION} of @cite{@value{TITLE}}, @*
for the @value{VERSION}.@value{PATCHLEVEL} version of the GNU implementation @*
of AWK.
@end ifinfo
@menu
* Preface:: What this @value{DOCUMENT} is about; brief
history and acknowledgements.
* What Is Awk:: What is the @code{awk} language; using this
@value{DOCUMENT}.
* Getting Started:: A basic introduction to using @code{awk}. How
to run an @code{awk} program. Command line
syntax.
* One-liners:: Short, sample @code{awk} programs.
* Regexp:: All about matching things using regular
expressions.
* Reading Files:: How to read files and manipulate fields.
* Printing:: How to print using @code{awk}. Describes the
@code{print} and @code{printf} statements.
Also describes redirection of output.
* Expressions:: Expressions are the basic building blocks of
statements.
* Patterns and Actions:: Overviews of patterns and actions.
* Statements:: The various control statements are described
in detail.
* Built-in Variables:: Built-in Variables
* Arrays:: The description and use of arrays. Also
includes array-oriented control statements.
* Built-in:: The built-in functions are summarized here.
* User-defined:: User-defined functions are described in
detail.
* Invoking Gawk:: How to run @code{gawk}.
* Library Functions:: A Library of @code{awk} Functions.
* Sample Programs:: Many @code{awk} programs with complete
explanations.
* Language History:: The evolution of the @code{awk} language.
* Gawk Summary:: @code{gawk} Options and Language Summary.
* Installation:: Installing @code{gawk} under various operating
systems.
* Notes:: Something about the implementation of
@code{gawk}.
* Glossary:: An explanation of some unfamiliar terms.
* Copying:: Your right to copy and distribute @code{gawk}.
* Index:: Concept and Variable Index.
* History:: The history of @code{gawk} and @code{awk}.
* Manual History:: Brief history of the GNU project and this
@value{DOCUMENT}.
* Acknowledgements:: Acknowledgements.
* This Manual:: Using this @value{DOCUMENT}. Includes sample
input files that you can use.
* Conventions:: Typographical Conventions.
* Sample Data Files:: Sample data files for use in the @code{awk}
programs illustrated in this @value{DOCUMENT}.
* Names:: What name to use to find @code{awk}.
* Running gawk:: How to run @code{gawk} programs; includes
command line syntax.
* One-shot:: Running a short throw-away @code{awk} program.
* Read Terminal:: Using no input files (input from terminal
instead).
* Long:: Putting permanent @code{awk} programs in
files.
* Executable Scripts:: Making self-contained @code{awk} programs.
* Comments:: Adding documentation to @code{gawk} programs.
* Very Simple:: A very simple example.
* Two Rules:: A less simple one-line example with two rules.
* More Complex:: A more complex example.
* Statements/Lines:: Subdividing or combining statements into
lines.
* Other Features:: Other Features of @code{awk}.
* When:: When to use @code{gawk} and when to use other
things.
* Regexp Usage:: How to Use Regular Expressions.
* Escape Sequences:: How to write non-printing characters.
* Regexp Operators:: Regular Expression Operators.
* GNU Regexp Operators:: Operators specific to GNU software.
* Case-sensitivity:: How to do case-insensitive matching.
* Leftmost Longest:: How much text matches.
* Computed Regexps:: Using Dynamic Regexps.
* Records:: Controlling how data is split into records.
* Fields:: An introduction to fields.
* Non-Constant Fields:: Non-constant Field Numbers.
* Changing Fields:: Changing the Contents of a Field.
* Field Separators:: The field separator and how to change it.
* Basic Field Splitting:: How fields are split with single characters or
simple strings.
* Regexp Field Splitting:: Using regexps as the field separator.
* Single Character Fields:: Making each character a separate field.
* Command Line Field Separator:: Setting @code{FS} from the command line.
* Field Splitting Summary:: Some final points and a summary table.
* Constant Size:: Reading constant width data.
* Multiple Line:: Reading multi-line records.
* Getline:: Reading files under explicit program control
using the @code{getline} function.
* Getline Intro:: Introduction to the @code{getline} function.
* Plain Getline:: Using @code{getline} with no arguments.
* Getline/Variable:: Using @code{getline} into a variable.
* Getline/File:: Using @code{getline} from a file.
* Getline/Variable/File:: Using @code{getline} into a variable from a
file.
* Getline/Pipe:: Using @code{getline} from a pipe.
* Getline/Variable/Pipe:: Using @code{getline} into a variable from a
pipe.
* Getline Summary:: Summary Of @code{getline} Variants.
* Print:: The @code{print} statement.
* Print Examples:: Simple examples of @code{print} statements.
* Output Separators:: The output separators and how to change them.
* OFMT:: Controlling Numeric Output With @code{print}.
* Printf:: The @code{printf} statement.
* Basic Printf:: Syntax of the @code{printf} statement.
* Control Letters:: Format-control letters.
* Format Modifiers:: Format-specification modifiers.
* Printf Examples:: Several examples.
* Redirection:: How to redirect output to multiple files and
pipes.
* Special Files:: File name interpretation in @code{gawk}.
@code{gawk} allows access to inherited file
descriptors.
* Close Files And Pipes:: Closing Input and Output Files and Pipes.
* Constants:: String, numeric, and regexp constants.
* Scalar Constants:: Numeric and string constants.
* Regexp Constants:: Regular Expression constants.
* Using Constant Regexps:: When and how to use a regexp constant.
* Variables:: Variables give names to values for later use.
* Using Variables:: Using variables in your programs.
* Assignment Options:: Setting variables on the command line and a
summary of command line syntax. This is an
advanced method of input.
* Conversion:: The conversion of strings to numbers and vice
versa.
* Arithmetic Ops:: Arithmetic operations (@samp{+}, @samp{-},
etc.)
* Concatenation:: Concatenating strings.
* Assignment Ops:: Changing the value of a variable or a field.
* Increment Ops:: Incrementing the numeric value of a variable.
* Truth Values:: What is ``true'' and what is ``false''.
* Typing and Comparison:: How variables acquire types, and how this
affects comparison of numbers and strings with
@samp{<}, etc.
* Boolean Ops:: Combining comparison expressions using boolean
operators @samp{||} (``or''), @samp{&&}
(``and'') and @samp{!} (``not'').
* Conditional Exp:: Conditional expressions select between two
subexpressions under control of a third
subexpression.
* Function Calls:: A function call is an expression.
* Precedence:: How various operators nest.
* Pattern Overview:: What goes into a pattern.
* Kinds of Patterns:: A list of all kinds of patterns.
* Regexp Patterns:: Using regexps as patterns.
* Expression Patterns:: Any expression can be used as a pattern.
* Ranges:: Pairs of patterns specify record ranges.
* BEGIN/END:: Specifying initialization and cleanup rules.
* Using BEGIN/END:: How and why to use BEGIN/END rules.
* I/O And BEGIN/END:: I/O issues in BEGIN/END rules.
* Empty:: The empty pattern, which matches every record.
* Action Overview:: What goes into an action.
* If Statement:: Conditionally execute some @code{awk}
statements.
* While Statement:: Loop until some condition is satisfied.
* Do Statement:: Do specified action while looping until some
condition is satisfied.
* For Statement:: Another looping statement, that provides
initialization and increment clauses.
* Break Statement:: Immediately exit the innermost enclosing loop.
* Continue Statement:: Skip to the end of the innermost enclosing
loop.
* Next Statement:: Stop processing the current input record.
* Nextfile Statement:: Stop processing the current file.
* Exit Statement:: Stop execution of @code{awk}.
* User-modified:: Built-in variables that you change to control
@code{awk}.
* Auto-set:: Built-in variables where @code{awk} gives you
information.
* ARGC and ARGV:: Ways to use @code{ARGC} and @code{ARGV}.
* Array Intro:: Introduction to Arrays
* Reference to Elements:: How to examine one element of an array.
* Assigning Elements:: How to change an element of an array.
* Array Example:: Basic Example of an Array
* Scanning an Array:: A variation of the @code{for} statement. It
loops through the indices of an array's
existing elements.
* Delete:: The @code{delete} statement removes an element
from an array.
* Numeric Array Subscripts:: How to use numbers as subscripts in
@code{awk}.
* Uninitialized Subscripts:: Using Uninitialized variables as subscripts.
* Multi-dimensional:: Emulating multi-dimensional arrays in
@code{awk}.
* Multi-scanning:: Scanning multi-dimensional arrays.
* Calling Built-in:: How to call built-in functions.
* Numeric Functions:: Functions that work with numbers, including
@code{int}, @code{sin} and @code{rand}.
* String Functions:: Functions for string manipulation, such as
@code{split}, @code{match}, and
@code{sprintf}.
* I/O Functions:: Functions for files and shell commands.
* Time Functions:: Functions for dealing with time stamps.
* Definition Syntax:: How to write definitions and what they mean.
* Function Example:: An example function definition and what it
does.
* Function Caveats:: Things to watch out for.
* Return Statement:: Specifying the value a function returns.
* Options:: Command line options and their meanings.
* Other Arguments:: Input file names and variable assignments.
* AWKPATH Variable:: Searching directories for @code{awk} programs.
* Obsolete:: Obsolete Options and/or features.
* Undocumented:: Undocumented Options and Features.
* Known Bugs:: Known Bugs in @code{gawk}.
* Portability Notes:: What to do if you don't have @code{gawk}.
* Nextfile Function:: Two implementations of a @code{nextfile}
function.
* Assert Function:: A function for assertions in @code{awk}
programs.
* Round Function:: A function for rounding if @code{sprintf} does
not do it correctly.
* Ordinal Functions:: Functions for using characters as numbers and
vice versa.
* Join Function:: A function to join an array into a string.
* Mktime Function:: A function to turn a date into a timestamp.
* Gettimeofday Function:: A function to get formatted times.
* Filetrans Function:: A function for handling data file transitions.
* Getopt Function:: A function for processing command line
arguments.
* Passwd Functions:: Functions for getting user information.
* Group Functions:: Functions for getting group information.
* Library Names:: How to best name private global variables in
library functions.
* Clones:: Clones of common utilities.
* Cut Program:: The @code{cut} utility.
* Egrep Program:: The @code{egrep} utility.
* Id Program:: The @code{id} utility.
* Split Program:: The @code{split} utility.
* Tee Program:: The @code{tee} utility.
* Uniq Program:: The @code{uniq} utility.
* Wc Program:: The @code{wc} utility.
* Miscellaneous Programs:: Some interesting @code{awk} programs.
* Dupword Program:: Finding duplicated words in a document.
* Alarm Program:: An alarm clock.
* Translate Program:: A program similar to the @code{tr} utility.
* Labels Program:: Printing mailing labels.
* Word Sorting:: A program to produce a word usage count.
* History Sorting:: Eliminating duplicate entries from a history
file.
* Extract Program:: Pulling out programs from Texinfo source
files.
* Simple Sed:: A Simple Stream Editor.
* Igawk Program:: A wrapper for @code{awk} that includes files.
* V7/SVR3.1:: The major changes between V7 and System V
Release 3.1.
* SVR4:: Minor changes between System V Releases 3.1
and 4.
* POSIX:: New features from the POSIX standard.
* BTL:: New features from the Bell Laboratories
version of @code{awk}.
* POSIX/GNU:: The extensions in @code{gawk} not in POSIX
@code{awk}.
* Command Line Summary:: Recapitulation of the command line.
* Language Summary:: A terse review of the language.
* Variables/Fields:: Variables, fields, and arrays.
* Fields Summary:: Input field splitting.
* Built-in Summary:: @code{awk}'s built-in variables.
* Arrays Summary:: Using arrays.
* Data Type Summary:: Values in @code{awk} are numbers or strings.
* Rules Summary:: Patterns and Actions, and their component
parts.
* Pattern Summary:: Quick overview of patterns.
* Regexp Summary:: Quick overview of regular expressions.
* Actions Summary:: Quick overview of actions.
* Operator Summary:: @code{awk} operators.
* Control Flow Summary:: The control statements.
* I/O Summary:: The I/O statements.
* Printf Summary:: A summary of @code{printf}.
* Special File Summary:: Special file names interpreted internally.
* Built-in Functions Summary:: Built-in numeric and string functions.
* Time Functions Summary:: Built-in time functions.
* String Constants Summary:: Escape sequences in strings.
* Functions Summary:: Defining and calling functions.
* Historical Features:: Some undocumented but supported ``features''.
* Gawk Distribution:: What is in the @code{gawk} distribution.
* Getting:: How to get the distribution.
* Extracting:: How to extract the distribution.
* Distribution contents:: What is in the distribution.
* Unix Installation:: Installing @code{gawk} under various versions
of Unix.
* Quick Installation:: Compiling @code{gawk} under Unix.
* Configuration Philosophy:: How it's all supposed to work.
* VMS Installation:: Installing @code{gawk} on VMS.
* VMS Compilation:: How to compile @code{gawk} under VMS.
* VMS Installation Details:: How to install @code{gawk} under VMS.
* VMS Running:: How to run @code{gawk} under VMS.
* VMS POSIX:: Alternate instructions for VMS POSIX.
* PC Installation:: Installing and Compiling @code{gawk} on MS-DOS
and OS/2
* Atari Installation:: Installing @code{gawk} on the Atari ST.
* Atari Compiling:: Compiling @code{gawk} on Atari
* Atari Using:: Running @code{gawk} on Atari
* Amiga Installation:: Installing @code{gawk} on an Amiga.
* Bugs:: Reporting Problems and Bugs.
* Other Versions:: Other freely available @code{awk}
implementations.
* Compatibility Mode:: How to disable certain @code{gawk} extensions.
* Additions:: Making Additions To @code{gawk}.
* Adding Code:: Adding code to the main body of @code{gawk}.
* New Ports:: Porting @code{gawk} to a new operating system.
* Future Extensions:: New features that may be implemented one day.
* Improvements:: Suggestions for improvements by volunteers.
@end menu
@c dedication for Info file
@ifinfo
@center To Miriam, for making me complete.
@sp 1
@center To Chana, for the joy you bring us.
@sp 1
@center To Rivka, for the exponential increase.
@sp 1
@center To Nachum, for the added dimension.
@end ifinfo
@node Preface, What Is Awk, Top, Top
@unnumbered Preface
@c I saw a comment somewhere that the preface should describe the book itself,
@c and the introduction should describe what the book covers.
This @value{DOCUMENT} teaches you about the @code{awk} language and
how you can use it effectively. You should already be familiar with basic
system commands, such as @code{cat} and @code{ls},@footnote{These commands
are available on POSIX compliant systems, as well as on traditional Unix
based systems. If you are using some other operating system, you still need to
be familiar with the ideas of I/O redirection and pipes.} and basic shell
facilities, such as Input/Output (I/O) redirection and pipes.
Implementations of the @code{awk} language are available for many different
computing environments. This @value{DOCUMENT}, while describing the @code{awk} language
in general, also describes a particular implementation of @code{awk} called
@code{gawk} (which stands for ``GNU Awk''). @code{gawk} runs on a broad range
of Unix systems, ranging from 80386 PC-based computers, up through large scale
systems, such as Crays. @code{gawk} has also been ported to MS-DOS and
OS/2 PC's, Atari and Amiga micro-computers, and VMS.
@menu
* History:: The history of @code{gawk} and @code{awk}.
* Manual History:: Brief history of the GNU project and this
@value{DOCUMENT}.
* Acknowledgements:: Acknowledgements.
@end menu
@node History, Manual History, Preface, Preface
@unnumberedsec History of @code{awk} and @code{gawk}
@cindex acronym
@cindex history of @code{awk}
@cindex Aho, Alfred
@cindex Weinberger, Peter
@cindex Kernighan, Brian
@cindex old @code{awk}
@cindex new @code{awk}
The name @code{awk} comes from the initials of its designers: Alfred V.@:
Aho, Peter J.@: Weinberger, and Brian W.@: Kernighan. The original version of
@code{awk} was written in 1977 at AT&T Bell Laboratories.
In 1985 a new version made the programming
language more powerful, introducing user-defined functions, multiple input
streams, and computed regular expressions.
This new version became generally available with Unix System V Release 3.1.
The version in System V Release 4 added some new features and also cleaned
up the behavior in some of the ``dark corners'' of the language.
The specification for @code{awk} in the POSIX Command Language
and Utilities standard further clarified the language based on feedback
from both the @code{gawk} designers, and the original Bell Labs @code{awk}
designers.
The GNU implementation, @code{gawk}, was written in 1986 by Paul Rubin
and Jay Fenlason, with advice from Richard Stallman. John Woods
contributed parts of the code as well. In 1988 and 1989, David Trueman, with
help from Arnold Robbins, thoroughly reworked @code{gawk} for compatibility
with the newer @code{awk}. Current development focuses on bug fixes,
performance improvements, standards compliance, and occasionally, new features.
@node Manual History, Acknowledgements, History, Preface
@unnumberedsec The GNU Project and This Book
@cindex Free Software Foundation
@cindex Stallman, Richard
The Free Software Foundation (FSF) is a non-profit organization dedicated
to the production and distribution of freely distributable software.
It was founded by Richard M.@: Stallman, the author of the original
Emacs editor. GNU Emacs is the most widely used version of Emacs today.
@cindex GNU Project
The GNU project is an on-going effort on the part of the Free Software
Foundation to create a complete, freely distributable, POSIX compliant
computing environment. (GNU stands for ``GNU's not Unix''.)
The FSF uses the ``GNU General Public License'' (or GPL) to ensure that
source code for their software is always available to the end user. A
copy of the GPL is included for your reference
(@pxref{Copying, ,GNU GENERAL PUBLIC LICENSE}).
The GPL applies to the C language source code for @code{gawk}.
A shell, an editor (Emacs), highly portable optimizing C, C++, and
Objective-C compilers, a symbolic debugger, and dozens of large and
small utilities (such as @code{gawk}), have all been completed and are
freely available. As of this writing (early 1997), the GNU operating
system kernel (the HURD), has been released, but is still in an early
stage of development.
@cindex Linux
@cindex NetBSD
@cindex FreeBSD
Until the GNU operating system is more fully developed, you should
consider using Linux, a freely distributable, Unix-like operating
system for 80386, DEC Alpha, Sun SPARC and other systems. There are
many books on Linux. One freely available one is @cite{Linux
Installation and Getting Started}, by Matt Welsh.
Many Linux distributions are available, often in computer stores or
bundled on CD-ROM with books about Linux.
(There are three other freely available, Unix-like operating systems for
80386 and other systems, NetBSD, FreeBSD,and OpenBSD. All are based on the
4.4-Lite Berkeley Software Distribution, and they use recent versions
of @code{gawk} for their versions of @code{awk}.)
@iftex
This @value{DOCUMENT} you are reading now is actually free. The
information in it is freely available to anyone, the machine readable
source code for the @value{DOCUMENT} comes with @code{gawk}, and anyone
may take this @value{DOCUMENT} to a copying machine and make as many
copies of it as they like. (Take a moment to check the copying
permissions on the Copyright page.)
If you paid money for this @value{DOCUMENT}, what you actually paid for
was the @value{DOCUMENT}'s nice printing and binding, and the
publisher's associated costs to produce it. We have made an effort to
keep these costs reasonable; most people would prefer a bound book to
over 330 pages of photo-copied text that would then have to be held in
a loose-leaf binder (not to mention the time and labor involved in
doing the copying). The same is true of producing this
@value{DOCUMENT} from the machine readable source; the retail price is
only slightly more than the cost per page of printing it
on a laser printer.
@end iftex
This @value{DOCUMENT} itself has gone through several previous,
preliminary editions. I started working on a preliminary draft of
@cite{The GAWK Manual}, by Diane Close, Paul Rubin, and Richard
Stallman in the fall of 1988.
It was around 90 pages long, and barely described the original, ``old''
version of @code{awk}. After substantial revision, the first version of
the @cite{The GAWK Manual} to be released was Edition 0.11 Beta in
October of 1989. The manual then underwent more substantial revision
for Edition 0.13 of December 1991.
David Trueman, Pat Rankin, and Michal Jaegermann contributed sections
of the manual for Edition 0.13.
That edition was published by the
FSF as a bound book early in 1992. Since then there have been several
minor revisions, notably Edition 0.14 of November 1992 that was published
by the FSF in January of 1993, and Edition 0.16 of August 1993.
Edition 1.0 of @cite{@value{TITLE}} represents a significant re-working
of @cite{The GAWK Manual}, with much additional material.
The FSF and I agree that I am now the primary author.
I also felt that it needed a more descriptive title.
@cite{@value{TITLE}} will undoubtedly continue to evolve.
An electronic version
comes with the @code{gawk} distribution from the FSF.
If you find an error in this @value{DOCUMENT}, please report it!
@xref{Bugs, ,Reporting Problems and Bugs}, for information on submitting
problem reports electronically, or write to me in care of the FSF.
@node Acknowledgements, , Manual History, Preface
@unnumberedsec Acknowledgements
@cindex Stallman, Richard
I would like to acknowledge Richard M.@: Stallman, for his vision of a
better world, and for his courage in founding the FSF and starting the
GNU project.
The initial draft of @cite{The GAWK Manual} had the following acknowledgements:
@quotation
Many people need to be thanked for their assistance in producing this
manual. Jay Fenlason contributed many ideas and sample programs. Richard
Mlynarik and Robert Chassell gave helpful comments on drafts of this
manual. The paper @cite{A Supplemental Document for @code{awk}} by John W.@:
Pierce of the Chemistry Department at UC San Diego, pinpointed several
issues relevant both to @code{awk} implementation and to this manual, that
would otherwise have escaped us.
@end quotation
The following people provided many helpful comments on Edition 0.13 of
@cite{The GAWK Manual}: Rick Adams, Michael Brennan, Rich Burridge, Diane Close,
Christopher (``Topher'') Eliot, Michael Lijewski, Pat Rankin, Miriam Robbins,
and Michal Jaegermann.
The following people provided many helpful comments for Edition 1.0 of
@cite{@value{TITLE}}: Karl Berry, Michael Brennan, Darrel
Hankerson, Michal Jaegermann, Michael Lijewski, and Miriam Robbins.
Pat Rankin, Michal Jaegermann, Darrel Hankerson and Scott Deifik
updated their respective sections for Edition 1.0.
Robert J.@: Chassell provided much valuable advice on
the use of Texinfo. He also deserves special thanks for
convincing me @emph{not} to title this @value{DOCUMENT}
@cite{How To Gawk Politely}.
Karl Berry helped significantly with the @TeX{} part of Texinfo.
@cindex Trueman, David
David Trueman deserves special credit; he has done a yeoman job
of evolving @code{gawk} so that it performs well, and without bugs.
Although he is no longer involved with @code{gawk},
working with him on this project was a significant pleasure.
@cindex Deifik, Scott
@cindex Hankerson, Darrel
@cindex Rommel, Kai Uwe
@cindex Rankin, Pat
@cindex Jaegermann, Michal
Scott Deifik, Darrel Hankerson, Kai Uwe Rommel, Pat Rankin, and Michal
Jaegermann (in no particular order) are long time members of the
@code{gawk} ``crack portability team.'' Without their hard work and
help, @code{gawk} would not be nearly the fine program it is today. It
has been and continues to be a pleasure working with this team of fine
people.
@cindex Friedl, Jeffrey
Jeffrey Friedl provided invaluable help in tracking down a number
of last minute problems with regular expressions in @code{gawk} 3.0.
@cindex Kernighan, Brian
David and I would like to thank Brian Kernighan of Bell Labs for
invaluable assistance during the testing and debugging of @code{gawk}, and for
help in clarifying numerous points about the language. We could not have
done nearly as good a job on either @code{gawk} or its documentation without
his help.
@cindex Hughes, Phil
I would like to thank Marshall and Elaine Hartholz of Seattle, and Dr.@:
Bert and Rita Schreiber of Detroit for large amounts of quiet vacation
time in their homes, which allowed me to make significant progress on
this @value{DOCUMENT} and on @code{gawk} itself. Phil Hughes of SSC
contributed in a very important way by loaning me his laptop Linux
system, not once, but twice, allowing me to do a lot of work while
away from home.
@cindex Robbins, Miriam
Finally, I must thank my wonderful wife, Miriam, for her patience through
the many versions of this project, for her proof-reading,
and for sharing me with the computer.
I would like to thank my parents for their love, and for the grace with
which they raised and educated me.
I also must acknowledge my gratitude to G-d, for the many opportunities
He has sent my way, as well as for the gifts He has given me with which to
take advantage of those opportunities.
@sp 2
@noindent
Arnold Robbins @*
Atlanta, Georgia @*
February, 1997
@ignore
Stuff still not covered anywhere:
BASICS:
Integer vs. floating point
Hex vs. octal vs. decimal
Interpreter vs compiler
input/output
@end ignore
@node What Is Awk, Getting Started, Preface, Top
@chapter Introduction
If you are like many computer users, you would frequently like to make
changes in various text files wherever certain patterns appear, or
extract data from parts of certain lines while discarding the rest. To
write a program to do this in a language such as C or Pascal is a
time-consuming inconvenience that may take many lines of code. The job
may be easier with @code{awk}.
The @code{awk} utility interprets a special-purpose programming language
that makes it possible to handle simple data-reformatting jobs
with just a few lines of code.
The GNU implementation of @code{awk} is called @code{gawk}; it is fully
upward compatible with the System V Release 4 version of
@code{awk}. @code{gawk} is also upward compatible with the POSIX
specification of the @code{awk} language. This means that all
properly written @code{awk} programs should work with @code{gawk}.
Thus, we usually don't distinguish between @code{gawk} and other @code{awk}
implementations.
@cindex uses of @code{awk}
Using @code{awk} you can:
@itemize @bullet
@item
manage small, personal databases
@item
generate reports
@item
validate data
@item
produce indexes, and perform other document preparation tasks
@item
even experiment with algorithms that can be adapted later to other computer
languages
@end itemize
@menu
* This Manual:: Using this @value{DOCUMENT}. Includes sample
input files that you can use.
* Conventions:: Typographical Conventions.
* Sample Data Files:: Sample data files for use in the @code{awk}
programs illustrated in this @value{DOCUMENT}.
@end menu
@node This Manual, Conventions, What Is Awk, What Is Awk
@section Using This Book
@cindex book, using this
@cindex using this book
@cindex language, @code{awk}
@cindex program, @code{awk}
@ignore
@cindex @code{awk} language
@cindex @code{awk} program
@end ignore
The term @code{awk} refers to a particular program, and to the language you
use to tell this program what to do. When we need to be careful, we call
the program ``the @code{awk} utility'' and the language ``the @code{awk}
language.'' The term @code{gawk} refers to a version of @code{awk} developed
as part the GNU project. The purpose of this @value{DOCUMENT} is to explain
both the @code{awk} language and how to run the @code{awk} utility.
The main purpose of the @value{DOCUMENT} is to explain the features
of @code{awk}, as defined in the POSIX standard. It does so in the context
of one particular implementation, @code{gawk}. While doing so, it will also
attempt to describe important differences between @code{gawk} and other
@code{awk} implementations. Finally, any @code{gawk} features that
are not in the POSIX standard for @code{awk} will be noted.
@iftex
This @value{DOCUMENT} has the difficult task of being both tutorial and reference.
If you are a novice, feel free to skip over details that seem too complex.
You should also ignore the many cross references; they are for the
expert user, and for the on-line Info version of the document.
@end iftex
The term @dfn{@code{awk} program} refers to a program written by you in
the @code{awk} programming language.
@xref{Getting Started, ,Getting Started with @code{awk}}, for the bare
essentials you need to know to start using @code{awk}.
Some useful ``one-liners'' are included to give you a feel for the
@code{awk} language (@pxref{One-liners, ,Useful One Line Programs}).
Many sample @code{awk} programs have been provided for you
(@pxref{Library Functions, ,A Library of @code{awk} Functions}; also
@pxref{Sample Programs, ,Practical @code{awk} Programs}).
The entire @code{awk} language is summarized for quick reference in
@ref{Gawk Summary, ,@code{gawk} Summary}. Look there if you just need
to refresh your memory about a particular feature.
If you find terms that you aren't familiar with, try looking them
up in the glossary (@pxref{Glossary}).
Most of the time complete @code{awk} programs are used as examples, but in
some of the more advanced sections, only the part of the @code{awk} program
that illustrates the concept being described is shown.
While this @value{DOCUMENT} is aimed principally at people who have not been
exposed
to @code{awk}, there is a lot of information here that even the @code{awk}
expert should find useful. In particular, the description of POSIX
@code{awk}, and the example programs in
@ref{Library Functions, ,A Library of @code{awk} Functions}, and
@ref{Sample Programs, ,Practical @code{awk} Programs},
should be of interest.
@c fakenode --- for prepinfo
@unnumberedsubsec Dark Corners
@display
@i{Who opened that window shade?!?}
Count Dracula
@end display
@sp 1
@cindex d.c., see ``dark corner''
@cindex dark corner
Until the POSIX standard (and @cite{The Gawk Manual}),
many features of @code{awk} were either poorly documented, or not
documented at all. Descriptions of such features
(often called ``dark corners'') are noted in this @value{DOCUMENT} with
``(d.c.)''.
They also appear in the index under the heading ``dark corner.''
@node Conventions, Sample Data Files, This Manual, What Is Awk
@section Typographical Conventions
This @value{DOCUMENT} is written using Texinfo, the GNU documentation formatting language.
A single Texinfo source file is used to produce both the printed and on-line
versions of the documentation.
@iftex
Because of this, the typographical conventions
are slightly different than in other books you may have read.
@end iftex
@ifinfo
This section briefly documents the typographical conventions used in Texinfo.
@end ifinfo
Examples you would type at the command line are preceded by the common
shell primary and secondary prompts, @samp{$} and @samp{>}.
Output from the command is preceded by the glyph ``@print{}''.
This typically represents the command's standard output.
Error messages, and other output on the command's standard error, are preceded
by the glyph ``@error{}''. For example:
@example
@group
$ echo hi on stdout
@print{} hi on stdout
$ echo hello on stderr 1>&2
@error{} hello on stderr
@end group
@end example
@iftex
In the text, command names appear in @code{this font}, while code segments
appear in the same font and quoted, @samp{like this}. Some things will
be emphasized @emph{like this}, and if a point needs to be made
strongly, it will be done @strong{like this}. The first occurrence of
a new term is usually its @dfn{definition}, and appears in the same
font as the previous occurrence of ``definition'' in this sentence.
File names are indicated like this: @file{/path/to/ourfile}.
@end iftex
Characters that you type at the keyboard look @kbd{like this}. In particular,
there are special characters called ``control characters.'' These are
characters that you type by holding down both the @kbd{CONTROL} key and
another key, at the same time. For example, a @kbd{Control-d} is typed
by first pressing and holding the @kbd{CONTROL} key, next
pressing the @kbd{d} key, and finally releasing both keys.
@node Sample Data Files, , Conventions, What Is Awk
@section Data Files for the Examples
@cindex input file, sample
@cindex sample input file
@cindex @file{BBS-list} file
Many of the examples in this @value{DOCUMENT} take their input from two sample
data files. The first, called @file{BBS-list}, represents a list of
computer bulletin board systems together with information about those systems.
The second data file, called @file{inventory-shipped}, contains
information about shipments on a monthly basis. In both files,
each line is considered to be one @dfn{record}.
In the file @file{BBS-list}, each record contains the name of a computer
bulletin board, its phone number, the board's baud rate(s), and a code for
the number of hours it is operational. An @samp{A} in the last column
means the board operates 24 hours a day. A @samp{B} in the last
column means the board operates evening and weekend hours, only. A
@samp{C} means the board operates only on weekends.
@c 2e: Update the baud rates to reflect today's faster modems
@example
@c system mkdir eg
@c system mkdir eg/lib
@c system mkdir eg/data
@c system mkdir eg/prog
@c system mkdir eg/misc
@c file eg/data/BBS-list
aardvark 555-5553 1200/300 B
alpo-net 555-3412 2400/1200/300 A
barfly 555-7685 1200/300 A
bites 555-1675 2400/1200/300 A
camelot 555-0542 300 C
core 555-2912 1200/300 C
fooey 555-1234 2400/1200/300 B
foot 555-6699 1200/300 B
macfoo 555-6480 1200/300 A
sdace 555-3430 2400/1200/300 A
sabafoo 555-2127 1200/300 C
@c endfile
@end example
@cindex @file{inventory-shipped} file
The second data file, called @file{inventory-shipped}, represents
information about shipments during the year.
Each record contains the month of the year, the number
of green crates shipped, the number of red boxes shipped, the number of
orange bags shipped, and the number of blue packages shipped,
respectively. There are 16 entries, covering the 12 months of one year
and four months of the next year.
@example
@c file eg/data/inventory-shipped
Jan 13 25 15 115
Feb 15 32 24 226
Mar 15 24 34 228
Apr 31 52 63 420
May 16 34 29 208
Jun 31 42 75 492
Jul 24 34 67 436
Aug 15 34 47 316
Sep 13 55 37 277
Oct 29 54 68 525
Nov 20 87 82 577
Dec 17 35 61 401
Jan 21 36 64 620
Feb 26 58 80 652
Mar 24 75 70 495
Apr 21 70 74 514
@c endfile
@end example
@ifinfo
If you are reading this in GNU Emacs using Info, you can copy the regions
of text showing these sample files into your own test files. This way you
can try out the examples shown in the remainder of this document. You do
this by using the command @kbd{M-x write-region} to copy text from the Info
file into a file for use with @code{awk}
(@xref{Misc File Ops, , Miscellaneous File Operations, emacs, GNU Emacs Manual},
for more information). Using this information, create your own
@file{BBS-list} and @file{inventory-shipped} files, and practice what you
learn in this @value{DOCUMENT}.
If you are using the stand-alone version of Info,
see @ref{Extract Program, ,Extracting Programs from Texinfo Source Files},
for an @code{awk} program that will extract these data files from
@file{gawk.texi}, the Texinfo source file for this Info file.
@end ifinfo
@node Getting Started, One-liners, What Is Awk, Top
@chapter Getting Started with @code{awk}
@cindex script, definition of
@cindex rule, definition of
@cindex program, definition of
@cindex basic function of @code{awk}
The basic function of @code{awk} is to search files for lines (or other
units of text) that contain certain patterns. When a line matches one
of the patterns, @code{awk} performs specified actions on that line.
@code{awk} keeps processing input lines in this way until the end of the
input files are reached.
@cindex data-driven languages
@cindex procedural languages
@cindex language, data-driven
@cindex language, procedural
Programs in @code{awk} are different from programs in most other languages,
because @code{awk} programs are @dfn{data-driven}; that is, you describe
the data you wish to work with, and then what to do when you find it.
Most other languages are @dfn{procedural}; you have to describe, in great
detail, every step the program is to take. When working with procedural
languages, it is usually much
harder to clearly describe the data your program will process.
For this reason, @code{awk} programs are often refreshingly easy to both
write and read.
@cindex program, definition of
@cindex rule, definition of
When you run @code{awk}, you specify an @code{awk} @dfn{program} that
tells @code{awk} what to do. The program consists of a series of
@dfn{rules}. (It may also contain @dfn{function definitions},
an advanced feature which we will ignore for now.
@xref{User-defined, ,User-defined Functions}.) Each rule specifies one
pattern to search for, and one action to perform when that pattern is found.
Syntactically, a rule consists of a pattern followed by an action. The
action is enclosed in curly braces to separate it from the pattern.
Rules are usually separated by newlines. Therefore, an @code{awk}
program looks like this:
@example
@var{pattern} @{ @var{action} @}
@var{pattern} @{ @var{action} @}
@dots{}
@end example
@menu
* Names:: What name to use to find @code{awk}.
* Running gawk:: How to run @code{gawk} programs; includes
command line syntax.
* Very Simple:: A very simple example.
* Two Rules:: A less simple one-line example with two rules.
* More Complex:: A more complex example.
* Statements/Lines:: Subdividing or combining statements into
lines.
* Other Features:: Other Features of @code{awk}.
* When:: When to use @code{gawk} and when to use other
things.
@end menu
@node Names, Running gawk , Getting Started, Getting Started
@section A Rose By Any Other Name
@cindex old @code{awk} vs. new @code{awk}
@cindex new @code{awk} vs. old @code{awk}
The @code{awk} language has evolved over the years. Full details are
provided in @ref{Language History, ,The Evolution of the @code{awk} Language}.
The language described in this @value{DOCUMENT}
is often referred to as ``new @code{awk}.''
Because of this, many systems have multiple
versions of @code{awk}.
Some systems have an @code{awk} utility that implements the
original version of the @code{awk} language, and a @code{nawk} utility
for the new version. Others have an @code{oawk} for the ``old @code{awk}''
language, and plain @code{awk} for the new one. Still others only
have one version, usually the new one.@footnote{Often, these systems
use @code{gawk} for their @code{awk} implementation!}
All in all, this makes it difficult for you to know which version of
@code{awk} you should run when writing your programs. The best advice
we can give here is to check your local documentation. Look for @code{awk},
@code{oawk}, and @code{nawk}, as well as for @code{gawk}. Chances are, you
will have some version of new @code{awk} on your system, and that is what
you should use when running your programs. (Of course, if you're reading
this @value{DOCUMENT}, chances are good that you have @code{gawk}!)
Throughout this @value{DOCUMENT}, whenever we refer to a language feature
that should be available in any complete implementation of POSIX @code{awk},
we simply use the term @code{awk}. When referring to a feature that is
specific to the GNU implementation, we use the term @code{gawk}.
@node Running gawk, Very Simple, Names, Getting Started
@section How to Run @code{awk} Programs
@cindex command line formats
@cindex running @code{awk} programs
There are several ways to run an @code{awk} program. If the program is
short, it is easiest to include it in the command that runs @code{awk},
like this:
@example
awk '@var{program}' @var{input-file1} @var{input-file2} @dots{}
@end example
@noindent
where @var{program} consists of a series of patterns and actions, as
described earlier.
(The reason for the single quotes is described below, in
@ref{One-shot, ,One-shot Throw-away @code{awk} Programs}.)
When the program is long, it is usually more convenient to put it in a file
and run it with a command like this:
@example
awk -f @var{program-file} @var{input-file1} @var{input-file2} @dots{}
@end example
@menu
* One-shot:: Running a short throw-away @code{awk} program.
* Read Terminal:: Using no input files (input from terminal
instead).
* Long:: Putting permanent @code{awk} programs in
files.
* Executable Scripts:: Making self-contained @code{awk} programs.
* Comments:: Adding documentation to @code{gawk} programs.
@end menu
@node One-shot, Read Terminal, Running gawk, Running gawk
@subsection One-shot Throw-away @code{awk} Programs
Once you are familiar with @code{awk}, you will often type in simple
programs the moment you want to use them. Then you can write the
program as the first argument of the @code{awk} command, like this:
@example
awk '@var{program}' @var{input-file1} @var{input-file2} @dots{}
@end example
@noindent
where @var{program} consists of a series of @var{patterns} and
@var{actions}, as described earlier.
@cindex single quotes, why needed
This command format instructs the @dfn{shell}, or command interpreter,
to start @code{awk} and use the @var{program} to process records in the
input file(s). There are single quotes around @var{program} so that
the shell doesn't interpret any @code{awk} characters as special shell
characters. They also cause the shell to treat all of @var{program} as
a single argument for @code{awk} and allow @var{program} to be more
than one line long.
This format is also useful for running short or medium-sized @code{awk}
programs from shell scripts, because it avoids the need for a separate
file for the @code{awk} program. A self-contained shell script is more
reliable since there are no other files to misplace.
@ref{One-liners, , Useful One Line Programs}, presents several short,
self-contained programs.
As an interesting side point, the command
@example
awk '/foo/' @var{files} @dots{}
@end example
@noindent
is essentially the same as
@cindex @code{egrep}
@example
egrep foo @var{files} @dots{}
@end example
@node Read Terminal, Long, One-shot, Running gawk
@subsection Running @code{awk} without Input Files
@cindex standard input
@cindex input, standard
You can also run @code{awk} without any input files. If you type the
command line:
@example
awk '@var{program}'
@end example
@noindent
then @code{awk} applies the @var{program} to the @dfn{standard input},
which usually means whatever you type on the terminal. This continues
until you indicate end-of-file by typing @kbd{Control-d}.
(On other operating systems, the end-of-file character may be different.
For example, on OS/2 and MS-DOS, it is @kbd{Control-z}.)
For example, the following program prints a friendly piece of advice
(from Douglas Adams' @cite{The Hitchhiker's Guide to the Galaxy}),
to keep you from worrying about the complexities of computer programming
(@samp{BEGIN} is a feature we haven't discussed yet).
@example
$ awk "BEGIN @{ print \"Don't Panic!\" @}"
@print{} Don't Panic!
@end example
@cindex quoting, shell
@cindex shell quoting
This program does not read any input. The @samp{\} before each of the
inner double quotes is necessary because of the shell's quoting rules,
in particular because it mixes both single quotes and double quotes.
This next simple @code{awk} program
emulates the @code{cat} utility; it copies whatever you type at the
keyboard to its standard output. (Why this works is explained shortly.)
@example
$ awk '@{ print @}'
Now is the time for all good men
@print{} Now is the time for all good men
to come to the aid of their country.
@print{} to come to the aid of their country.
Four score and seven years ago, ...
@print{} Four score and seven years ago, ...
What, me worry?
@print{} What, me worry?
@kbd{Control-d}
@end example
@node Long, Executable Scripts, Read Terminal, Running gawk
@subsection Running Long Programs
@cindex running long programs
@cindex @code{-f} option
@cindex program file
@cindex file, @code{awk} program
Sometimes your @code{awk} programs can be very long. In this case it is
more convenient to put the program into a separate file. To tell
@code{awk} to use that file for its program, you type:
@example
awk -f @var{source-file} @var{input-file1} @var{input-file2} @dots{}
@end example
The @samp{-f} instructs the @code{awk} utility to get the @code{awk} program
from the file @var{source-file}. Any file name can be used for
@var{source-file}. For example, you could put the program:
@example
BEGIN @{ print "Don't Panic!" @}
@end example
@noindent
into the file @file{advice}. Then this command:
@example
awk -f advice
@end example
@noindent
does the same thing as this one:
@example
awk "BEGIN @{ print \"Don't Panic!\" @}"
@end example
@cindex quoting, shell
@cindex shell quoting
@noindent
which was explained earlier (@pxref{Read Terminal, ,Running @code{awk} without Input Files}).
Note that you don't usually need single quotes around the file name that you
specify with @samp{-f}, because most file names don't contain any of the shell's
special characters. Notice that in @file{advice}, the @code{awk}
program did not have single quotes around it. The quotes are only needed
for programs that are provided on the @code{awk} command line.
If you want to identify your @code{awk} program files clearly as such,
you can add the extension @file{.awk} to the file name. This doesn't
affect the execution of the @code{awk} program, but it does make
``housekeeping'' easier.
@node Executable Scripts, Comments, Long, Running gawk
@subsection Executable @code{awk} Programs
@cindex executable scripts
@cindex scripts, executable
@cindex self contained programs
@cindex program, self contained
@cindex @code{#!} (executable scripts)
Once you have learned @code{awk}, you may want to write self-contained
@code{awk} scripts, using the @samp{#!} script mechanism. You can do
this on many Unix systems@footnote{The @samp{#!} mechanism works on
Linux systems,
Unix systems derived from Berkeley Unix, System V Release 4, and some System
V Release 3 systems.} (and someday on the GNU system).
For example, you could update the file @file{advice} to look like this:
@example
#! /bin/awk -f
BEGIN @{ print "Don't Panic!" @}
@end example
@noindent
After making this file executable (with the @code{chmod} utility), you
can simply type @samp{advice}
at the shell, and the system will arrange to run @code{awk}@footnote{The
line beginning with @samp{#!} lists the full file name of an interpreter
to be run, and an optional initial command line argument to pass to that
interpreter. The operating system then runs the interpreter with the given
argument and the full argument list of the executed program. The first argument
in the list is the full file name of the @code{awk} program. The rest of the
argument list will either be options to @code{awk}, or data files,
or both.} as if you had typed @samp{awk -f advice}.
@example
@group
$ advice
@print{} Don't Panic!
@end group
@end example
@noindent
Self-contained @code{awk} scripts are useful when you want to write a
program which users can invoke without their having to know that the program is
written in @code{awk}.
@cindex shell scripts
@cindex scripts, shell
Some older systems do not support the @samp{#!} mechanism. You can get a
similar effect using a regular shell script. It would look something
like this:
@example
: The colon ensures execution by the standard shell.
awk '@var{program}' "$@@"
@end example
Using this technique, it is @emph{vital} to enclose the @var{program} in
single quotes to protect it from interpretation by the shell. If you
omit the quotes, only a shell wizard can predict the results.
The @code{"$@@"} causes the shell to forward all the command line
arguments to the @code{awk} program, without interpretation. The first
line, which starts with a colon, is used so that this shell script will
work even if invoked by a user who uses the C shell. (Not all older systems
obey this convention, but many do.)
@c 2e:
@c Someday: (See @cite{The Bourne Again Shell}, by ??.)
@node Comments, , Executable Scripts, Running gawk
@subsection Comments in @code{awk} Programs
@cindex @code{#} (comment)
@cindex comments
@cindex use of comments
@cindex documenting @code{awk} programs
@cindex programs, documenting
A @dfn{comment} is some text that is included in a program for the sake
of human readers; it is not really part of the program. Comments
can explain what the program does, and how it works. Nearly all
programming languages have provisions for comments, because programs are
typically hard to understand without their extra help.
In the @code{awk} language, a comment starts with the sharp sign
character, @samp{#}, and continues to the end of the line.
The @samp{#} does not have to be the first character on the line. The
@code{awk} language ignores the rest of a line following a sharp sign.
For example, we could have put the following into @file{advice}:
@example
# This program prints a nice friendly message. It helps
# keep novice users from being afraid of the computer.
BEGIN @{ print "Don't Panic!" @}
@end example
You can put comment lines into keyboard-composed throw-away @code{awk}
programs also, but this usually isn't very useful; the purpose of a
comment is to help you or another person understand the program at
a later time.
@node Very Simple, Two Rules, Running gawk, Getting Started
@section A Very Simple Example
The following command runs a simple @code{awk} program that searches the
input file @file{BBS-list} for the string of characters: @samp{foo}. (A
string of characters is usually called a @dfn{string}.
The term @dfn{string} is perhaps based on similar usage in English, such
as ``a string of pearls,'' or, ``a string of cars in a train.'')
@example
awk '/foo/ @{ print $0 @}' BBS-list
@end example
@noindent
When lines containing @samp{foo} are found, they are printed, because
@w{@samp{print $0}} means print the current line. (Just @samp{print} by
itself means the same thing, so we could have written that
instead.)
You will notice that slashes, @samp{/}, surround the string @samp{foo}
in the @code{awk} program. The slashes indicate that @samp{foo}
is a pattern to search for. This type of pattern is called a
@dfn{regular expression}, and is covered in more detail later
(@pxref{Regexp, ,Regular Expressions}).
The pattern is allowed to match parts of words.
There are
single-quotes around the @code{awk} program so that the shell won't
interpret any of it as special shell characters.
Here is what this program prints:
@example
@group
$ awk '/foo/ @{ print $0 @}' BBS-list
@print{} fooey 555-1234 2400/1200/300 B
@print{} foot 555-6699 1200/300 B
@print{} macfoo 555-6480 1200/300 A
@print{} sabafoo 555-2127 1200/300 C
@end group
@end example
@cindex action, default
@cindex pattern, default
@cindex default action
@cindex default pattern
In an @code{awk} rule, either the pattern or the action can be omitted,
but not both. If the pattern is omitted, then the action is performed
for @emph{every} input line. If the action is omitted, the default
action is to print all lines that match the pattern.
@cindex empty action
@cindex action, empty
Thus, we could leave out the action (the @code{print} statement and the curly
braces) in the above example, and the result would be the same: all
lines matching the pattern @samp{foo} would be printed. By comparison,
omitting the @code{print} statement but retaining the curly braces makes an
empty action that does nothing; then no lines would be printed.
@node Two Rules, More Complex, Very Simple, Getting Started
@section An Example with Two Rules
@cindex how @code{awk} works
The @code{awk} utility reads the input files one line at a
time. For each line, @code{awk} tries the patterns of each of the rules.
If several patterns match then several actions are run, in the order in
which they appear in the @code{awk} program. If no patterns match, then
no actions are run.
After processing all the rules (perhaps none) that match the line,
@code{awk} reads the next line (however,
@pxref{Next Statement, ,The @code{next} Statement},
and also @pxref{Nextfile Statement, ,The @code{nextfile} Statement}).
This continues until the end of the file is reached.
For example, the @code{awk} program:
@example
/12/ @{ print $0 @}
/21/ @{ print $0 @}
@end example
@noindent
contains two rules. The first rule has the string @samp{12} as the
pattern and @samp{print $0} as the action. The second rule has the
string @samp{21} as the pattern and also has @samp{print $0} as the
action. Each rule's action is enclosed in its own pair of braces.
This @code{awk} program prints every line that contains the string
@samp{12} @emph{or} the string @samp{21}. If a line contains both
strings, it is printed twice, once by each rule.
This is what happens if we run this program on our two sample data files,
@file{BBS-list} and @file{inventory-shipped}, as shown here:
@example
$ awk '/12/ @{ print $0 @}
> /21/ @{ print $0 @}' BBS-list inventory-shipped
@print{} aardvark 555-5553 1200/300 B
@print{} alpo-net 555-3412 2400/1200/300 A
@print{} barfly 555-7685 1200/300 A
@print{} bites 555-1675 2400/1200/300 A
@print{} core 555-2912 1200/300 C
@print{} fooey 555-1234 2400/1200/300 B
@print{} foot 555-6699 1200/300 B
@print{} macfoo 555-6480 1200/300 A
@print{} sdace 555-3430 2400/1200/300 A
@print{} sabafoo 555-2127 1200/300 C
@print{} sabafoo 555-2127 1200/300 C
@print{} Jan 21 36 64 620
@print{} Apr 21 70 74 514
@end example
@noindent
Note how the line in @file{BBS-list} beginning with @samp{sabafoo}
was printed twice, once for each rule.
@node More Complex, Statements/Lines, Two Rules, Getting Started
@section A More Complex Example
@ignore
We have to use ls -lg here to get portable output across Unix systems.
The POSIX ls matches this behavior too. Sigh.
@end ignore
Here is an example to give you an idea of what typical @code{awk}
programs do. This example shows how @code{awk} can be used to
summarize, select, and rearrange the output of another utility. It uses
features that haven't been covered yet, so don't worry if you don't
understand all the details.
@example
ls -lg | awk '$6 == "Nov" @{ sum += $5 @}
END @{ print sum @}'
@end example
@cindex @code{csh}, backslash continuation
@cindex backslash continuation in @code{csh}
This command prints the total number of bytes in all the files in the
current directory that were last modified in November (of any year).
(In the C shell you would need to type a semicolon and then a backslash
at the end of the first line; in a POSIX-compliant shell, such as the
Bourne shell or Bash, the GNU Bourne-Again shell, you can type the example
as shown.)
@ignore
FIXME: how can users tell what shell they are running? Need a footnote
or something, but getting into this is a distraction.
@end ignore
The @w{@samp{ls -lg}} part of this example is a system command that gives
you a listing of the files in a directory, including file size and the date
the file was last modified. Its output looks like this:
@example
-rw-r--r-- 1 arnold user 1933 Nov 7 13:05 Makefile
-rw-r--r-- 1 arnold user 10809 Nov 7 13:03 gawk.h
-rw-r--r-- 1 arnold user 983 Apr 13 12:14 gawk.tab.h
-rw-r--r-- 1 arnold user 31869 Jun 15 12:20 gawk.y
-rw-r--r-- 1 arnold user 22414 Nov 7 13:03 gawk1.c
-rw-r--r-- 1 arnold user 37455 Nov 7 13:03 gawk2.c
-rw-r--r-- 1 arnold user 27511 Dec 9 13:07 gawk3.c
-rw-r--r-- 1 arnold user 7989 Nov 7 13:03 gawk4.c
@end example
@noindent
The first field contains read-write permissions, the second field contains
the number of links to the file, and the third field identifies the owner of
the file. The fourth field identifies the group of the file.
The fifth field contains the size of the file in bytes. The
sixth, seventh and eighth fields contain the month, day, and time,
respectively, that the file was last modified. Finally, the ninth field
contains the name of the file.
@cindex automatic initialization
@cindex initialization, automatic
The @samp{$6 == "Nov"} in our @code{awk} program is an expression that
tests whether the sixth field of the output from @w{@samp{ls -lg}}
matches the string @samp{Nov}. Each time a line has the string
@samp{Nov} for its sixth field, the action @samp{sum += $5} is
performed. This adds the fifth field (the file size) to the variable
@code{sum}. As a result, when @code{awk} has finished reading all the
input lines, @code{sum} is the sum of the sizes of files whose
lines matched the pattern. (This works because @code{awk} variables
are automatically initialized to zero.)
After the last line of output from @code{ls} has been processed, the
@code{END} rule is executed, and the value of @code{sum} is
printed. In this example, the value of @code{sum} would be 80600.
These more advanced @code{awk} techniques are covered in later sections
(@pxref{Action Overview, ,Overview of Actions}). Before you can move on to more
advanced @code{awk} programming, you have to know how @code{awk} interprets
your input and displays your output. By manipulating fields and using
@code{print} statements, you can produce some very useful and impressive
looking reports.
@node Statements/Lines, Other Features, More Complex, Getting Started
@section @code{awk} Statements Versus Lines
@cindex line break
@cindex newline
Most often, each line in an @code{awk} program is a separate statement or
separate rule, like this:
@example
awk '/12/ @{ print $0 @}
/21/ @{ print $0 @}' BBS-list inventory-shipped
@end example
However, @code{gawk} will ignore newlines after any of the following:
@example
, @{ ? : || && do else
@end example
@noindent
A newline at any other point is considered the end of the statement.
(Splitting lines after @samp{?} and @samp{:} is a minor @code{gawk}
extension. The @samp{?} and @samp{:} referred to here is the
three operand conditional expression described in
@ref{Conditional Exp, ,Conditional Expressions}.)
@cindex backslash continuation
@cindex continuation of lines
@cindex line continuation
If you would like to split a single statement into two lines at a point
where a newline would terminate it, you can @dfn{continue} it by ending the
first line with a backslash character, @samp{\}. The backslash must be
the final character on the line to be recognized as a continuation
character. This is allowed absolutely anywhere in the statement, even
in the middle of a string or regular expression. For example:
@example
awk '/This regular expression is too long, so continue it\
on the next line/ @{ print $1 @}'
@end example
@noindent
@cindex portability issues
We have generally not used backslash continuation in the sample programs
in this @value{DOCUMENT}. Since in @code{gawk} there is no limit on the
length of a line, it is never strictly necessary; it just makes programs
more readable. For this same reason, as well as for clarity, we have
kept most statements short in the sample programs presented throughout
the @value{DOCUMENT}. Backslash continuation is most useful when your
@code{awk} program is in a separate source file, instead of typed in on
the command line. You should also note that many @code{awk}
implementations are more particular about where you may use backslash
continuation. For example, they may not allow you to split a string
constant using backslash continuation. Thus, for maximal portability of
your @code{awk} programs, it is best not to split your lines in the
middle of a regular expression or a string.
@cindex @code{csh}, backslash continuation
@cindex backslash continuation in @code{csh}
@strong{Caution: backslash continuation does not work as described above
with the C shell.} Continuation with backslash works for @code{awk}
programs in files, and also for one-shot programs @emph{provided} you
are using a POSIX-compliant shell, such as the Bourne shell or Bash, the
GNU Bourne-Again shell. But the C shell (@code{csh}) behaves
differently! There, you must use two backslashes in a row, followed by
a newline. Note also that when using the C shell, @emph{every} newline
in your awk program must be escaped with a backslash. To illustrate:
@example
% awk 'BEGIN @{ \
? print \\
? "hello, world" \
? @}'
@print{} hello, world
@end example
@noindent
Here, the @samp{%} and @samp{?} are the C shell's primary and secondary
prompts, analogous to the standard shell's @samp{$} and @samp{>}.
@code{awk} is a line-oriented language. Each rule's action has to
begin on the same line as the pattern. To have the pattern and action
on separate lines, you @emph{must} use backslash continuation---there
is no other way.
@cindex backslash continuation and comments
@cindex comments and backslash continuation
Note that backslash continuation and comments do not mix. As soon
as @code{awk} sees the @samp{#} that starts a comment, it ignores
@emph{everything} on the rest of the line. For example:
@example
@group
$ gawk 'BEGIN @{ print "dont panic" # a friendly \
> BEGIN rule
> @}'
@error{} gawk: cmd. line:2: BEGIN rule
@error{} gawk: cmd. line:2: ^ parse error
@end group
@end example
@noindent
Here, it looks like the backslash would continue the comment onto the
next line. However, the backslash-newline combination is never even
noticed, since it is ``hidden'' inside the comment. Thus, the
@samp{BEGIN} is noted as a syntax error.
@cindex multiple statements on one line
When @code{awk} statements within one rule are short, you might want to put
more than one of them on a line. You do this by separating the statements
with a semicolon, @samp{;}.
This also applies to the rules themselves.
Thus, the previous program could have been written:
@example
/12/ @{ print $0 @} ; /21/ @{ print $0 @}
@end example
@noindent
@strong{Note:} the requirement that rules on the same line must be
separated with a semicolon was not in the original @code{awk}
language; it was added for consistency with the treatment of statements
within an action.
@node Other Features, When, Statements/Lines, Getting Started
@section Other Features of @code{awk}
The @code{awk} language provides a number of predefined, or built-in variables, which
your programs can use to get information from @code{awk}. There are other
variables your program can set to control how @code{awk} processes your
data.
In addition, @code{awk} provides a number of built-in functions for doing
common computational and string related operations.
As we develop our presentation of the @code{awk} language, we introduce
most of the variables and many of the functions. They are defined
systematically in @ref{Built-in Variables}, and
@ref{Built-in, ,Built-in Functions}.
@node When, , Other Features, Getting Started
@section When to Use @code{awk}
@cindex when to use @code{awk}
@cindex applications of @code{awk}
You might wonder how @code{awk} might be useful for you. Using
utility programs, advanced patterns, field separators, arithmetic
statements, and other selection criteria, you can produce much more
complex output. The @code{awk} language is very useful for producing
reports from large amounts of raw data, such as summarizing information
from the output of other utility programs like @code{ls}.
(@xref{More Complex, ,A More Complex Example}.)
Programs written with @code{awk} are usually much smaller than they would
be in other languages. This makes @code{awk} programs easy to compose and
use. Often, @code{awk} programs can be quickly composed at your terminal,
used once, and thrown away. Since @code{awk} programs are interpreted, you
can avoid the (usually lengthy) compilation part of the typical
edit-compile-test-debug cycle of software development.
Complex programs have been written in @code{awk}, including a complete
retargetable assembler for eight-bit microprocessors (@pxref{Glossary}, for
more information) and a microcode assembler for a special purpose Prolog
computer. However, @code{awk}'s capabilities are strained by tasks of
such complexity.
If you find yourself writing @code{awk} scripts of more than, say, a few
hundred lines, you might consider using a different programming
language. Emacs Lisp is a good choice if you need sophisticated string
or pattern matching capabilities. The shell is also good at string and
pattern matching; in addition, it allows powerful use of the system
utilities. More conventional languages, such as C, C++, and Lisp, offer
better facilities for system programming and for managing the complexity
of large programs. Programs in these languages may require more lines
of source code than the equivalent @code{awk} programs, but they are
easier to maintain and usually run more efficiently.
@node One-liners, Regexp, Getting Started, Top
@chapter Useful One Line Programs
@cindex one-liners
Many useful @code{awk} programs are short, just a line or two. Here is a
collection of useful, short programs to get you started. Some of these
programs contain constructs that haven't been covered yet. The description
of the program will give you a good idea of what is going on, but please
read the rest of the @value{DOCUMENT} to become an @code{awk} expert!
Most of the examples use a data file named @file{data}. This is just a
placeholder; if you were to use these programs yourself, you would substitute
your own file names for @file{data}.
@ifinfo
Since you are reading this in Info, each line of the example code is
enclosed in quotes, to represent text that you would type literally.
The examples themselves represent shell commands that use single quotes
to keep the shell from interpreting the contents of the program.
When reading the examples, focus on the text between the open and close
quotes.
@end ifinfo
@table @code
@item awk '@{ if (length($0) > max) max = length($0) @}
@itemx @ @ @ @ @ END @{ print max @}' data
This program prints the length of the longest input line.
@item awk 'length($0) > 80' data
This program prints every line that is longer than 80 characters. The sole
rule has a relational expression as its pattern, and has no action (so the
default action, printing the record, is used).
@item expand@ data@ |@ awk@ '@{ if (x < length()) x = length() @}
@itemx @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ END @{ print "maximum line length is " x @}'
This program prints the length of the longest line in @file{data}. The input
is processed by the @code{expand} program to change tabs into spaces,
so the widths compared are actually the right-margin columns.
@item awk 'NF > 0' data
This program prints every line that has at least one field. This is an
easy way to delete blank lines from a file (or rather, to create a new
file similar to the old file but from which the blank lines have been
deleted).
@c Karl Berry points out that new users probably don't want to see
@c multiple ways to do things, just the `best' way. He's probably
@c right. At some point it might be worth adding something about there
@c often being multiple ways to do things in awk, but for now we'll
@c just take this one out.
@ignore
@item awk '@{ if (NF > 0) print @}' data
This program also prints every line that has at least one field. Here we
allow the rule to match every line, and then decide in the action whether
to print.
@end ignore
@item awk@ 'BEGIN@ @{@ for (i = 1; i <= 7; i++)
@itemx @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ print int(101 * rand()) @}'
This program prints seven random numbers from zero to 100, inclusive.
@item ls -lg @var{files} | awk '@{ x += $5 @} ; END @{ print "total bytes: " x @}'
This program prints the total number of bytes used by @var{files}.
@item ls -lg @var{files} | awk '@{ x += $5 @}
@itemx @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ END @{ print "total K-bytes: " (x + 1023)/1024 @}'
This program prints the total number of kilobytes used by @var{files}.
@item awk -F: '@{ print $1 @}' /etc/passwd | sort
This program prints a sorted list of the login names of all users.
@item awk 'END @{ print NR @}' data
This program counts lines in a file.
@item awk 'NR % 2 == 0' data
This program prints the even numbered lines in the data file.
If you were to use the expression @samp{NR % 2 == 1} instead,
it would print the odd numbered lines.
@end table
@node Regexp, Reading Files, One-liners, Top
@chapter Regular Expressions
@cindex pattern, regular expressions
@cindex regexp
@cindex regular expression
@cindex regular expressions as patterns
A @dfn{regular expression}, or @dfn{regexp}, is a way of describing a
set of strings.
Because regular expressions are such a fundamental part of @code{awk}
programming, their format and use deserve a separate chapter.
A regular expression enclosed in slashes (@samp{/})
is an @code{awk} pattern that matches every input record whose text
belongs to that set.
The simplest regular expression is a sequence of letters, numbers, or
both. Such a regexp matches any string that contains that sequence.
Thus, the regexp @samp{foo} matches any string containing @samp{foo}.
Therefore, the pattern @code{/foo/} matches any input record containing
the three characters @samp{foo}, @emph{anywhere} in the record. Other
kinds of regexps let you specify more complicated classes of strings.
@iftex
Initially, the examples will be simple. As we explain more about how
regular expressions work, we will present more complicated examples.
@end iftex
@menu
* Regexp Usage:: How to Use Regular Expressions.
* Escape Sequences:: How to write non-printing characters.
* Regexp Operators:: Regular Expression Operators.
* GNU Regexp Operators:: Operators specific to GNU software.
* Case-sensitivity:: How to do case-insensitive matching.
* Leftmost Longest:: How much text matches.
* Computed Regexps:: Using Dynamic Regexps.
@end menu
@node Regexp Usage, Escape Sequences, Regexp, Regexp
@section How to Use Regular Expressions
A regular expression can be used as a pattern by enclosing it in
slashes. Then the regular expression is tested against the
entire text of each record. (Normally, it only needs
to match some part of the text in order to succeed.) For example, this
prints the second field of each record that contains the three
characters @samp{foo} anywhere in it:
@example
@group
$ awk '/foo/ @{ print $2 @}' BBS-list
@print{} 555-1234
@print{} 555-6699
@print{} 555-6480
@print{} 555-2127
@end group
@end example
@cindex regexp matching operators
@cindex string-matching operators
@cindex operators, string-matching
@cindex operators, regexp matching
@cindex regexp match/non-match operators
@cindex @code{~} operator
@cindex @code{!~} operator
Regular expressions can also be used in matching expressions. These
expressions allow you to specify the string to match against; it need
not be the entire current input record. The two operators, @samp{~}
and @samp{!~}, perform regular expression comparisons. Expressions
using these operators can be used as patterns or in @code{if},
@code{while}, @code{for}, and @code{do} statements.
@ifinfo
@c adding this xref in TeX screws up the formatting too much
(@xref{Statements, ,Control Statements in Actions}.)
@end ifinfo
@table @code
@item @var{exp} ~ /@var{regexp}/
This is true if the expression @var{exp} (taken as a string)
is matched by @var{regexp}. The following example matches, or selects,
all input records with the upper-case letter @samp{J} somewhere in the
first field:
@example
@group
$ awk '$1 ~ /J/' inventory-shipped
@print{} Jan 13 25 15 115
@print{} Jun 31 42 75 492
@print{} Jul 24 34 67 436
@print{} Jan 21 36 64 620
@end group
@end example
So does this:
@example
awk '@{ if ($1 ~ /J/) print @}' inventory-shipped
@end example
@item @var{exp} !~ /@var{regexp}/
This is true if the expression @var{exp} (taken as a character string)
is @emph{not} matched by @var{regexp}. The following example matches,
or selects, all input records whose first field @emph{does not} contain
the upper-case letter @samp{J}:
@example
@group
$ awk '$1 !~ /J/' inventory-shipped
@print{} Feb 15 32 24 226
@print{} Mar 15 24 34 228
@print{} Apr 31 52 63 420
@print{} May 16 34 29 208
@dots{}
@end group
@end example
@end table
@cindex regexp constant
When a regexp is written enclosed in slashes, like @code{/foo/}, we call it
a @dfn{regexp constant}, much like @code{5.27} is a numeric constant, and
@code{"foo"} is a string constant.
@node Escape Sequences, Regexp Operators, Regexp Usage, Regexp
@section Escape Sequences
@cindex escape sequence notation
Some characters cannot be included literally in string constants
(@code{"foo"}) or regexp constants (@code{/foo/}). You represent them
instead with @dfn{escape sequences}, which are character sequences
beginning with a backslash (@samp{\}).
One use of an escape sequence is to include a double-quote character in
a string constant. Since a plain double-quote would end the string, you
must use @samp{\"} to represent an actual double-quote character as a
part of the string. For example:
@example
$ awk 'BEGIN @{ print "He said \"hi!\" to her." @}'
@print{} He said "hi!" to her.
@end example
The backslash character itself is another character that cannot be
included normally; you write @samp{\\} to put one backslash in the
string or regexp. Thus, the string whose contents are the two characters
@samp{"} and @samp{\} must be written @code{"\"\\"}.
Another use of backslash is to represent unprintable characters
such as tab or newline. While there is nothing to stop you from entering most
unprintable characters directly in a string constant or regexp constant,
they may look ugly.
Here is a table of all the escape sequences used in @code{awk}, and
what they represent. Unless noted otherwise, all of these escape
sequences apply to both string constants and regexp constants.
@c @cartouche
@table @code
@item \\
A literal backslash, @samp{\}.
@cindex @code{awk} language, V.4 version
@item \a
The ``alert'' character, @kbd{Control-g}, ASCII code 7 (BEL).
@item \b
Backspace, @kbd{Control-h}, ASCII code 8 (BS).
@item \f
Formfeed, @kbd{Control-l}, ASCII code 12 (FF).
@item \n
Newline, @kbd{Control-j}, ASCII code 10 (LF).
@item \r
Carriage return, @kbd{Control-m}, ASCII code 13 (CR).
@item \t
Horizontal tab, @kbd{Control-i}, ASCII code 9 (HT).
@cindex @code{awk} language, V.4 version
@item \v
Vertical tab, @kbd{Control-k}, ASCII code 11 (VT).
@item \@var{nnn}
The octal value @var{nnn}, where @var{nnn} are one to three digits
between @samp{0} and @samp{7}. For example, the code for the ASCII ESC
(escape) character is @samp{\033}.
@cindex @code{awk} language, V.4 version
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@item \x@var{hh}@dots{}
The hexadecimal value @var{hh}, where @var{hh} are hexadecimal
digits (@samp{0} through @samp{9} and either @samp{A} through @samp{F} or
@samp{a} through @samp{f}). Like the same construct in ANSI C, the escape
sequence continues until the first non-hexadecimal digit is seen. However,
using more than two hexadecimal digits produces undefined results. (The
@samp{\x} escape sequence is not allowed in POSIX @code{awk}.)
@item \/
A literal slash (necessary for regexp constants only).
You use this when you wish to write a regexp
constant that contains a slash. Since the regexp is delimited by
slashes, you need to escape the slash that is part of the pattern,
in order to tell @code{awk} to keep processing the rest of the regexp.
@item \"
A literal double-quote (necessary for string constants only).
You use this when you wish to write a string
constant that contains a double-quote. Since the string is delimited by
double-quotes, you need to escape the quote that is part of the string,
in order to tell @code{awk} to keep processing the rest of the string.
@end table
@c @end cartouche
In @code{gawk}, there are additional two character sequences that begin
with backslash that have special meaning in regexps.
@xref{GNU Regexp Operators, ,Additional Regexp Operators Only in @code{gawk}}.
In a string constant,
what happens if you place a backslash before something that is not one of
the characters listed above? POSIX @code{awk} purposely leaves this case
undefined. There are two choices.
@itemize @bullet
@item
Strip the backslash out. This is what Unix @code{awk} and @code{gawk} both do.
For example, @code{"a\qc"} is the same as @code{"aqc"}.
@item
Leave the backslash alone. Some other @code{awk} implementations do this.
In such implementations, @code{"a\qc"} is the same as if you had typed
@code{"a\\qc"}.
@end itemize
In a regexp, a backslash before any character that is not in the above table,
and not listed in
@ref{GNU Regexp Operators, ,Additional Regexp Operators Only in @code{gawk}},
means that the next character should be taken literally, even if it would
normally be a regexp operator. E.g., @code{/a\+b/} matches the three
characters @samp{a+b}.
@cindex portability issues
For complete portability, do not use a backslash before any character not
listed in the table above.
Another interesting question arises. Suppose you use an octal or hexadecimal
escape to represent a regexp metacharacter
(@pxref{Regexp Operators, , Regular Expression Operators}).
Does @code{awk} treat the character as literal character, or as a regexp
operator?
@cindex dark corner
It turns out that historically, such characters were taken literally (d.c.).
However, the POSIX standard indicates that they should be treated
as real metacharacters, and this is what @code{gawk} does.
However, in compatibility mode (@pxref{Options, ,Command Line Options}),
@code{gawk} treats the characters represented by octal and hexadecimal
escape sequences literally when used in regexp constants. Thus,
@code{/a\52b/} is equivalent to @code{/a\*b/}.
To summarize:
@enumerate 1
@item
The escape sequences in the table above are always processed first,
for both string constants and regexp constants. This happens very early,
as soon as @code{awk} reads your program.
@item
@code{gawk} processes both regexp constants and dynamic regexps
(@pxref{Computed Regexps, ,Using Dynamic Regexps}),
for the special operators listed in
@ref{GNU Regexp Operators, ,Additional Regexp Operators Only in @code{gawk}}.
@item
A backslash before any other character means to treat that character
literally.
@end enumerate
@node Regexp Operators, GNU Regexp Operators, Escape Sequences, Regexp
@section Regular Expression Operators
@cindex metacharacters
@cindex regular expression metacharacters
@cindex regexp operators
You can combine regular expressions with the following characters,
called @dfn{regular expression operators}, or @dfn{metacharacters}, to
increase the power and versatility of regular expressions.
The escape sequences described
@iftex
above
@end iftex
in @ref{Escape Sequences},
are valid inside a regexp. They are introduced by a @samp{\}. They
are recognized and converted into the corresponding real characters as
the very first step in processing regexps.
Here is a table of metacharacters. All characters that are not escape
sequences and that are not listed in the table stand for themselves.
@table @code
@item \
This is used to suppress the special meaning of a character when
matching. For example:
@example
\$
@end example
@noindent
matches the character @samp{$}.
@c NEEDED
@page
@cindex anchors in regexps
@cindex regexp, anchors
@item ^
This matches the beginning of a string. For example:
@example
^@@chapter
@end example
@noindent
matches the @samp{@@chapter} at the beginning of a string, and can be used
to identify chapter beginnings in Texinfo source files.
The @samp{^} is known as an @dfn{anchor}, since it anchors the pattern to
matching only at the beginning of the string.
It is important to realize that @samp{^} does not match the beginning of
a line embedded in a string. In this example the condition is not true:
@example
if ("line1\nLINE 2" ~ /^L/) @dots{}
@end example
@item $
This is similar to @samp{^}, but it matches only at the end of a string.
For example:
@example
p$
@end example
@noindent
matches a record that ends with a @samp{p}. The @samp{$} is also an anchor,
and also does not match the end of a line embedded in a string. In this
example the condition is not true:
@example
if ("line1\nLINE 2" ~ /1$/) @dots{}
@end example
@item .
The period, or dot, matches any single character,
@emph{including} the newline character. For example:
@example
.P
@end example
@noindent
matches any single character followed by a @samp{P} in a string. Using
concatenation we can make a regular expression like @samp{U.A}, which
matches any three-character sequence that begins with @samp{U} and ends
with @samp{A}.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
In strict POSIX mode (@pxref{Options, ,Command Line Options}),
@samp{.} does not match the @sc{nul}
character, which is a character with all bits equal to zero.
Otherwise, @sc{nul} is just another character. Other versions of @code{awk}
may not be able to match the @sc{nul} character.
@ignore
2e: Add stuff that character list is the POSIX terminology. In other
literature known as character set or character class.
@end ignore
@cindex character list
@item [@dots{}]
This is called a @dfn{character list}. It matches any @emph{one} of the
characters that are enclosed in the square brackets. For example:
@example
[MVX]
@end example
@noindent
matches any one of the characters @samp{M}, @samp{V}, or @samp{X} in a
string.
Ranges of characters are indicated by using a hyphen between the beginning
and ending characters, and enclosing the whole thing in brackets. For
example:
@example
[0-9]
@end example
@noindent
matches any digit.
Multiple ranges are allowed. E.g., the list @code{@w{[A-Za-z0-9]}} is a
common way to express the idea of ``all alphanumeric characters.''
To include one of the characters @samp{\}, @samp{]}, @samp{-} or @samp{^} in a
character list, put a @samp{\} in front of it. For example:
@example
[d\]]
@end example
@noindent
matches either @samp{d}, or @samp{]}.
@cindex @code{egrep}
This treatment of @samp{\} in character lists
is compatible with other @code{awk}
implementations, and is also mandated by POSIX.
The regular expressions in @code{awk} are a superset
of the POSIX specification for Extended Regular Expressions (EREs).
POSIX EREs are based on the regular expressions accepted by the
traditional @code{egrep} utility.
@cindex character classes
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@dfn{Character classes} are a new feature introduced in the POSIX standard.
A character class is a special notation for describing
lists of characters that have a specific attribute, but where the
actual characters themselves can vary from country to country and/or
from character set to character set. For example, the notion of what
is an alphabetic character differs in the USA and in France.
A character class is only valid in a regexp @emph{inside} the
brackets of a character list. Character classes consist of @samp{[:},
a keyword denoting the class, and @samp{:]}. Here are the character
classes defined by the POSIX standard.
@table @code
@item [:alnum:]
Alphanumeric characters.
@item [:alpha:]
Alphabetic characters.
@item [:blank:]
Space and tab characters.
@item [:cntrl:]
Control characters.
@item [:digit:]
Numeric characters.
@item [:graph:]
Characters that are printable and are also visible.
(A space is printable, but not visible, while an @samp{a} is both.)
@item [:lower:]
Lower-case alphabetic characters.
@item [:print:]
Printable characters (characters that are not control characters.)
@item [:punct:]
Punctuation characters (characters that are not letter, digits,
control characters, or space characters).
@item [:space:]
Space characters (such as space, tab, and formfeed, to name a few).
@item [:upper:]
Upper-case alphabetic characters.
@item [:xdigit:]
Characters that are hexadecimal digits.
@end table
For example, before the POSIX standard, to match alphanumeric
characters, you had to write @code{/[A-Za-z0-9]/}. If your
character set had other alphabetic characters in it, this would not
match them. With the POSIX character classes, you can write
@code{/[[:alnum:]]/}, and this will match @emph{all} the alphabetic
and numeric characters in your character set.
@cindex collating elements
Two additional special sequences can appear in character lists.
These apply to non-ASCII character sets, which can have single symbols
(called @dfn{collating elements}) that are represented with more than one
character, as well as several characters that are equivalent for
@dfn{collating}, or sorting, purposes. (E.g., in French, a plain ``e''
and a grave-accented ``@`e'' are equivalent.)
@table @asis
@cindex collating symbols
@item Collating Symbols
A @dfn{collating symbol} is a multi-character collating element enclosed in
@samp{[.} and @samp{.]}. For example, if @samp{ch} is a collating element,
then @code{[[.ch.]]} is a regexp that matches this collating element, while
@code{[ch]} is a regexp that matches either @samp{c} or @samp{h}.
@cindex equivalence classes
@item Equivalence Classes
An @dfn{equivalence class} is a locale-specific name for a list of
characters that are equivalent. The name is enclosed in
@samp{[=} and @samp{=]}.
For example, the name @samp{e} might be used to represent all of
``e,'' ``@`e,'' and ``@'e.'' In this case, @code{[[=e]]} is a regexp
that matches any of @samp{e}, @samp{@'e}, or @samp{@`e}.
@end table
These features are very valuable in non-English speaking locales.
@strong{Caution:} The library functions that @code{gawk} uses for regular
expression matching currently only recognize POSIX character classes;
they do not recognize collating symbols or equivalence classes.
@c maybe one day ...
@cindex complemented character list
@cindex character list, complemented
@item [^ @dots{}]
This is a @dfn{complemented character list}. The first character after
the @samp{[} @emph{must} be a @samp{^}. It matches any characters
@emph{except} those in the square brackets. For example:
@example
[^0-9]
@end example
@noindent
matches any character that is not a digit.
@item |
This is the @dfn{alternation operator}, and it is used to specify
alternatives. For example:
@example
^P|[0-9]
@end example
@noindent
matches any string that matches either @samp{^P} or @samp{[0-9]}. This
means it matches any string that starts with @samp{P} or contains a digit.
The alternation applies to the largest possible regexps on either side.
In other words, @samp{|} has the lowest precedence of all the regular
expression operators.
@item (@dots{})
Parentheses are used for grouping in regular expressions as in
arithmetic. They can be used to concatenate regular expressions
containing the alternation operator, @samp{|}. For example,
@samp{@@(samp|code)\@{[^@}]+\@}} matches both @samp{@@code@{foo@}} and
@samp{@@samp@{bar@}}. (These are Texinfo formatting control sequences.)
@item *
This symbol means that the preceding regular expression is to be
repeated as many times as necessary to find a match. For example:
@example
ph*
@end example
@noindent
applies the @samp{*} symbol to the preceding @samp{h} and looks for matches
of one @samp{p} followed by any number of @samp{h}s. This will also match
just @samp{p} if no @samp{h}s are present.
The @samp{*} repeats the @emph{smallest} possible preceding expression.
(Use parentheses if you wish to repeat a larger expression.) It finds
as many repetitions as possible. For example:
@example
awk '/\(c[ad][ad]*r x\)/ @{ print @}' sample
@end example
@noindent
prints every record in @file{sample} containing a string of the form
@samp{(car x)}, @samp{(cdr x)}, @samp{(cadr x)}, and so on.
Notice the escaping of the parentheses by preceding them
with backslashes.
@item +
This symbol is similar to @samp{*}, but the preceding expression must be
matched at least once. This means that:
@example
wh+y
@end example
@noindent
would match @samp{why} and @samp{whhy} but not @samp{wy}, whereas
@samp{wh*y} would match all three of these strings. This is a simpler
way of writing the last @samp{*} example:
@example
awk '/\(c[ad]+r x\)/ @{ print @}' sample
@end example
@item ?
This symbol is similar to @samp{*}, but the preceding expression can be
matched either once or not at all. For example:
@example
fe?d
@end example
@noindent
will match @samp{fed} and @samp{fd}, but nothing else.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@cindex interval expressions
@item @{@var{n}@}
@itemx @{@var{n},@}
@itemx @{@var{n},@var{m}@}
One or two numbers inside braces denote an @dfn{interval expression}.
If there is one number in the braces, the preceding regexp is repeated
@var{n} times.
If there are two numbers separated by a comma, the preceding regexp is
repeated @var{n} to @var{m} times.
If there is one number followed by a comma, then the preceding regexp
is repeated at least @var{n} times.
@table @code
@item wh@{3@}y
matches @samp{whhhy} but not @samp{why} or @samp{whhhhy}.
@item wh@{3,5@}y
matches @samp{whhhy} or @samp{whhhhy} or @samp{whhhhhy}, only.
@item wh@{2,@}y
matches @samp{whhy} or @samp{whhhy}, and so on.
@end table
Interval expressions were not traditionally available in @code{awk}.
As part of the POSIX standard they were added, to make @code{awk}
and @code{egrep} consistent with each other.
However, since old programs may use @samp{@{} and @samp{@}} in regexp
constants, by default @code{gawk} does @emph{not} match interval expressions
in regexps. If either @samp{--posix} or @samp{--re-interval} are specified
(@pxref{Options, , Command Line Options}), then interval expressions
are allowed in regexps.
@end table
@cindex precedence, regexp operators
@cindex regexp operators, precedence of
In regular expressions, the @samp{*}, @samp{+}, and @samp{?} operators,
as well as the braces @samp{@{} and @samp{@}},
have
the highest precedence, followed by concatenation, and finally by @samp{|}.
As in arithmetic, parentheses can change how operators are grouped.
If @code{gawk} is in compatibility mode
(@pxref{Options, ,Command Line Options}),
character classes and interval expressions are not available in
regular expressions.
The next
@ifinfo
node
@end ifinfo
@iftex
section
@end iftex
discusses the GNU-specific regexp operators, and provides
more detail concerning how command line options affect the way @code{gawk}
interprets the characters in regular expressions.
@node GNU Regexp Operators, Case-sensitivity, Regexp Operators, Regexp
@section Additional Regexp Operators Only in @code{gawk}
@c This section adapted from the regex-0.12 manual
@cindex regexp operators, GNU specific
GNU software that deals with regular expressions provides a number of
additional regexp operators. These operators are described in this
section, and are specific to @code{gawk}; they are not available in other
@code{awk} implementations.
@cindex word, regexp definition of
Most of the additional operators are for dealing with word matching.
For our purposes, a @dfn{word} is a sequence of one or more letters, digits,
or underscores (@samp{_}).
@table @code
@cindex @code{\w} regexp operator
@item \w
This operator matches any word-constituent character, i.e.@: any
letter, digit, or underscore. Think of it as a short-hand for
@c @w{@code{[A-Za-z0-9_]}} or
@w{@code{[[:alnum:]_]}}.
@cindex @code{\W} regexp operator
@item \W
This operator matches any character that is not word-constituent.
Think of it as a short-hand for
@c @w{@code{[^A-Za-z0-9_]}} or
@w{@code{[^[:alnum:]_]}}.
@cindex @code{\<} regexp operator
@item \<
This operator matches the empty string at the beginning of a word.
For example, @code{/\<away/} matches @samp{away}, but not
@samp{stowaway}.
@cindex @code{\>} regexp operator
@item \>
This operator matches the empty string at the end of a word.
For example, @code{/stow\>/} matches @samp{stow}, but not @samp{stowaway}.
@cindex @code{\y} regexp operator
@cindex word boundaries, matching
@item \y
This operator matches the empty string at either the beginning or the
end of a word (the word boundar@strong{y}). For example, @samp{\yballs?\y}
matches either @samp{ball} or @samp{balls} as a separate word.
@cindex @code{\B} regexp operator
@item \B
This operator matches the empty string within a word. In other words,
@samp{\B} matches the empty string that occurs between two
word-constituent characters. For example,
@code{/\Brat\B/} matches @samp{crate}, but it does not match @samp{dirty rat}.
@samp{\B} is essentially the opposite of @samp{\y}.
@end table
There are two other operators that work on buffers. In Emacs, a
@dfn{buffer} is, naturally, an Emacs buffer. For other programs, the
regexp library routines that @code{gawk} uses consider the entire
string to be matched as the buffer.
For @code{awk}, since @samp{^} and @samp{$} always work in terms
of the beginning and end of strings, these operators don't add any
new capabilities. They are provided for compatibility with other GNU
software.
@cindex buffer matching operators
@table @code
@cindex @code{\`} regexp operator
@item \`
This operator matches the empty string at the
beginning of the buffer.
@cindex @code{\'} regexp operator
@item \'
This operator matches the empty string at the
end of the buffer.
@end table
In other GNU software, the word boundary operator is @samp{\b}. However,
that conflicts with the @code{awk} language's definition of @samp{\b}
as backspace, so @code{gawk} uses a different letter.
An alternative method would have been to require two backslashes in the
GNU operators, but this was deemed to be too confusing, and the current
method of using @samp{\y} for the GNU @samp{\b} appears to be the
lesser of two evils.
@c NOTE!!! Keep this in sync with the same table in the summary appendix!
@cindex regexp, effect of command line options
The various command line options
(@pxref{Options, ,Command Line Options})
control how @code{gawk} interprets characters in regexps.
@table @asis
@item No options
In the default case, @code{gawk} provide all the facilities of
POSIX regexps and the GNU regexp operators described
@iftex
above.
@end iftex
@ifinfo
in @ref{Regexp Operators, ,Regular Expression Operators}.
@end ifinfo
However, interval expressions are not supported.
@item @code{--posix}
Only POSIX regexps are supported, the GNU operators are not special
(e.g., @samp{\w} matches a literal @samp{w}). Interval expressions
are allowed.
@item @code{--traditional}
Traditional Unix @code{awk} regexps are matched. The GNU operators
are not special, interval expressions are not available, and neither
are the POSIX character classes (@code{[[:alnum:]]} and so on).
Characters described by octal and hexadecimal escape sequences are
treated literally, even if they represent regexp metacharacters.
@item @code{--re-interval}
Allow interval expressions in regexps, even if @samp{--traditional}
has been provided.
@end table
@node Case-sensitivity, Leftmost Longest, GNU Regexp Operators, Regexp
@section Case-sensitivity in Matching
@cindex case sensitivity
@cindex ignoring case
Case is normally significant in regular expressions, both when matching
ordinary characters (i.e.@: not metacharacters), and inside character
sets. Thus a @samp{w} in a regular expression matches only a lower-case
@samp{w} and not an upper-case @samp{W}.
The simplest way to do a case-independent match is to use a character
list: @samp{[Ww]}. However, this can be cumbersome if you need to use it
often; and it can make the regular expressions harder to
read. There are two alternatives that you might prefer.
One way to do a case-insensitive match at a particular point in the
program is to convert the data to a single case, using the
@code{tolower} or @code{toupper} built-in string functions (which we
haven't discussed yet;
@pxref{String Functions, ,Built-in Functions for String Manipulation}).
For example:
@example
tolower($1) ~ /foo/ @{ @dots{} @}
@end example
@noindent
converts the first field to lower-case before matching against it.
This will work in any POSIX-compliant implementation of @code{awk}.
@cindex differences between @code{gawk} and @code{awk}
@cindex @code{~} operator
@cindex @code{!~} operator
@vindex IGNORECASE
Another method, specific to @code{gawk}, is to set the variable
@code{IGNORECASE} to a non-zero value (@pxref{Built-in Variables}).
When @code{IGNORECASE} is not zero, @emph{all} regexp and string
operations ignore case. Changing the value of
@code{IGNORECASE} dynamically controls the case sensitivity of your
program as it runs. Case is significant by default because
@code{IGNORECASE} (like most variables) is initialized to zero.
@example
@group
x = "aB"
if (x ~ /ab/) @dots{} # this test will fail
@end group
@group
IGNORECASE = 1
if (x ~ /ab/) @dots{} # now it will succeed
@end group
@end example
In general, you cannot use @code{IGNORECASE} to make certain rules
case-insensitive and other rules case-sensitive, because there is no way
to set @code{IGNORECASE} just for the pattern of a particular rule.
@ignore
This isn't quite true. Consider:
IGNORECASE=1 && /foObAr/ { .... }
IGNORECASE=0 || /foobar/ { .... }
But that's pretty bad style and I don't want to get into it at this
late date.
@end ignore
To do this, you must use character lists or @code{tolower}. However, one
thing you can do only with @code{IGNORECASE} is turn case-sensitivity on
or off dynamically for all the rules at once.
@code{IGNORECASE} can be set on the command line, or in a @code{BEGIN} rule
(@pxref{Other Arguments, ,Other Command Line Arguments}; also
@pxref{Using BEGIN/END, ,Startup and Cleanup Actions}).
Setting @code{IGNORECASE} from the command line is a way to make
a program case-insensitive without having to edit it.
Prior to version 3.0 of @code{gawk}, the value of @code{IGNORECASE}
only affected regexp operations. It did not affect string comparison
with @samp{==}, @samp{!=}, and so on.
Beginning with version 3.0, both regexp and string comparison
operations are affected by @code{IGNORECASE}.
@cindex ISO 8859-1
@cindex ISO Latin-1
Beginning with version 3.0 of @code{gawk}, the equivalences between upper-case
and lower-case characters are based on the ISO-8859-1 (ISO Latin-1)
character set. This character set is a superset of the traditional 128
ASCII characters, that also provides a number of characters suitable
for use with European languages.
@ignore
A pure ASCII character set can be used instead if @code{gawk} is compiled
with @samp{-DUSE_PURE_ASCII}.
@end ignore
The value of @code{IGNORECASE} has no effect if @code{gawk} is in
compatibility mode (@pxref{Options, ,Command Line Options}).
Case is always significant in compatibility mode.
@node Leftmost Longest, Computed Regexps, Case-sensitivity, Regexp
@section How Much Text Matches?
@cindex leftmost longest match
@cindex matching, leftmost longest
Consider the following example:
@example
echo aaaabcd | awk '@{ sub(/a+/, "<A>"); print @}'
@end example
This example uses the @code{sub} function (which we haven't discussed yet,
@pxref{String Functions, ,Built-in Functions for String Manipulation})
to make a change to the input record. Here, the regexp @code{/a+/}
indicates ``one or more @samp{a} characters,'' and the replacement
text is @samp{<A>}.
The input contains four @samp{a} characters. What will the output be?
In other words, how many is ``one or more''---will @code{awk} match two,
three, or all four @samp{a} characters?
The answer is, @code{awk} (and POSIX) regular expressions always match
the leftmost, @emph{longest} sequence of input characters that can
match. Thus, in this example, all four @samp{a} characters are
replaced with @samp{<A>}.
@example
$ echo aaaabcd | awk '@{ sub(/a+/, "<A>"); print @}'
@print{} <A>bcd
@end example
For simple match/no-match tests, this is not so important. But when doing
regexp-based field and record splitting, and
text matching and substitutions with the @code{match}, @code{sub}, @code{gsub},
and @code{gensub} functions, it is very important.
@ifinfo
@xref{String Functions, ,Built-in Functions for String Manipulation},
for more information on these functions.
@end ifinfo
Understanding this principle is also important for regexp-based record
and field splitting (@pxref{Records, ,How Input is Split into Records},
and also @pxref{Field Separators, ,Specifying How Fields are Separated}).
@node Computed Regexps, , Leftmost Longest, Regexp
@section Using Dynamic Regexps
@cindex computed regular expressions
@cindex regular expressions, computed
@cindex dynamic regular expressions
@cindex regexp, dynamic
@cindex @code{~} operator
@cindex @code{!~} operator
The right hand side of a @samp{~} or @samp{!~} operator need not be a
regexp constant (i.e.@: a string of characters between slashes). It may
be any expression. The expression is evaluated, and converted if
necessary to a string; the contents of the string are used as the
regexp. A regexp that is computed in this way is called a @dfn{dynamic
regexp}. For example:
@example
BEGIN @{ identifier_regexp = "[A-Za-z_][A-Za-z_0-9]+" @}
$0 ~ identifier_regexp @{ print @}
@end example
@noindent
sets @code{identifier_regexp} to a regexp that describes @code{awk}
variable names, and tests if the input record matches this regexp.
@strong{Caution:} When using the @samp{~} and @samp{!~}
operators, there is a difference between a regexp constant
enclosed in slashes, and a string constant enclosed in double quotes.
If you are going to use a string constant, you have to understand that
the string is in essence scanned @emph{twice}; the first time when
@code{awk} reads your program, and the second time when it goes to
match the string on the left-hand side of the operator with the pattern
on the right. This is true of any string valued expression (such as
@code{identifier_regexp} above), not just string constants.
@cindex regexp constants, difference between slashes and quotes
What difference does it make if the string is
scanned twice? The answer has to do with escape sequences, and particularly
with backslashes. To get a backslash into a regular expression inside a
string, you have to type two backslashes.
For example, @code{/\*/} is a regexp constant for a literal @samp{*}.
Only one backslash is needed. To do the same thing with a string,
you would have to type @code{"\\*"}. The first backslash escapes the
second one, so that the string actually contains the
two characters @samp{\} and @samp{*}.
@cindex common mistakes
@cindex mistakes, common
@cindex errors, common
Given that you can use both regexp and string constants to describe
regular expressions, which should you use? The answer is ``regexp
constants,'' for several reasons.
@enumerate 1
@item
String constants are more complicated to write, and
more difficult to read. Using regexp constants makes your programs
less error-prone. Not understanding the difference between the two
kinds of constants is a common source of errors.
@item
It is also more efficient to use regexp constants: @code{awk} can note
that you have supplied a regexp and store it internally in a form that
makes pattern matching more efficient. When using a string constant,
@code{awk} must first convert the string into this internal form, and
then perform the pattern matching.
@item
Using regexp constants is better style; it shows clearly that you
intend a regexp match.
@end enumerate
@node Reading Files, Printing, Regexp, Top
@chapter Reading Input Files
@cindex reading files
@cindex input
@cindex standard input
@vindex FILENAME
In the typical @code{awk} program, all input is read either from the
standard input (by default the keyboard, but often a pipe from another
command) or from files whose names you specify on the @code{awk} command
line. If you specify input files, @code{awk} reads them in order, reading
all the data from one before going on to the next. The name of the current
input file can be found in the built-in variable @code{FILENAME}
(@pxref{Built-in Variables}).
The input is read in units called @dfn{records}, and processed by the
rules of your program one record at a time.
By default, each record is one line. Each
record is automatically split into chunks called @dfn{fields}.
This makes it more convenient for programs to work on the parts of a record.
On rare occasions you will need to use the @code{getline} command.
The @code{getline} command is valuable, both because it
can do explicit input from any number of files, and because the files
used with it do not have to be named on the @code{awk} command line
(@pxref{Getline, ,Explicit Input with @code{getline}}).
@menu
* Records:: Controlling how data is split into records.
* Fields:: An introduction to fields.
* Non-Constant Fields:: Non-constant Field Numbers.
* Changing Fields:: Changing the Contents of a Field.
* Field Separators:: The field separator and how to change it.
* Constant Size:: Reading constant width data.
* Multiple Line:: Reading multi-line records.
* Getline:: Reading files under explicit program control
using the @code{getline} function.
@end menu
@node Records, Fields, Reading Files, Reading Files
@section How Input is Split into Records
@cindex record separator, @code{RS}
@cindex changing the record separator
@cindex record, definition of
@vindex RS
The @code{awk} utility divides the input for your @code{awk}
program into records and fields.
Records are separated by a character called the @dfn{record separator}.
By default, the record separator is the newline character.
This is why records are, by default, single lines.
You can use a different character for the record separator by
assigning the character to the built-in variable @code{RS}.
You can change the value of @code{RS} in the @code{awk} program,
like any other variable, with the
assignment operator, @samp{=} (@pxref{Assignment Ops, ,Assignment Expressions}).
The new record-separator character should be enclosed in quotation marks,
which indicate
a string constant. Often the right time to do this is at the beginning
of execution, before any input has been processed, so that the very
first record will be read with the proper separator. To do this, use
the special @code{BEGIN} pattern
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}). For
example:
@example
awk 'BEGIN @{ RS = "/" @} ; @{ print $0 @}' BBS-list
@end example
@noindent
changes the value of @code{RS} to @code{"/"}, before reading any input.
This is a string whose first character is a slash; as a result, records
are separated by slashes. Then the input file is read, and the second
rule in the @code{awk} program (the action with no pattern) prints each
record. Since each @code{print} statement adds a newline at the end of
its output, the effect of this @code{awk} program is to copy the input
with each slash changed to a newline. Here are the results of running
the program on @file{BBS-list}:
@example
@group
$ awk 'BEGIN @{ RS = "/" @} ; @{ print $0 @}' BBS-list
@print{} aardvark 555-5553 1200
@print{} 300 B
@print{} alpo-net 555-3412 2400
@print{} 1200
@print{} 300 A
@print{} barfly 555-7685 1200
@print{} 300 A
@print{} bites 555-1675 2400
@print{} 1200
@print{} 300 A
@print{} camelot 555-0542 300 C
@print{} core 555-2912 1200
@print{} 300 C
@print{} fooey 555-1234 2400
@print{} 1200
@print{} 300 B
@print{} foot 555-6699 1200
@print{} 300 B
@print{} macfoo 555-6480 1200
@print{} 300 A
@print{} sdace 555-3430 2400
@print{} 1200
@print{} 300 A
@print{} sabafoo 555-2127 1200
@print{} 300 C
@print{}
@end group
@end example
@noindent
Note that the entry for the @samp{camelot} BBS is not split.
In the original data file
(@pxref{Sample Data Files, , Data Files for the Examples}),
the line looks like this:
@example
camelot 555-0542 300 C
@end example
@noindent
It only has one baud rate; there are no slashes in the record.
Another way to change the record separator is on the command line,
using the variable-assignment feature
(@pxref{Other Arguments, ,Other Command Line Arguments}).
@example
awk '@{ print $0 @}' RS="/" BBS-list
@end example
@noindent
This sets @code{RS} to @samp{/} before processing @file{BBS-list}.
Using an unusual character such as @samp{/} for the record separator
produces correct behavior in the vast majority of cases. However,
the following (extreme) pipeline prints a surprising @samp{1}. There
is one field, consisting of a newline. The value of the built-in
variable @code{NF} is the number of fields in the current record.
@example
$ echo | awk 'BEGIN @{ RS = "a" @} ; @{ print NF @}'
@print{} 1
@end example
@cindex dark corner
@noindent
Reaching the end of an input file terminates the current input record,
even if the last character in the file is not the character in @code{RS}
(d.c.).
@cindex empty string
The empty string, @code{""} (a string of no characters), has a special meaning
as the value of @code{RS}: it means that records are separated
by one or more blank lines, and nothing else.
@xref{Multiple Line, ,Multiple-Line Records}, for more details.
If you change the value of @code{RS} in the middle of an @code{awk} run,
the new value is used to delimit subsequent records, but the record
currently being processed (and records already processed) are not
affected.
@vindex RT
@cindex record terminator, @code{RT}
@cindex terminator, record
@cindex differences between @code{gawk} and @code{awk}
After the end of the record has been determined, @code{gawk}
sets the variable @code{RT} to the text in the input that matched
@code{RS}.
@cindex regular expressions as record separators
The value of @code{RS} is in fact not limited to a one-character
string. It can be any regular expression
(@pxref{Regexp, ,Regular Expressions}).
In general, each record
ends at the next string that matches the regular expression; the next
record starts at the end of the matching string. This general rule is
actually at work in the usual case, where @code{RS} contains just a
newline: a record ends at the beginning of the next matching string (the
next newline in the input) and the following record starts just after
the end of this string (at the first character of the following line).
The newline, since it matches @code{RS}, is not part of either record.
When @code{RS} is a single character, @code{RT} will
contain the same single character. However, when @code{RS} is a
regular expression, then @code{RT} becomes more useful; it contains
the actual input text that matched the regular expression.
The following example illustrates both of these features.
It sets @code{RS} equal to a regular expression that
matches either a newline, or a series of one or more upper-case letters
with optional leading and/or trailing white space
(@pxref{Regexp, , Regular Expressions}).
@example
$ echo record 1 AAAA record 2 BBBB record 3 |
> gawk 'BEGIN @{ RS = "\n|( *[[:upper:]]+ *)" @}
> @{ print "Record =", $0, "and RT =", RT @}'
@print{} Record = record 1 and RT = AAAA
@print{} Record = record 2 and RT = BBBB
@print{} Record = record 3 and RT =
@print{}
@end example
@noindent
The final line of output has an extra blank line. This is because the
value of @code{RT} is a newline, and then the @code{print} statement
supplies its own terminating newline.
@xref{Simple Sed, ,A Simple Stream Editor}, for a more useful example
of @code{RS} as a regexp and @code{RT}.
@cindex differences between @code{gawk} and @code{awk}
The use of @code{RS} as a regular expression and the @code{RT}
variable are @code{gawk} extensions; they are not available in
compatibility mode
(@pxref{Options, ,Command Line Options}).
In compatibility mode, only the first character of the value of
@code{RS} is used to determine the end of the record.
@cindex number of records, @code{NR}, @code{FNR}
@vindex NR
@vindex FNR
The @code{awk} utility keeps track of the number of records that have
been read so far from the current input file. This value is stored in a
built-in variable called @code{FNR}. It is reset to zero when a new
file is started. Another built-in variable, @code{NR}, is the total
number of input records read so far from all data files. It starts at zero
but is never automatically reset to zero.
@node Fields, Non-Constant Fields, Records, Reading Files
@section Examining Fields
@cindex examining fields
@cindex fields
@cindex accessing fields
When @code{awk} reads an input record, the record is
automatically separated or @dfn{parsed} by the interpreter into chunks
called @dfn{fields}. By default, fields are separated by whitespace,
like words in a line.
Whitespace in @code{awk} means any string of one or more spaces,
tabs or newlines;@footnote{In POSIX @code{awk}, newlines are not
considered whitespace for separating fields.} other characters such as
formfeed, and so on, that are
considered whitespace by other languages are @emph{not} considered
whitespace by @code{awk}.
The purpose of fields is to make it more convenient for you to refer to
these pieces of the record. You don't have to use them---you can
operate on the whole record if you wish---but fields are what make
simple @code{awk} programs so powerful.
@cindex @code{$} (field operator)
@cindex field operator @code{$}
To refer to a field in an @code{awk} program, you use a dollar-sign,
@samp{$}, followed by the number of the field you want. Thus, @code{$1}
refers to the first field, @code{$2} to the second, and so on. For
example, suppose the following is a line of input:
@example
This seems like a pretty nice example.
@end example
@noindent
Here the first field, or @code{$1}, is @samp{This}; the second field, or
@code{$2}, is @samp{seems}; and so on. Note that the last field,
@code{$7}, is @samp{example.}. Because there is no space between the
@samp{e} and the @samp{.}, the period is considered part of the seventh
field.
@vindex NF
@cindex number of fields, @code{NF}
@code{NF} is a built-in variable whose value
is the number of fields in the current record.
@code{awk} updates the value of @code{NF} automatically, each time
a record is read.
No matter how many fields there are, the last field in a record can be
represented by @code{$NF}. So, in the example above, @code{$NF} would
be the same as @code{$7}, which is @samp{example.}. Why this works is
explained below (@pxref{Non-Constant Fields, ,Non-constant Field Numbers}).
If you try to reference a field beyond the last one, such as @code{$8}
when the record has only seven fields, you get the empty string.
@c the empty string acts like 0 in some contexts, but I don't want to
@c get into that here....
@code{$0}, which looks like a reference to the ``zeroth'' field, is
a special case: it represents the whole input record. @code{$0} is
used when you are not interested in fields.
Here are some more examples:
@example
@group
$ awk '$1 ~ /foo/ @{ print $0 @}' BBS-list
@print{} fooey 555-1234 2400/1200/300 B
@print{} foot 555-6699 1200/300 B
@print{} macfoo 555-6480 1200/300 A
@print{} sabafoo 555-2127 1200/300 C
@end group
@end example
@noindent
This example prints each record in the file @file{BBS-list} whose first
field contains the string @samp{foo}. The operator @samp{~} is called a
@dfn{matching operator}
(@pxref{Regexp Usage, , How to Use Regular Expressions});
it tests whether a string (here, the field @code{$1}) matches a given regular
expression.
By contrast, the following example
looks for @samp{foo} in @emph{the entire record} and prints the first
field and the last field for each input record containing a
match.
@example
@group
$ awk '/foo/ @{ print $1, $NF @}' BBS-list
@print{} fooey B
@print{} foot B
@print{} macfoo A
@print{} sabafoo C
@end group
@end example
@node Non-Constant Fields, Changing Fields, Fields, Reading Files
@section Non-constant Field Numbers
The number of a field does not need to be a constant. Any expression in
the @code{awk} language can be used after a @samp{$} to refer to a
field. The value of the expression specifies the field number. If the
value is a string, rather than a number, it is converted to a number.
Consider this example:
@example
awk '@{ print $NR @}'
@end example
@noindent
Recall that @code{NR} is the number of records read so far: one in the
first record, two in the second, etc. So this example prints the first
field of the first record, the second field of the second record, and so
on. For the twentieth record, field number 20 is printed; most likely,
the record has fewer than 20 fields, so this prints a blank line.
Here is another example of using expressions as field numbers:
@example
awk '@{ print $(2*2) @}' BBS-list
@end example
@code{awk} must evaluate the expression @samp{(2*2)} and use
its value as the number of the field to print. The @samp{*} sign
represents multiplication, so the expression @samp{2*2} evaluates to four.
The parentheses are used so that the multiplication is done before the
@samp{$} operation; they are necessary whenever there is a binary
operator in the field-number expression. This example, then, prints the
hours of operation (the fourth field) for every line of the file
@file{BBS-list}. (All of the @code{awk} operators are listed, in
order of decreasing precedence, in
@ref{Precedence, , Operator Precedence (How Operators Nest)}.)
If the field number you compute is zero, you get the entire record.
Thus, @code{$(2-2)} has the same value as @code{$0}. Negative field
numbers are not allowed; trying to reference one will usually terminate
your running @code{awk} program. (The POSIX standard does not define
what happens when you reference a negative field number. @code{gawk}
will notice this and terminate your program. Other @code{awk}
implementations may behave differently.)
As mentioned in @ref{Fields, ,Examining Fields},
the number of fields in the current record is stored in the built-in
variable @code{NF} (also @pxref{Built-in Variables}). The expression
@code{$NF} is not a special feature: it is the direct consequence of
evaluating @code{NF} and using its value as a field number.
@node Changing Fields, Field Separators, Non-Constant Fields, Reading Files
@section Changing the Contents of a Field
@cindex field, changing contents of
@cindex changing contents of a field
@cindex assignment to fields
You can change the contents of a field as seen by @code{awk} within an
@code{awk} program; this changes what @code{awk} perceives as the
current input record. (The actual input is untouched; @code{awk} @emph{never}
modifies the input file.)
Consider this example and its output:
@example
@group
$ awk '@{ $3 = $2 - 10; print $2, $3 @}' inventory-shipped
@print{} 13 3
@print{} 15 5
@print{} 15 5
@dots{}
@end group
@end example
@noindent
The @samp{-} sign represents subtraction, so this program reassigns
field three, @code{$3}, to be the value of field two minus ten,
@samp{$2 - 10}. (@xref{Arithmetic Ops, ,Arithmetic Operators}.)
Then field two, and the new value for field three, are printed.
In order for this to work, the text in field @code{$2} must make sense
as a number; the string of characters must be converted to a number in
order for the computer to do arithmetic on it. The number resulting
from the subtraction is converted back to a string of characters which
then becomes field three.
@xref{Conversion, ,Conversion of Strings and Numbers}.
When you change the value of a field (as perceived by @code{awk}), the
text of the input record is recalculated to contain the new field where
the old one was. Therefore, @code{$0} changes to reflect the altered
field. Thus, this program
prints a copy of the input file, with 10 subtracted from the second
field of each line.
@example
@group
$ awk '@{ $2 = $2 - 10; print $0 @}' inventory-shipped
@print{} Jan 3 25 15 115
@print{} Feb 5 32 24 226
@print{} Mar 5 24 34 228
@dots{}
@end group
@end example
You can also assign contents to fields that are out of range. For
example:
@example
$ awk '@{ $6 = ($5 + $4 + $3 + $2)
> print $6 @}' inventory-shipped
@print{} 168
@print{} 297
@print{} 301
@dots{}
@end example
@noindent
We've just created @code{$6}, whose value is the sum of fields
@code{$2}, @code{$3}, @code{$4}, and @code{$5}. The @samp{+} sign
represents addition. For the file @file{inventory-shipped}, @code{$6}
represents the total number of parcels shipped for a particular month.
Creating a new field changes @code{awk}'s internal copy of the current
input record---the value of @code{$0}. Thus, if you do @samp{print $0}
after adding a field, the record printed includes the new field, with
the appropriate number of field separators between it and the previously
existing fields.
This recomputation affects and is affected by
@code{NF} (the number of fields; @pxref{Fields, ,Examining Fields}),
and by a feature that has not been discussed yet,
the @dfn{output field separator}, @code{OFS},
which is used to separate the fields (@pxref{Output Separators}).
For example, the value of @code{NF} is set to the number of the highest
field you create.
Note, however, that merely @emph{referencing} an out-of-range field
does @emph{not} change the value of either @code{$0} or @code{NF}.
Referencing an out-of-range field only produces an empty string. For
example:
@example
if ($(NF+1) != "")
print "can't happen"
else
print "everything is normal"
@end example
@noindent
should print @samp{everything is normal}, because @code{NF+1} is certain
to be out of range. (@xref{If Statement, ,The @code{if}-@code{else} Statement},
for more information about @code{awk}'s @code{if-else} statements.
@xref{Typing and Comparison, ,Variable Typing and Comparison Expressions},
for more information about the @samp{!=} operator.)
It is important to note that making an assignment to an existing field
will change the
value of @code{$0}, but will not change the value of @code{NF},
even when you assign the empty string to a field. For example:
@example
@group
$ echo a b c d | awk '@{ OFS = ":"; $2 = ""
> print $0; print NF @}'
@print{} a::c:d
@print{} 4
@end group
@end example
@noindent
The field is still there; it just has an empty value. You can tell
because there are two colons in a row.
This example shows what happens if you create a new field.
@example
$ echo a b c d | awk '@{ OFS = ":"; $2 = ""; $6 = "new"
> print $0; print NF @}'
@print{} a::c:d::new
@print{} 6
@end example
@noindent
The intervening field, @code{$5} is created with an empty value
(indicated by the second pair of adjacent colons),
and @code{NF} is updated with the value six.
Finally, decrementing @code{NF} will lose the values of the fields
after the new value of @code{NF}, and @code{$0} will be recomputed.
Here is an example:
@example
$ echo a b c d e f | ../gawk '@{ print "NF =", NF;
> NF = 3; print $0 @}'
@print{} NF = 6
@print{} a b c
@end example
@node Field Separators, Constant Size, Changing Fields, Reading Files
@section Specifying How Fields are Separated
This section is rather long; it describes one of the most fundamental
operations in @code{awk}.
@menu
* Basic Field Splitting:: How fields are split with single characters
or simple strings.
* Regexp Field Splitting:: Using regexps as the field separator.
* Single Character Fields:: Making each character a separate field.
* Command Line Field Separator:: Setting @code{FS} from the command line.
* Field Splitting Summary:: Some final points and a summary table.
@end menu
@node Basic Field Splitting, Regexp Field Splitting, Field Separators, Field Separators
@subsection The Basics of Field Separating
@vindex FS
@cindex fields, separating
@cindex field separator, @code{FS}
The @dfn{field separator}, which is either a single character or a regular
expression, controls the way @code{awk} splits an input record into fields.
@code{awk} scans the input record for character sequences that
match the separator; the fields themselves are the text between the matches.
In the examples below, we use the bullet symbol ``@bullet{}'' to represent
spaces in the output.
If the field separator is @samp{oo}, then the following line:
@example
moo goo gai pan
@end example
@noindent
would be split into three fields: @samp{m}, @samp{@bullet{}g} and
@samp{@bullet{}gai@bullet{}pan}.
Note the leading spaces in the values of the second and third fields.
@cindex common mistakes
@cindex mistakes, common
@cindex errors, common
The field separator is represented by the built-in variable @code{FS}.
Shell programmers take note! @code{awk} does @emph{not} use the name @code{IFS}
which is used by the POSIX compatible shells (such as the Bourne shell,
@code{sh}, or the GNU Bourne-Again Shell, Bash).
You can change the value of @code{FS} in the @code{awk} program with the
assignment operator, @samp{=} (@pxref{Assignment Ops, ,Assignment Expressions}).
Often the right time to do this is at the beginning of execution,
before any input has been processed, so that the very first record
will be read with the proper separator. To do this, use the special
@code{BEGIN} pattern
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}).
For example, here we set the value of @code{FS} to the string
@code{","}:
@example
awk 'BEGIN @{ FS = "," @} ; @{ print $2 @}'
@end example
@noindent
Given the input line,
@example
John Q. Smith, 29 Oak St., Walamazoo, MI 42139
@end example
@noindent
this @code{awk} program extracts and prints the string
@samp{@bullet{}29@bullet{}Oak@bullet{}St.}.
@cindex field separator, choice of
@cindex regular expressions as field separators
Sometimes your input data will contain separator characters that don't
separate fields the way you thought they would. For instance, the
person's name in the example we just used might have a title or
suffix attached, such as @samp{John Q. Smith, LXIX}. From input
containing such a name:
@example
John Q. Smith, LXIX, 29 Oak St., Walamazoo, MI 42139
@end example
@noindent
@c careful of an overfull hbox here!
the above program would extract @samp{@bullet{}LXIX}, instead of
@samp{@bullet{}29@bullet{}Oak@bullet{}St.}.
If you were expecting the program to print the
address, you would be surprised. The moral is: choose your data layout and
separator characters carefully to prevent such problems.
@iftex
As you know, normally,
@end iftex
@ifinfo
Normally,
@end ifinfo
fields are separated by whitespace sequences
(spaces, tabs and newlines), not by single spaces: two spaces in a row do not
delimit an empty field. The default value of the field separator @code{FS}
is a string containing a single space, @w{@code{" "}}. If this value were
interpreted in the usual way, each space character would separate
fields, so two spaces in a row would make an empty field between them.
The reason this does not happen is that a single space as the value of
@code{FS} is a special case: it is taken to specify the default manner
of delimiting fields.
If @code{FS} is any other single character, such as @code{","}, then
each occurrence of that character separates two fields. Two consecutive
occurrences delimit an empty field. If the character occurs at the
beginning or the end of the line, that too delimits an empty field. The
space character is the only single character which does not follow these
rules.
@node Regexp Field Splitting, Single Character Fields, Basic Field Splitting, Field Separators
@subsection Using Regular Expressions to Separate Fields
The previous
@iftex
subsection
@end iftex
@ifinfo
node
@end ifinfo
discussed the use of single characters or simple strings as the
value of @code{FS}.
More generally, the value of @code{FS} may be a string containing any
regular expression. In this case, each match in the record for the regular
expression separates fields. For example, the assignment:
@example
FS = ", \t"
@end example
@noindent
makes every area of an input line that consists of a comma followed by a
space and a tab, into a field separator. (@samp{\t}
is an @dfn{escape sequence} that stands for a tab;
@pxref{Escape Sequences},
for the complete list of similar escape sequences.)
For a less trivial example of a regular expression, suppose you want
single spaces to separate fields the way single commas were used above.
You can set @code{FS} to @w{@code{"[@ ]"}} (left bracket, space, right
bracket). This regular expression matches a single space and nothing else
(@pxref{Regexp, ,Regular Expressions}).
There is an important difference between the two cases of @samp{FS = @w{" "}}
(a single space) and @samp{FS = @w{"[ \t\n]+"}} (left bracket, space,
backslash, ``t'', backslash, ``n'', right bracket, which is a regular
expression matching one or more spaces, tabs, or newlines). For both
values of @code{FS}, fields are separated by runs of spaces, tabs
and/or newlines. However, when the value of @code{FS} is @w{@code{"
"}}, @code{awk} will first strip leading and trailing whitespace from
the record, and then decide where the fields are.
For example, the following pipeline prints @samp{b}:
@example
$ echo ' a b c d ' | awk '@{ print $2 @}'
@print{} b
@end example
@noindent
However, this pipeline prints @samp{a} (note the extra spaces around
each letter):
@example
$ echo ' a b c d ' | awk 'BEGIN @{ FS = "[ \t]+" @}
> @{ print $2 @}'
@print{} a
@end example
@noindent
@cindex null string
@cindex empty string
In this case, the first field is @dfn{null}, or empty.
The stripping of leading and trailing whitespace also comes into
play whenever @code{$0} is recomputed. For instance, study this pipeline:
@example
$ echo ' a b c d' | awk '@{ print; $2 = $2; print @}'
@print{} a b c d
@print{} a b c d
@end example
@noindent
The first @code{print} statement prints the record as it was read,
with leading whitespace intact. The assignment to @code{$2} rebuilds
@code{$0} by concatenating @code{$1} through @code{$NF} together,
separated by the value of @code{OFS}. Since the leading whitespace
was ignored when finding @code{$1}, it is not part of the new @code{$0}.
Finally, the last @code{print} statement prints the new @code{$0}.
@node Single Character Fields, Command Line Field Separator, Regexp Field Splitting, Field Separators
@subsection Making Each Character a Separate Field
@cindex differences between @code{gawk} and @code{awk}
@cindex single character fields
There are times when you may want to examine each character
of a record separately. In @code{gawk}, this is easy to do, you
simply assign the null string (@code{""}) to @code{FS}. In this case,
each individual character in the record will become a separate field.
Here is an example:
@example
@group
$ echo a b | gawk 'BEGIN @{ FS = "" @}
> @{
> for (i = 1; i <= NF; i = i + 1)
> print "Field", i, "is", $i
> @}'
@print{} Field 1 is a
@print{} Field 2 is
@print{} Field 3 is b
@end group
@end example
@cindex dark corner
Traditionally, the behavior for @code{FS} equal to @code{""} was not defined.
In this case, Unix @code{awk} would simply treat the entire record
as only having one field (d.c.). In compatibility mode
(@pxref{Options, ,Command Line Options}),
if @code{FS} is the null string, then @code{gawk} will also
behave this way.
@node Command Line Field Separator, Field Splitting Summary, Single Character Fields, Field Separators
@subsection Setting @code{FS} from the Command Line
@cindex @code{-F} option
@cindex field separator, on command line
@cindex command line, setting @code{FS} on
@code{FS} can be set on the command line. You use the @samp{-F} option to
do so. For example:
@example
awk -F, '@var{program}' @var{input-files}
@end example
@noindent
sets @code{FS} to be the @samp{,} character. Notice that the option uses
a capital @samp{F}. Contrast this with @samp{-f}, which specifies a file
containing an @code{awk} program. Case is significant in command line options:
the @samp{-F} and @samp{-f} options have nothing to do with each other.
You can use both options at the same time to set the @code{FS} variable
@emph{and} get an @code{awk} program from a file.
The value used for the argument to @samp{-F} is processed in exactly the
same way as assignments to the built-in variable @code{FS}. This means that
if the field separator contains special characters, they must be escaped
appropriately. For example, to use a @samp{\} as the field separator, you
would have to type:
@example
# same as FS = "\\"
awk -F\\\\ '@dots{}' files @dots{}
@end example
@noindent
Since @samp{\} is used for quoting in the shell, @code{awk} will see
@samp{-F\\}. Then @code{awk} processes the @samp{\\} for escape
characters (@pxref{Escape Sequences}), finally yielding
a single @samp{\} to be used for the field separator.
@cindex historical features
As a special case, in compatibility mode
(@pxref{Options, ,Command Line Options}), if the
argument to @samp{-F} is @samp{t}, then @code{FS} is set to the tab
character. This is because if you type @samp{-F\t} at the shell,
without any quotes, the @samp{\} gets deleted, so @code{awk} figures that you
really want your fields to be separated with tabs, and not @samp{t}s.
Use @samp{-v FS="t"} on the command line if you really do want to separate
your fields with @samp{t}s
(@pxref{Options, ,Command Line Options}).
For example, let's use an @code{awk} program file called @file{baud.awk}
that contains the pattern @code{/300/}, and the action @samp{print $1}.
Here is the program:
@example
/300/ @{ print $1 @}
@end example
Let's also set @code{FS} to be the @samp{-} character, and run the
program on the file @file{BBS-list}. The following command prints a
list of the names of the bulletin boards that operate at 300 baud and
the first three digits of their phone numbers:
@c tweaked to make the tex output look better in @smallbook
@example
@group
$ awk -F- -f baud.awk BBS-list
@print{} aardvark 555
@print{} alpo
@print{} barfly 555
@dots{}
@end group
@ignore
@print{} bites 555
@print{} camelot 555
@print{} core 555
@print{} fooey 555
@print{} foot 555
@print{} macfoo 555
@print{} sdace 555
@print{} sabafoo 555
@end ignore
@end example
@noindent
Note the second line of output. In the original file
(@pxref{Sample Data Files, ,Data Files for the Examples}),
the second line looked like this:
@example
alpo-net 555-3412 2400/1200/300 A
@end example
The @samp{-} as part of the system's name was used as the field
separator, instead of the @samp{-} in the phone number that was
originally intended. This demonstrates why you have to be careful in
choosing your field and record separators.
On many Unix systems, each user has a separate entry in the system password
file, one line per user. The information in these lines is separated
by colons. The first field is the user's logon name, and the second is
the user's encrypted password. A password file entry might look like this:
@example
arnold:xyzzy:2076:10:Arnold Robbins:/home/arnold:/bin/sh
@end example
The following program searches the system password file, and prints
the entries for users who have no password:
@example
awk -F: '$2 == ""' /etc/passwd
@end example
@node Field Splitting Summary, , Command Line Field Separator, Field Separators
@subsection Field Splitting Summary
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
According to the POSIX standard, @code{awk} is supposed to behave
as if each record is split into fields at the time that it is read.
In particular, this means that you can change the value of @code{FS}
after a record is read, and the value of the fields (i.e.@: how they were split)
should reflect the old value of @code{FS}, not the new one.
@cindex dark corner
@cindex @code{sed} utility
@cindex stream editor
However, many implementations of @code{awk} do not work this way. Instead,
they defer splitting the fields until a field is actually
referenced. The fields will be split
using the @emph{current} value of @code{FS}! (d.c.)
This behavior can be difficult
to diagnose. The following example illustrates the difference
between the two methods.
(The @code{sed}@footnote{The @code{sed} utility is a ``stream editor.''
Its behavior is also defined by the POSIX standard.}
command prints just the first line of @file{/etc/passwd}.)
@example
sed 1q /etc/passwd | awk '@{ FS = ":" ; print $1 @}'
@end example
@noindent
will usually print
@example
root
@end example
@noindent
on an incorrect implementation of @code{awk}, while @code{gawk}
will print something like
@example
root:nSijPlPhZZwgE:0:0:Root:/:
@end example
The following table summarizes how fields are split, based on the
value of @code{FS}. (@samp{==} means ``is equal to.'')
@c @cartouche
@table @code
@item FS == " "
Fields are separated by runs of whitespace. Leading and trailing
whitespace are ignored. This is the default.
@item FS == @var{any other single character}
Fields are separated by each occurrence of the character. Multiple
successive occurrences delimit empty fields, as do leading and
trailing occurrences.
The character can even be a regexp metacharacter; it does not need
to be escaped.
@item FS == @var{regexp}
Fields are separated by occurrences of characters that match @var{regexp}.
Leading and trailing matches of @var{regexp} delimit empty fields.
@item FS == ""
Each individual character in the record becomes a separate field.
@end table
@c @end cartouche
@node Constant Size, Multiple Line, Field Separators, Reading Files
@section Reading Fixed-width Data
(This section discusses an advanced, experimental feature. If you are
a novice @code{awk} user, you may wish to skip it on the first reading.)
@code{gawk} version 2.13 introduced a new facility for dealing with
fixed-width fields with no distinctive field separator. Data of this
nature arises, for example, in the input for old FORTRAN programs where
numbers are run together; or in the output of programs that did not
anticipate the use of their output as input for other programs.
An example of the latter is a table where all the columns are lined up by
the use of a variable number of spaces and @emph{empty fields are just
spaces}. Clearly, @code{awk}'s normal field splitting based on @code{FS}
will not work well in this case. Although a portable @code{awk} program
can use a series of @code{substr} calls on @code{$0}
(@pxref{String Functions, ,Built-in Functions for String Manipulation}),
this is awkward and inefficient for a large number of fields.
The splitting of an input record into fixed-width fields is specified by
assigning a string containing space-separated numbers to the built-in
variable @code{FIELDWIDTHS}. Each number specifies the width of the field
@emph{including} columns between fields. If you want to ignore the columns
between fields, you can specify the width as a separate field that is
subsequently ignored.
The following data is the output of the Unix @code{w} utility. It is useful
to illustrate the use of @code{FIELDWIDTHS}.
@example
@group
10:06pm up 21 days, 14:04, 23 users
User tty login@ idle JCPU PCPU what
hzuo ttyV0 8:58pm 9 5 vi p24.tex
hzang ttyV3 6:37pm 50 -csh
eklye ttyV5 9:53pm 7 1 em thes.tex
dportein ttyV6 8:17pm 1:47 -csh
gierd ttyD3 10:00pm 1 elm
dave ttyD4 9:47pm 4 4 w
brent ttyp0 26Jun91 4:46 26:46 4:41 bash
dave ttyq4 26Jun9115days 46 46 wnewmail
@end group
@end example
The following program takes the above input, converts the idle time to
number of seconds and prints out the first two fields and the calculated
idle time. (This program uses a number of @code{awk} features that
haven't been introduced yet.)
@example
@group
BEGIN @{ FIELDWIDTHS = "9 6 10 6 7 7 35" @}
NR > 2 @{
idle = $4
sub(/^ */, "", idle) # strip leading spaces
if (idle == "")
idle = 0
if (idle ~ /:/) @{
split(idle, t, ":")
idle = t[1] * 60 + t[2]
@}
if (idle ~ /days/)
idle *= 24 * 60 * 60
print $1, $2, idle
@}
@end group
@end example
Here is the result of running the program on the data:
@example
hzuo ttyV0 0
hzang ttyV3 50
eklye ttyV5 0
dportein ttyV6 107
gierd ttyD3 1
dave ttyD4 0
brent ttyp0 286
dave ttyq4 1296000
@end example
Another (possibly more practical) example of fixed-width input data
would be the input from a deck of balloting cards. In some parts of
the United States, voters mark their choices by punching holes in computer
cards. These cards are then processed to count the votes for any particular
candidate or on any particular issue. Since a voter may choose not to
vote on some issue, any column on the card may be empty. An @code{awk}
program for processing such data could use the @code{FIELDWIDTHS} feature
to simplify reading the data. (Of course, getting @code{gawk} to run on
a system with card readers is another story!)
@ignore
Exercise: Write a ballot card reading program
@end ignore
Assigning a value to @code{FS} causes @code{gawk} to return to using
@code{FS} for field splitting. Use @samp{FS = FS} to make this happen,
without having to know the current value of @code{FS}.
This feature is still experimental, and may evolve over time.
Note that in particular, @code{gawk} does not attempt to verify
the sanity of the values used in the value of @code{FIELDWIDTHS}.
@node Multiple Line, Getline, Constant Size, Reading Files
@section Multiple-Line Records
@cindex multiple line records
@cindex input, multiple line records
@cindex reading files, multiple line records
@cindex records, multiple line
In some data bases, a single line cannot conveniently hold all the
information in one entry. In such cases, you can use multi-line
records.
The first step in doing this is to choose your data format: when records
are not defined as single lines, how do you want to define them?
What should separate records?
One technique is to use an unusual character or string to separate
records. For example, you could use the formfeed character (written
@samp{\f} in @code{awk}, as in C) to separate them, making each record
a page of the file. To do this, just set the variable @code{RS} to
@code{"\f"} (a string containing the formfeed character). Any
other character could equally well be used, as long as it won't be part
of the data in a record.
Another technique is to have blank lines separate records. By a special
dispensation, an empty string as the value of @code{RS} indicates that
records are separated by one or more blank lines. If you set @code{RS}
to the empty string, a record always ends at the first blank line
encountered. And the next record doesn't start until the first non-blank
line that follows---no matter how many blank lines appear in a row, they
are considered one record-separator.
@cindex leftmost longest match
@cindex matching, leftmost longest
You can achieve the same effect as @samp{RS = ""} by assigning the
string @code{"\n\n+"} to @code{RS}. This regexp matches the newline
at the end of the record, and one or more blank lines after the record.
In addition, a regular expression always matches the longest possible
sequence when there is a choice
(@pxref{Leftmost Longest, ,How Much Text Matches?}).
So the next record doesn't start until
the first non-blank line that follows---no matter how many blank lines
appear in a row, they are considered one record-separator.
@cindex dark corner
There is an important difference between @samp{RS = ""} and
@samp{RS = "\n\n+"}. In the first case, leading newlines in the input
data file are ignored, and if a file ends without extra blank lines
after the last record, the final newline is removed from the record.
In the second case, this special processing is not done (d.c.).
Now that the input is separated into records, the second step is to
separate the fields in the record. One way to do this is to divide each
of the lines into fields in the normal manner. This happens by default
as the result of a special feature: when @code{RS} is set to the empty
string, the newline character @emph{always} acts as a field separator.
This is in addition to whatever field separations result from @code{FS}.
The original motivation for this special exception was probably to provide
useful behavior in the default case (i.e.@: @code{FS} is equal
to @w{@code{" "}}). This feature can be a problem if you really don't
want the newline character to separate fields, since there is no way to
prevent it. However, you can work around this by using the @code{split}
function to break up the record manually
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
Another way to separate fields is to
put each field on a separate line: to do this, just set the
variable @code{FS} to the string @code{"\n"}. (This simple regular
expression matches a single newline.)
A practical example of a data file organized this way might be a mailing
list, where each entry is separated by blank lines. If we have a mailing
list in a file named @file{addresses}, that looks like this:
@example
Jane Doe
123 Main Street
Anywhere, SE 12345-6789
John Smith
456 Tree-lined Avenue
Smallville, MW 98765-4321
@dots{}
@end example
@noindent
A simple program to process this file would look like this:
@example
@group
# addrs.awk --- simple mailing list program
# Records are separated by blank lines.
# Each line is one field.
BEGIN @{ RS = "" ; FS = "\n" @}
@{
print "Name is:", $1
print "Address is:", $2
print "City and State are:", $3
print ""
@}
@end group
@end example
Running the program produces the following output:
@example
@group
$ awk -f addrs.awk addresses
@print{} Name is: Jane Doe
@print{} Address is: 123 Main Street
@print{} City and State are: Anywhere, SE 12345-6789
@print{}
@end group
@group
@print{} Name is: John Smith
@print{} Address is: 456 Tree-lined Avenue
@print{} City and State are: Smallville, MW 98765-4321
@print{}
@dots{}
@end group
@end example
@xref{Labels Program, ,Printing Mailing Labels}, for a more realistic
program that deals with address lists.
The following table summarizes how records are split, based on the
value of @code{RS}. (@samp{==} means ``is equal to.'')
@c @cartouche
@table @code
@item RS == "\n"
Records are separated by the newline character (@samp{\n}). In effect,
every line in the data file is a separate record, including blank lines.
This is the default.
@item RS == @var{any single character}
Records are separated by each occurrence of the character. Multiple
successive occurrences delimit empty records.
@item RS == ""
Records are separated by runs of blank lines. The newline character
always serves as a field separator, in addition to whatever value
@code{FS} may have. Leading and trailing newlines in a file are ignored.
@item RS == @var{regexp}
Records are separated by occurrences of characters that match @var{regexp}.
Leading and trailing matches of @var{regexp} delimit empty records.
@end table
@c @end cartouche
@vindex RT
In all cases, @code{gawk} sets @code{RT} to the input text that matched the
value specified by @code{RS}.
@node Getline, , Multiple Line, Reading Files
@section Explicit Input with @code{getline}
@findex getline
@cindex input, explicit
@cindex explicit input
@cindex input, @code{getline} command
@cindex reading files, @code{getline} command
So far we have been getting our input data from @code{awk}'s main
input stream---either the standard input (usually your terminal, sometimes
the output from another program) or from the
files specified on the command line. The @code{awk} language has a
special built-in command called @code{getline} that
can be used to read input under your explicit control.
@menu
* Getline Intro:: Introduction to the @code{getline} function.
* Plain Getline:: Using @code{getline} with no arguments.
* Getline/Variable:: Using @code{getline} into a variable.
* Getline/File:: Using @code{getline} from a file.
* Getline/Variable/File:: Using @code{getline} into a variable from a
file.
* Getline/Pipe:: Using @code{getline} from a pipe.
* Getline/Variable/Pipe:: Using @code{getline} into a variable from a
pipe.
* Getline Summary:: Summary Of @code{getline} Variants.
@end menu
@node Getline Intro, Plain Getline, Getline, Getline
@subsection Introduction to @code{getline}
This command is used in several different ways, and should @emph{not} be
used by beginners. It is covered here because this is the chapter on input.
The examples that follow the explanation of the @code{getline} command
include material that has not been covered yet. Therefore, come back
and study the @code{getline} command @emph{after} you have reviewed the
rest of this @value{DOCUMENT} and have a good knowledge of how @code{awk} works.
@vindex ERRNO
@cindex differences between @code{gawk} and @code{awk}
@cindex @code{getline}, return values
@code{getline} returns one if it finds a record, and zero if the end of the
file is encountered. If there is some error in getting a record, such
as a file that cannot be opened, then @code{getline} returns @minus{}1.
In this case, @code{gawk} sets the variable @code{ERRNO} to a string
describing the error that occurred.
In the following examples, @var{command} stands for a string value that
represents a shell command.
@node Plain Getline, Getline/Variable, Getline Intro, Getline
@subsection Using @code{getline} with No Arguments
The @code{getline} command can be used without arguments to read input
from the current input file. All it does in this case is read the next
input record and split it up into fields. This is useful if you've
finished processing the current record, but you want to do some special
processing @emph{right now} on the next record. Here's an
example:
@example
@group
awk '@{
if ((t = index($0, "/*")) != 0) @{
# value will be "" if t is 1
tmp = substr($0, 1, t - 1)
u = index(substr($0, t + 2), "*/")
while (u == 0) @{
if (getline <= 0) @{
m = "unexpected EOF or error"
m = (m ": " ERRNO)
print m > "/dev/stderr"
exit
@}
t = -1
u = index($0, "*/")
@}
@end group
@group
# substr expression will be "" if */
# occurred at end of line
$0 = tmp substr($0, t + u + 3)
@}
print $0
@}'
@end group
@end example
This @code{awk} program deletes all C-style comments, @samp{/* @dots{}
*/}, from the input. By replacing the @samp{print $0} with other
statements, you could perform more complicated processing on the
decommented input, like searching for matches of a regular
expression. This program has a subtle problem---it does not work if one
comment ends and another begins on the same line.
@ignore
Exercise,
write a program that does handle multiple comments on the line.
@end ignore
This form of the @code{getline} command sets @code{NF} (the number of
fields; @pxref{Fields, ,Examining Fields}), @code{NR} (the number of
records read so far; @pxref{Records, ,How Input is Split into Records}),
@code{FNR} (the number of records read from this input file), and the
value of @code{$0}.
@cindex dark corner
@strong{Note:} the new value of @code{$0} is used in testing
the patterns of any subsequent rules. The original value
of @code{$0} that triggered the rule which executed @code{getline}
is lost (d.c.).
By contrast, the @code{next} statement reads a new record
but immediately begins processing it normally, starting with the first
rule in the program. @xref{Next Statement, ,The @code{next} Statement}.
@node Getline/Variable, Getline/File, Plain Getline, Getline
@subsection Using @code{getline} Into a Variable
You can use @samp{getline @var{var}} to read the next record from
@code{awk}'s input into the variable @var{var}. No other processing is
done.
For example, suppose the next line is a comment, or a special string,
and you want to read it, without triggering
any rules. This form of @code{getline} allows you to read that line
and store it in a variable so that the main
read-a-line-and-check-each-rule loop of @code{awk} never sees it.
The following example swaps every two lines of input. For example, given:
@example
wan
tew
free
phore
@end example
@noindent
it outputs:
@example
tew
wan
phore
free
@end example
@noindent
Here's the program:
@example
@group
awk '@{
if ((getline tmp) > 0) @{
print tmp
print $0
@} else
print $0
@}'
@end group
@end example
The @code{getline} command used in this way sets only the variables
@code{NR} and @code{FNR} (and of course, @var{var}). The record is not
split into fields, so the values of the fields (including @code{$0}) and
the value of @code{NF} do not change.
@node Getline/File, Getline/Variable/File, Getline/Variable, Getline
@subsection Using @code{getline} from a File
@cindex input redirection
@cindex redirection of input
Use @samp{getline < @var{file}} to read
the next record from the file
@var{file}. Here @var{file} is a string-valued expression that
specifies the file name. @samp{< @var{file}} is called a @dfn{redirection}
since it directs input to come from a different place.
For example, the following
program reads its input record from the file @file{secondary.input} when it
encounters a first field with a value equal to 10 in the current input
file.
@example
@group
awk '@{
if ($1 == 10) @{
getline < "secondary.input"
print
@} else
print
@}'
@end group
@end example
Since the main input stream is not used, the values of @code{NR} and
@code{FNR} are not changed. But the record read is split into fields in
the normal manner, so the values of @code{$0} and other fields are
changed. So is the value of @code{NF}.
@c Thanks to Paul Eggert for initial wording here
According to POSIX, @samp{getline < @var{expression}} is ambiguous if
@var{expression} contains unparenthesized operators other than
@samp{$}; for example, @samp{getline < dir "/" file} is ambiguous
because the concatenation operator is not parenthesized, and you should
write it as @samp{getline < (dir "/" file)} if you want your program
to be portable to other @code{awk} implementations.
@node Getline/Variable/File, Getline/Pipe, Getline/File, Getline
@subsection Using @code{getline} Into a Variable from a File
Use @samp{getline @var{var} < @var{file}} to read input
the file
@var{file} and put it in the variable @var{var}. As above, @var{file}
is a string-valued expression that specifies the file from which to read.
In this version of @code{getline}, none of the built-in variables are
changed, and the record is not split into fields. The only variable
changed is @var{var}.
@ifinfo
@c Thanks to Paul Eggert for initial wording here
According to POSIX, @samp{getline @var{var} < @var{expression}} is ambiguous if
@var{expression} contains unparenthesized operators other than
@samp{$}; for example, @samp{getline < dir "/" file} is ambiguous
because the concatenation operator is not parenthesized, and you should
write it as @samp{getline < (dir "/" file)} if you want your program
to be portable to other @code{awk} implementations.
@end ifinfo
For example, the following program copies all the input files to the
output, except for records that say @w{@samp{@@include @var{filename}}}.
Such a record is replaced by the contents of the file
@var{filename}.
@example
@group
awk '@{
if (NF == 2 && $1 == "@@include") @{
while ((getline line < $2) > 0)
print line
close($2)
@} else
print
@}'
@end group
@end example
Note here how the name of the extra input file is not built into
the program; it is taken directly from the data, from the second field on
the @samp{@@include} line.
The @code{close} function is called to ensure that if two identical
@samp{@@include} lines appear in the input, the entire specified file is
included twice.
@xref{Close Files And Pipes, ,Closing Input and Output Files and Pipes}.
One deficiency of this program is that it does not process nested
@samp{@@include} statements
(@samp{@@include} statements in included files)
the way a true macro preprocessor would.
@xref{Igawk Program, ,An Easy Way to Use Library Functions}, for a program
that does handle nested @samp{@@include} statements.
@node Getline/Pipe, Getline/Variable/Pipe, Getline/Variable/File, Getline
@subsection Using @code{getline} from a Pipe
@cindex input pipeline
@cindex pipeline, input
You can pipe the output of a command into @code{getline}, using
@samp{@var{command} | getline}. In
this case, the string @var{command} is run as a shell command and its output
is piped into @code{awk} to be used as input. This form of @code{getline}
reads one record at a time from the pipe.
For example, the following program copies its input to its output, except for
lines that begin with @samp{@@execute}, which are replaced by the output
produced by running the rest of the line as a shell command:
@example
@group
awk '@{
if ($1 == "@@execute") @{
tmp = substr($0, 10)
while ((tmp | getline) > 0)
print
close(tmp)
@} else
print
@}'
@end group
@end example
@noindent
The @code{close} function is called to ensure that if two identical
@samp{@@execute} lines appear in the input, the command is run for
each one.
@xref{Close Files And Pipes, ,Closing Input and Output Files and Pipes}.
@c Exercise!!
@c This example is unrealistic, since you could just use system
@c NEEDED
@page
Given the input:
@example
@group
foo
bar
baz
@@execute who
bletch
@end group
@end example
@noindent
the program might produce:
@example
@group
foo
bar
baz
arnold ttyv0 Jul 13 14:22
miriam ttyp0 Jul 13 14:23 (murphy:0)
bill ttyp1 Jul 13 14:23 (murphy:0)
bletch
@end group
@end example
@noindent
Notice that this program ran the command @code{who} and printed the result.
(If you try this program yourself, you will of course get different results,
showing you who is logged in on your system.)
This variation of @code{getline} splits the record into fields, sets the
value of @code{NF} and recomputes the value of @code{$0}. The values of
@code{NR} and @code{FNR} are not changed.
@c Thanks to Paul Eggert for initial wording here
According to POSIX, @samp{@var{expression} | getline} is ambiguous if
@var{expression} contains unparenthesized operators other than
@samp{$}; for example, @samp{"echo " "date" | getline} is ambiguous
because the concatenation operator is not parenthesized, and you should
write it as @samp{("echo " "date") | getline} if you want your program
to be portable to other @code{awk} implementations.
@node Getline/Variable/Pipe, Getline Summary, Getline/Pipe, Getline
@subsection Using @code{getline} Into a Variable from a Pipe
When you use @samp{@var{command} | getline @var{var}}, the
output of the command @var{command} is sent through a pipe to
@code{getline} and into the variable @var{var}. For example, the
following program reads the current date and time into the variable
@code{current_time}, using the @code{date} utility, and then
prints it.
@example
@group
awk 'BEGIN @{
"date" | getline current_time
close("date")
print "Report printed on " current_time
@}'
@end group
@end example
In this version of @code{getline}, none of the built-in variables are
changed, and the record is not split into fields.
@ifinfo
@c Thanks to Paul Eggert for initial wording here
According to POSIX, @samp{@var{expression} | getline @var{var}} is ambiguous if
@var{expression} contains unparenthesized operators other than
@samp{$}; for example, @samp{"echo " "date" | getline @var{var}} is ambiguous
because the concatenation operator is not parenthesized, and you should
write it as @samp{("echo " "date") | getline @var{var}} if you want your
program to be portable to other @code{awk} implementations.
@end ifinfo
@node Getline Summary, , Getline/Variable/Pipe, Getline
@subsection Summary of @code{getline} Variants
With all the forms of @code{getline}, even though @code{$0} and @code{NF},
may be updated, the record will not be tested against all the patterns
in the @code{awk} program, in the way that would happen if the record
were read normally by the main processing loop of @code{awk}. However
the new record is tested against any subsequent rules.
@cindex differences between @code{gawk} and @code{awk}
@cindex limitations
@cindex implementation limits
Many @code{awk} implementations limit the number of pipelines an @code{awk}
program may have open to just one! In @code{gawk}, there is no such limit.
You can open as many pipelines as the underlying operating system will
permit.
@vindex FILENAME
@cindex dark corner
@cindex @code{getline}, setting @code{FILENAME}
@cindex @code{FILENAME}, being set by @code{getline}
An interesting side-effect occurs if you use @code{getline} (without a
redirection) inside a @code{BEGIN} rule. Since an unredirected @code{getline}
reads from the command line data files, the first @code{getline} command
causes @code{awk} to set the value of @code{FILENAME}. Normally,
@code{FILENAME} does not have a value inside @code{BEGIN} rules, since you
have not yet started to process the command line data files (d.c.).
(@xref{BEGIN/END, , The @code{BEGIN} and @code{END} Special Patterns},
also @pxref{Auto-set, , Built-in Variables that Convey Information}.)
The following table summarizes the six variants of @code{getline},
listing which built-in variables are set by each one.
@c @cartouche
@table @code
@item getline
sets @code{$0}, @code{NF}, @code{FNR}, and @code{NR}.
@item getline @var{var}
sets @var{var}, @code{FNR}, and @code{NR}.
@item getline < @var{file}
sets @code{$0}, and @code{NF}.
@item getline @var{var} < @var{file}
sets @var{var}.
@item @var{command} | getline
sets @code{$0}, and @code{NF}.
@item @var{command} | getline @var{var}
sets @var{var}.
@end table
@c @end cartouche
@node Printing, Expressions, Reading Files, Top
@chapter Printing Output
@cindex printing
@cindex output
One of the most common actions is to @dfn{print}, or output,
some or all of the input. You use the @code{print} statement
for simple output. You use the @code{printf} statement
for fancier formatting. Both are described in this chapter.
@menu
* Print:: The @code{print} statement.
* Print Examples:: Simple examples of @code{print} statements.
* Output Separators:: The output separators and how to change them.
* OFMT:: Controlling Numeric Output With @code{print}.
* Printf:: The @code{printf} statement.
* Redirection:: How to redirect output to multiple files and
pipes.
* Special Files:: File name interpretation in @code{gawk}.
@code{gawk} allows access to inherited file
descriptors.
* Close Files And Pipes:: Closing Input and Output Files and Pipes.
@end menu
@node Print, Print Examples, Printing, Printing
@section The @code{print} Statement
@cindex @code{print} statement
The @code{print} statement does output with simple, standardized
formatting. You specify only the strings or numbers to be printed, in a
list separated by commas. They are output, separated by single spaces,
followed by a newline. The statement looks like this:
@example
print @var{item1}, @var{item2}, @dots{}
@end example
@noindent
The entire list of items may optionally be enclosed in parentheses. The
parentheses are necessary if any of the item expressions uses the @samp{>}
relational operator; otherwise it could be confused with a redirection
(@pxref{Redirection, ,Redirecting Output of @code{print} and @code{printf}}).
The items to be printed can be constant strings or numbers, fields of the
current record (such as @code{$1}), variables, or any @code{awk}
expressions.
Numeric values are converted to strings, and then printed.
The @code{print} statement is completely general for
computing @emph{what} values to print. However, with two exceptions,
you cannot specify @emph{how} to print them---how many
columns, whether to use exponential notation or not, and so on.
(For the exceptions, @pxref{Output Separators}, and
@ref{OFMT, ,Controlling Numeric Output with @code{print}}.)
For that, you need the @code{printf} statement
(@pxref{Printf, ,Using @code{printf} Statements for Fancier Printing}).
The simple statement @samp{print} with no items is equivalent to
@samp{print $0}: it prints the entire current record. To print a blank
line, use @samp{print ""}, where @code{""} is the empty string.
To print a fixed piece of text, use a string constant such as
@w{@code{"Don't Panic"}} as one item. If you forget to use the
double-quote characters, your text will be taken as an @code{awk}
expression, and you will probably get an error. Keep in mind that a
space is printed between any two items.
Each @code{print} statement makes at least one line of output. But it
isn't limited to one line. If an item value is a string that contains a
newline, the newline is output along with the rest of the string. A
single @code{print} can make any number of lines this way.
@node Print Examples, Output Separators, Print, Printing
@section Examples of @code{print} Statements
Here is an example of printing a string that contains embedded newlines
(the @samp{\n} is an escape sequence, used to represent the newline
character; see @ref{Escape Sequences}):
@example
@group
$ awk 'BEGIN @{ print "line one\nline two\nline three" @}'
@print{} line one
@print{} line two
@print{} line three
@end group
@end example
Here is an example that prints the first two fields of each input record,
with a space between them:
@example
@group
$ awk '@{ print $1, $2 @}' inventory-shipped
@print{} Jan 13
@print{} Feb 15
@print{} Mar 15
@dots{}
@end group
@end example
@cindex common mistakes
@cindex mistakes, common
@cindex errors, common
A common mistake in using the @code{print} statement is to omit the comma
between two items. This often has the effect of making the items run
together in the output, with no space. The reason for this is that
juxtaposing two string expressions in @code{awk} means to concatenate
them. Here is the same program, without the comma:
@example
@group
$ awk '@{ print $1 $2 @}' inventory-shipped
@print{} Jan13
@print{} Feb15
@print{} Mar15
@dots{}
@end group
@end example
To someone unfamiliar with the file @file{inventory-shipped}, neither
example's output makes much sense. A heading line at the beginning
would make it clearer. Let's add some headings to our table of months
(@code{$1}) and green crates shipped (@code{$2}). We do this using the
@code{BEGIN} pattern
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns})
to force the headings to be printed only once:
@example
awk 'BEGIN @{ print "Month Crates"
print "----- ------" @}
@{ print $1, $2 @}' inventory-shipped
@end example
@noindent
Did you already guess what happens? When run, the program prints
the following:
@example
@group
Month Crates
----- ------
Jan 13
Feb 15
Mar 15
@dots{}
@end group
@end example
@noindent
The headings and the table data don't line up! We can fix this by printing
some spaces between the two fields:
@example
awk 'BEGIN @{ print "Month Crates"
print "----- ------" @}
@{ print $1, " ", $2 @}' inventory-shipped
@end example
You can imagine that this way of lining up columns can get pretty
complicated when you have many columns to fix. Counting spaces for two
or three columns can be simple, but more than this and you can get
lost quite easily. This is why the @code{printf} statement was
created (@pxref{Printf, ,Using @code{printf} Statements for Fancier Printing});
one of its specialties is lining up columns of data.
@cindex line continuation
As a side point,
you can continue either a @code{print} or @code{printf} statement simply
by putting a newline after any comma
(@pxref{Statements/Lines, ,@code{awk} Statements Versus Lines}).
@node Output Separators, OFMT, Print Examples, Printing
@section Output Separators
@cindex output field separator, @code{OFS}
@cindex output record separator, @code{ORS}
@vindex OFS
@vindex ORS
As mentioned previously, a @code{print} statement contains a list
of items, separated by commas. In the output, the items are normally
separated by single spaces. This need not be the case; a
single space is only the default. You can specify any string of
characters to use as the @dfn{output field separator} by setting the
built-in variable @code{OFS}. The initial value of this variable
is the string @w{@code{" "}}, that is, a single space.
The output from an entire @code{print} statement is called an
@dfn{output record}. Each @code{print} statement outputs one output
record and then outputs a string called the @dfn{output record separator}.
The built-in variable @code{ORS} specifies this string. The initial
value of @code{ORS} is the string @code{"\n"}, i.e.@: a newline
character; thus, normally each @code{print} statement makes a separate line.
You can change how output fields and records are separated by assigning
new values to the variables @code{OFS} and/or @code{ORS}. The usual
place to do this is in the @code{BEGIN} rule
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}), so
that it happens before any input is processed. You may also do this
with assignments on the command line, before the names of your input
files, or using the @samp{-v} command line option
(@pxref{Options, ,Command Line Options}).
@ignore
Exercise,
Rewrite the
@example
awk 'BEGIN @{ print "Month Crates"
print "----- ------" @}
@{ print $1, " ", $2 @}' inventory-shipped
@end example
program by using a new value of @code{OFS}.
@end ignore
The following example prints the first and second fields of each input
record separated by a semicolon, with a blank line added after each
line:
@example
@group
$ awk 'BEGIN @{ OFS = ";"; ORS = "\n\n" @}
> @{ print $1, $2 @}' BBS-list
@print{} aardvark;555-5553
@print{}
@print{} alpo-net;555-3412
@print{}
@print{} barfly;555-7685
@dots{}
@end group
@end example
If the value of @code{ORS} does not contain a newline, all your output
will be run together on a single line, unless you output newlines some
other way.
@node OFMT, Printf, Output Separators, Printing
@section Controlling Numeric Output with @code{print}
@vindex OFMT
@cindex numeric output format
@cindex format, numeric output
@cindex output format specifier, @code{OFMT}
When you use the @code{print} statement to print numeric values,
@code{awk} internally converts the number to a string of characters,
and prints that string. @code{awk} uses the @code{sprintf} function
to do this conversion
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
For now, it suffices to say that the @code{sprintf}
function accepts a @dfn{format specification} that tells it how to format
numbers (or strings), and that there are a number of different ways in which
numbers can be formatted. The different format specifications are discussed
more fully in
@ref{Control Letters, , Format-Control Letters}.
The built-in variable @code{OFMT} contains the default format specification
that @code{print} uses with @code{sprintf} when it wants to convert a
number to a string for printing.
The default value of @code{OFMT} is @code{"%.6g"}.
By supplying different format specifications
as the value of @code{OFMT}, you can change how @code{print} will print
your numbers. As a brief example:
@example
@group
$ awk 'BEGIN @{
> OFMT = "%.0f" # print numbers as integers (rounds)
> print 17.23 @}'
@print{} 17
@end group
@end example
@noindent
@cindex dark corner
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
According to the POSIX standard, @code{awk}'s behavior will be undefined
if @code{OFMT} contains anything but a floating point conversion specification
(d.c.).
@node Printf, Redirection, OFMT, Printing
@section Using @code{printf} Statements for Fancier Printing
@cindex formatted output
@cindex output, formatted
If you want more precise control over the output format than
@code{print} gives you, use @code{printf}. With @code{printf} you can
specify the width to use for each item, and you can specify various
formatting choices for numbers (such as what radix to use, whether to
print an exponent, whether to print a sign, and how many digits to print
after the decimal point). You do this by supplying a string, called
the @dfn{format string}, which controls how and where to print the other
arguments.
@menu
* Basic Printf:: Syntax of the @code{printf} statement.
* Control Letters:: Format-control letters.
* Format Modifiers:: Format-specification modifiers.
* Printf Examples:: Several examples.
@end menu
@node Basic Printf, Control Letters, Printf, Printf
@subsection Introduction to the @code{printf} Statement
@cindex @code{printf} statement, syntax of
The @code{printf} statement looks like this:
@example
printf @var{format}, @var{item1}, @var{item2}, @dots{}
@end example
@noindent
The entire list of arguments may optionally be enclosed in parentheses. The
parentheses are necessary if any of the item expressions use the @samp{>}
relational operator; otherwise it could be confused with a redirection
(@pxref{Redirection, ,Redirecting Output of @code{print} and @code{printf}}).
@cindex format string
The difference between @code{printf} and @code{print} is the @var{format}
argument. This is an expression whose value is taken as a string; it
specifies how to output each of the other arguments. It is called
the @dfn{format string}.
The format string is very similar to that in the ANSI C library function
@code{printf}. Most of @var{format} is text to be output verbatim.
Scattered among this text are @dfn{format specifiers}, one per item.
Each format specifier says to output the next item in the argument list
at that place in the format.
The @code{printf} statement does not automatically append a newline to its
output. It outputs only what the format string specifies. So if you want
a newline, you must include one in the format string. The output separator
variables @code{OFS} and @code{ORS} have no effect on @code{printf}
statements. For example:
@example
@group
BEGIN @{
ORS = "\nOUCH!\n"; OFS = "!"
msg = "Don't Panic!"; printf "%s\n", msg
@}
@end group
@end example
This program still prints the familiar @samp{Don't Panic!} message.
@node Control Letters, Format Modifiers, Basic Printf, Printf
@subsection Format-Control Letters
@cindex @code{printf}, format-control characters
@cindex format specifier
A format specifier starts with the character @samp{%} and ends with a
@dfn{format-control letter}; it tells the @code{printf} statement how
to output one item. (If you actually want to output a @samp{%}, write
@samp{%%}.) The format-control letter specifies what kind of value to
print. The rest of the format specifier is made up of optional
@dfn{modifiers} which are parameters to use, such as the field width.
Here is a list of the format-control letters:
@table @code
@item c
This prints a number as an ASCII character. Thus, @samp{printf "%c",
65} outputs the letter @samp{A}. The output for a string value is
the first character of the string.
@item d
@itemx i
These are equivalent. They both print a decimal integer.
The @samp{%i} specification is for compatibility with ANSI C.
@item e
@itemx E
This prints a number in scientific (exponential) notation.
For example,
@example
printf "%4.3e\n", 1950
@end example
@noindent
prints @samp{1.950e+03}, with a total of four significant figures of
which three follow the decimal point. The @samp{4.3} are modifiers,
discussed below. @samp{%E} uses @samp{E} instead of @samp{e} in the output.
@item f
This prints a number in floating point notation.
For example,
@example
printf "%4.3f", 1950
@end example
@noindent
prints @samp{1950.000}, with a total of four significant figures of
which three follow the decimal point. The @samp{4.3} are modifiers,
discussed below.
@item g
@itemx G
This prints a number in either scientific notation or floating point
notation, whichever uses fewer characters. If the result is printed in
scientific notation, @samp{%G} uses @samp{E} instead of @samp{e}.
@item o
This prints an unsigned octal integer.
(In octal, or base-eight notation, the digits run from @samp{0} to @samp{7};
the decimal number eight is represented as @samp{10} in octal.)
@item s
This prints a string.
@item x
@itemx X
This prints an unsigned hexadecimal integer.
(In hexadecimal, or base-16 notation, the digits are @samp{0} through @samp{9}
and @samp{a} through @samp{f}. The hexadecimal digit @samp{f} represents
the decimal number 15.) @samp{%X} uses the letters @samp{A} through @samp{F}
instead of @samp{a} through @samp{f}.
@item %
This isn't really a format-control letter, but it does have a meaning
when used after a @samp{%}: the sequence @samp{%%} outputs one
@samp{%}. It does not consume an argument, and it ignores any
modifiers.
@end table
@cindex dark corner
When using the integer format-control letters for values that are outside
the range of a C @code{long} integer, @code{gawk} will switch to the
@samp{%g} format specifier. Other versions of @code{awk} may print
invalid values, or do something else entirely (d.c.).
@node Format Modifiers, Printf Examples, Control Letters, Printf
@subsection Modifiers for @code{printf} Formats
@cindex @code{printf}, modifiers
@cindex modifiers (in format specifiers)
A format specification can also include @dfn{modifiers} that can control
how much of the item's value is printed and how much space it gets. The
modifiers come between the @samp{%} and the format-control letter.
In the examples below, we use the bullet symbol ``@bullet{}'' to represent
spaces in the output. Here are the possible modifiers, in the order in
which they may appear:
@table @code
@item -
The minus sign, used before the width modifier (see below),
says to left-justify
the argument within its specified width. Normally the argument
is printed right-justified in the specified width. Thus,
@example
printf "%-4s", "foo"
@end example
@noindent
prints @samp{foo@bullet{}}.
@item @var{space}
For numeric conversions, prefix positive values with a space, and
negative values with a minus sign.
@item +
The plus sign, used before the width modifier (see below),
says to always supply a sign for numeric conversions, even if the data
to be formatted is positive. The @samp{+} overrides the space modifier.
@item #
Use an ``alternate form'' for certain control letters.
For @samp{%o}, supply a leading zero.
For @samp{%x}, and @samp{%X}, supply a leading @samp{0x} or @samp{0X} for
a non-zero result.
For @samp{%e}, @samp{%E}, and @samp{%f}, the result will always contain a
decimal point.
For @samp{%g}, and @samp{%G}, trailing zeros are not removed from the result.
@cindex dark corner
@item 0
A leading @samp{0} (zero) acts as a flag, that indicates output should be
padded with zeros instead of spaces.
This applies even to non-numeric output formats (d.c.).
This flag only has an effect when the field width is wider than the
value to be printed.
@item @var{width}
This is a number specifying the desired minimum width of a field. Inserting any
number between the @samp{%} sign and the format control character forces the
field to be expanded to this width. The default way to do this is to
pad with spaces on the left. For example,
@example
printf "%4s", "foo"
@end example
@noindent
prints @samp{@bullet{}foo}.
The value of @var{width} is a minimum width, not a maximum. If the item
value requires more than @var{width} characters, it can be as wide as
necessary. Thus,
@example
printf "%4s", "foobar"
@end example
@noindent
prints @samp{foobar}.
Preceding the @var{width} with a minus sign causes the output to be
padded with spaces on the right, instead of on the left.
@item .@var{prec}
This is a number that specifies the precision to use when printing.
For the @samp{e}, @samp{E}, and @samp{f} formats, this specifies the
number of digits you want printed to the right of the decimal point.
For the @samp{g}, and @samp{G} formats, it specifies the maximum number
of significant digits. For the @samp{d}, @samp{o}, @samp{i}, @samp{u},
@samp{x}, and @samp{X} formats, it specifies the minimum number of
digits to print. For a string, it specifies the maximum number of
characters from the string that should be printed. Thus,
@example
printf "%.4s", "foobar"
@end example
@noindent
prints @samp{foob}.
@end table
The C library @code{printf}'s dynamic @var{width} and @var{prec}
capability (for example, @code{"%*.*s"}) is supported. Instead of
supplying explicit @var{width} and/or @var{prec} values in the format
string, you pass them in the argument list. For example:
@example
w = 5
p = 3
s = "abcdefg"
printf "%*.*s\n", w, p, s
@end example
@noindent
is exactly equivalent to
@example
s = "abcdefg"
printf "%5.3s\n", s
@end example
@noindent
Both programs output @samp{@w{@bullet{}@bullet{}abc}}.
Earlier versions of @code{awk} did not support this capability.
If you must use such a version, you may simulate this feature by using
concatenation to build up the format string, like so:
@example
w = 5
p = 3
s = "abcdefg"
printf "%" w "." p "s\n", s
@end example
@noindent
This is not particularly easy to read, but it does work.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
C programmers may be used to supplying additional @samp{l} and @samp{h}
flags in @code{printf} format strings. These are not valid in @code{awk}.
Most @code{awk} implementations silently ignore these flags.
If @samp{--lint} is provided on the command line
(@pxref{Options, ,Command Line Options}),
@code{gawk} will warn about their use. If @samp{--posix} is supplied,
their use is a fatal error.
@node Printf Examples, , Format Modifiers, Printf
@subsection Examples Using @code{printf}
Here is how to use @code{printf} to make an aligned table:
@example
awk '@{ printf "%-10s %s\n", $1, $2 @}' BBS-list
@end example
@noindent
prints the names of bulletin boards (@code{$1}) of the file
@file{BBS-list} as a string of 10 characters, left justified. It also
prints the phone numbers (@code{$2}) afterward on the line. This
produces an aligned two-column table of names and phone numbers:
@example
@group
$ awk '@{ printf "%-10s %s\n", $1, $2 @}' BBS-list
@print{} aardvark 555-5553
@print{} alpo-net 555-3412
@print{} barfly 555-7685
@print{} bites 555-1675
@print{} camelot 555-0542
@print{} core 555-2912
@print{} fooey 555-1234
@print{} foot 555-6699
@print{} macfoo 555-6480
@print{} sdace 555-3430
@print{} sabafoo 555-2127
@end group
@end example
Did you notice that we did not specify that the phone numbers be printed
as numbers? They had to be printed as strings because the numbers are
separated by a dash.
If we had tried to print the phone numbers as numbers, all we would have
gotten would have been the first three digits, @samp{555}.
This would have been pretty confusing.
We did not specify a width for the phone numbers because they are the
last things on their lines. We don't need to put spaces after them.
We could make our table look even nicer by adding headings to the tops
of the columns. To do this, we use the @code{BEGIN} pattern
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns})
to force the header to be printed only once, at the beginning of
the @code{awk} program:
@example
@group
awk 'BEGIN @{ print "Name Number"
print "---- ------" @}
@{ printf "%-10s %s\n", $1, $2 @}' BBS-list
@end group
@end example
Did you notice that we mixed @code{print} and @code{printf} statements in
the above example? We could have used just @code{printf} statements to get
the same results:
@example
@group
awk 'BEGIN @{ printf "%-10s %s\n", "Name", "Number"
printf "%-10s %s\n", "----", "------" @}
@{ printf "%-10s %s\n", $1, $2 @}' BBS-list
@end group
@end example
@noindent
By printing each column heading with the same format specification
used for the elements of the column, we have made sure that the headings
are aligned just like the columns.
The fact that the same format specification is used three times can be
emphasized by storing it in a variable, like this:
@example
@group
awk 'BEGIN @{ format = "%-10s %s\n"
printf format, "Name", "Number"
printf format, "----", "------" @}
@{ printf format, $1, $2 @}' BBS-list
@end group
@end example
@c !!! exercise
See if you can use the @code{printf} statement to line up the headings and
table data for our @file{inventory-shipped} example covered earlier in the
section on the @code{print} statement
(@pxref{Print, ,The @code{print} Statement}).
@node Redirection, Special Files, Printf, Printing
@section Redirecting Output of @code{print} and @code{printf}
@cindex output redirection
@cindex redirection of output
So far we have been dealing only with output that prints to the standard
output, usually your terminal. Both @code{print} and @code{printf} can
also send their output to other places.
This is called @dfn{redirection}.
A redirection appears after the @code{print} or @code{printf} statement.
Redirections in @code{awk} are written just like redirections in shell
commands, except that they are written inside the @code{awk} program.
There are three forms of output redirection: output to a file,
output appended to a file, and output through a pipe to another
command.
They are all shown for
the @code{print} statement, but they work identically for @code{printf}
also.
@table @code
@item print @var{items} > @var{output-file}
This type of redirection prints the items into the output file
@var{output-file}. The file name @var{output-file} can be any
expression. Its value is changed to a string and then used as a
file name (@pxref{Expressions}).
When this type of redirection is used, the @var{output-file} is erased
before the first output is written to it. Subsequent writes
to the same @var{output-file} do not
erase @var{output-file}, but append to it. If @var{output-file} does
not exist, then it is created.
For example, here is how an @code{awk} program can write a list of
BBS names to a file @file{name-list} and a list of phone numbers to a
file @file{phone-list}. Each output file contains one name or number
per line.
@example
@group
$ awk '@{ print $2 > "phone-list"
> print $1 > "name-list" @}' BBS-list
@end group
@group
$ cat phone-list
@print{} 555-5553
@print{} 555-3412
@dots{}
@end group
@group
$ cat name-list
@print{} aardvark
@print{} alpo-net
@dots{}
@end group
@end example
@item print @var{items} >> @var{output-file}
This type of redirection prints the items into the pre-existing output file
@var{output-file}. The difference between this and the
single-@samp{>} redirection is that the old contents (if any) of
@var{output-file} are not erased. Instead, the @code{awk} output is
appended to the file.
If @var{output-file} does not exist, then it is created.
@cindex pipes for output
@cindex output, piping
@item print @var{items} | @var{command}
It is also possible to send output to another program through a pipe
instead of into a
file. This type of redirection opens a pipe to @var{command} and writes
the values of @var{items} through this pipe, to another process created
to execute @var{command}.
The redirection argument @var{command} is actually an @code{awk}
expression. Its value is converted to a string, whose contents give the
shell command to be run.
For example, this produces two files, one unsorted list of BBS names
and one list sorted in reverse alphabetical order:
@example
awk '@{ print $1 > "names.unsorted"
command = "sort -r > names.sorted"
print $1 | command @}' BBS-list
@end example
Here the unsorted list is written with an ordinary redirection while
the sorted list is written by piping through the @code{sort} utility.
This example uses redirection to mail a message to a mailing
list @samp{bug-system}. This might be useful when trouble is encountered
in an @code{awk} script run periodically for system maintenance.
@example
report = "mail bug-system"
print "Awk script failed:", $0 | report
m = ("at record number " FNR " of " FILENAME)
print m | report
close(report)
@end example
The message is built using string concatenation and saved in the variable
@code{m}. It is then sent down the pipeline to the @code{mail} program.
We call the @code{close} function here because it's a good idea to close
the pipe as soon as all the intended output has been sent to it.
@xref{Close Files And Pipes, ,Closing Input and Output Files and Pipes},
for more information
on this. This example also illustrates the use of a variable to represent
a @var{file} or @var{command}: it is not necessary to always
use a string constant. Using a variable is generally a good idea,
since @code{awk} requires you to spell the string value identically
every time.
@end table
Redirecting output using @samp{>}, @samp{>>}, or @samp{|} asks the system
to open a file or pipe only if the particular @var{file} or @var{command}
you've specified has not already been written to by your program, or if
it has been closed since it was last written to.
@cindex differences between @code{gawk} and @code{awk}
@cindex limitations
@cindex implementation limits
@iftex
As mentioned earlier
(@pxref{Getline Summary, , Summary of @code{getline} Variants}),
many
@end iftex
@ifinfo
Many
@end ifinfo
@code{awk} implementations limit the number of pipelines an @code{awk}
program may have open to just one! In @code{gawk}, there is no such limit.
You can open as many pipelines as the underlying operating system will
permit.
@node Special Files, Close Files And Pipes , Redirection, Printing
@section Special File Names in @code{gawk}
@cindex standard input
@cindex standard output
@cindex standard error output
@cindex file descriptors
Running programs conventionally have three input and output streams
already available to them for reading and writing. These are known as
the @dfn{standard input}, @dfn{standard output}, and @dfn{standard error
output}. These streams are, by default, connected to your terminal, but
they are often redirected with the shell, via the @samp{<}, @samp{<<},
@samp{>}, @samp{>>}, @samp{>&} and @samp{|} operators. Standard error
is typically used for writing error messages; the reason we have two separate
streams, standard output and standard error, is so that they can be
redirected separately.
@cindex differences between @code{gawk} and @code{awk}
In other implementations of @code{awk}, the only way to write an error
message to standard error in an @code{awk} program is as follows:
@example
print "Serious error detected!" | "cat 1>&2"
@end example
@noindent
This works by opening a pipeline to a shell command which can access the
standard error stream which it inherits from the @code{awk} process.
This is far from elegant, and is also inefficient, since it requires a
separate process. So people writing @code{awk} programs often
neglect to do this. Instead, they send the error messages to the
terminal, like this:
@example
@group
print "Serious error detected!" > "/dev/tty"
@end group
@end example
@noindent
This usually has the same effect, but not always: although the
standard error stream is usually the terminal, it can be redirected, and
when that happens, writing to the terminal is not correct. In fact, if
@code{awk} is run from a background job, it may not have a terminal at all.
Then opening @file{/dev/tty} will fail.
@code{gawk} provides special file names for accessing the three standard
streams. When you redirect input or output in @code{gawk}, if the file name
matches one of these special names, then @code{gawk} directly uses the
stream it stands for.
@cindex @file{/dev/stdin}
@cindex @file{/dev/stdout}
@cindex @file{/dev/stderr}
@cindex @file{/dev/fd}
@c @cartouche
@table @file
@item /dev/stdin
The standard input (file descriptor 0).
@item /dev/stdout
The standard output (file descriptor 1).
@item /dev/stderr
The standard error output (file descriptor 2).
@item /dev/fd/@var{N}
The file associated with file descriptor @var{N}. Such a file must have
been opened by the program initiating the @code{awk} execution (typically
the shell). Unless you take special pains in the shell from which
you invoke @code{gawk}, only descriptors 0, 1 and 2 are available.
@end table
@c @end cartouche
The file names @file{/dev/stdin}, @file{/dev/stdout}, and @file{/dev/stderr}
are aliases for @file{/dev/fd/0}, @file{/dev/fd/1}, and @file{/dev/fd/2},
respectively, but they are more self-explanatory.
The proper way to write an error message in a @code{gawk} program
is to use @file{/dev/stderr}, like this:
@example
print "Serious error detected!" > "/dev/stderr"
@end example
@code{gawk} also provides special file names that give access to information
about the running @code{gawk} process. Each of these ``files'' provides
a single record of information. To read them more than once, you must
first close them with the @code{close} function
(@pxref{Close Files And Pipes, ,Closing Input and Output Files and Pipes}).
The filenames are:
@cindex process information
@cindex @file{/dev/pid}
@cindex @file{/dev/pgrpid}
@cindex @file{/dev/ppid}
@cindex @file{/dev/user}
@c @cartouche
@table @file
@item /dev/pid
Reading this file returns the process ID of the current process,
in decimal, terminated with a newline.
@item /dev/ppid
Reading this file returns the parent process ID of the current process,
in decimal, terminated with a newline.
@item /dev/pgrpid
Reading this file returns the process group ID of the current process,
in decimal, terminated with a newline.
@item /dev/user
Reading this file returns a single record terminated with a newline.
The fields are separated with spaces. The fields represent the
following information:
@table @code
@item $1
The return value of the @code{getuid} system call
(the real user ID number).
@item $2
The return value of the @code{geteuid} system call
(the effective user ID number).
@item $3
The return value of the @code{getgid} system call
(the real group ID number).
@item $4
The return value of the @code{getegid} system call
(the effective group ID number).
@end table
If there are any additional fields, they are the group IDs returned by
@code{getgroups} system call.
(Multiple groups may not be supported on all systems.)
@end table
@c @end cartouche
These special file names may be used on the command line as data
files, as well as for I/O redirections within an @code{awk} program.
They may not be used as source files with the @samp{-f} option.
Recognition of these special file names is disabled if @code{gawk} is in
compatibility mode (@pxref{Options, ,Command Line Options}).
@strong{Caution}: Unless your system actually has a @file{/dev/fd} directory
(or any of the other above listed special files),
the interpretation of these file names is done by @code{gawk} itself.
For example, using @samp{/dev/fd/4} for output will actually write on
file descriptor 4, and not on a new file descriptor that was @code{dup}'ed
from file descriptor 4. Most of the time this does not matter; however, it
is important to @emph{not} close any of the files related to file descriptors
0, 1, and 2. If you do close one of these files, unpredictable behavior
will result.
The special files that provide process-related information may disappear
in a future version of @code{gawk}.
@xref{Future Extensions, ,Probable Future Extensions}.
@node Close Files And Pipes, , Special Files, Printing
@section Closing Input and Output Files and Pipes
@cindex closing input files and pipes
@cindex closing output files and pipes
@findex close
If the same file name or the same shell command is used with
@code{getline}
(@pxref{Getline, ,Explicit Input with @code{getline}})
more than once during the execution of an @code{awk}
program, the file is opened (or the command is executed) only the first time.
At that time, the first record of input is read from that file or command.
The next time the same file or command is used in @code{getline}, another
record is read from it, and so on.
Similarly, when a file or pipe is opened for output, the file name or command
associated with
it is remembered by @code{awk} and subsequent writes to the same file or
command are appended to the previous writes. The file or pipe stays
open until @code{awk} exits.
This implies that if you want to start reading the same file again from
the beginning, or if you want to rerun a shell command (rather than
reading more output from the command), you must take special steps.
What you must do is use the @code{close} function, as follows:
@example
close(@var{filename})
@end example
@noindent
or
@example
close(@var{command})
@end example
The argument @var{filename} or @var{command} can be any expression. Its
value must @emph{exactly} match the string that was used to open the file or
start the command (spaces and other ``irrelevant'' characters
included). For example, if you open a pipe with this:
@example
"sort -r names" | getline foo
@end example
@noindent
then you must close it with this:
@example
close("sort -r names")
@end example
Once this function call is executed, the next @code{getline} from that
file or command, or the next @code{print} or @code{printf} to that
file or command, will reopen the file or rerun the command.
Because the expression that you use to close a file or pipeline must
exactly match the expression used to open the file or run the command,
it is good practice to use a variable to store the file name or command.
The previous example would become
@example
sortcom = "sort -r names"
sortcom | getline foo
@dots{}
close(sortcom)
@end example
@noindent
This helps avoid hard-to-find typographical errors in your @code{awk}
programs.
Here are some reasons why you might need to close an output file:
@itemize @bullet
@item
To write a file and read it back later on in the same @code{awk}
program. Close the file when you are finished writing it; then
you can start reading it with @code{getline}.
@item
To write numerous files, successively, in the same @code{awk}
program. If you don't close the files, eventually you may exceed a
system limit on the number of open files in one process. So close
each one when you are finished writing it.
@item
To make a command finish. When you redirect output through a pipe,
the command reading the pipe normally continues to try to read input
as long as the pipe is open. Often this means the command cannot
really do its work until the pipe is closed. For example, if you
redirect output to the @code{mail} program, the message is not
actually sent until the pipe is closed.
@item
To run the same program a second time, with the same arguments.
This is not the same thing as giving more input to the first run!
For example, suppose you pipe output to the @code{mail} program. If you
output several lines redirected to this pipe without closing it, they make
a single message of several lines. By contrast, if you close the pipe
after each line of output, then each line makes a separate message.
@end itemize
@vindex ERRNO
@cindex differences between @code{gawk} and @code{awk}
@code{close} returns a value of zero if the close succeeded.
Otherwise, the value will be non-zero.
In this case, @code{gawk} sets the variable @code{ERRNO} to a string
describing the error that occurred.
@cindex differences between @code{gawk} and @code{awk}
@cindex portability issues
If you use more files than the system allows you to have open,
@code{gawk} will attempt to multiplex the available open files among
your data files. @code{gawk}'s ability to do this depends upon the
facilities of your operating system: it may not always work. It is
therefore both good practice and good portability advice to always
use @code{close} on your files when you are done with them.
@node Expressions, Patterns and Actions, Printing, Top
@chapter Expressions
@cindex expression
Expressions are the basic building blocks of @code{awk} patterns
and actions. An expression evaluates to a value, which you can print, test,
store in a variable or pass to a function. Additionally, an expression
can assign a new value to a variable or a field, with an assignment operator.
An expression can serve as a pattern or action statement on its own.
Most other kinds of
statements contain one or more expressions which specify data on which to
operate. As in other languages, expressions in @code{awk} include
variables, array references, constants, and function calls, as well as
combinations of these with various operators.
@menu
* Constants:: String, numeric, and regexp constants.
* Using Constant Regexps:: When and how to use a regexp constant.
* Variables:: Variables give names to values for later use.
* Conversion:: The conversion of strings to numbers and vice
versa.
* Arithmetic Ops:: Arithmetic operations (@samp{+}, @samp{-},
etc.)
* Concatenation:: Concatenating strings.
* Assignment Ops:: Changing the value of a variable or a field.
* Increment Ops:: Incrementing the numeric value of a variable.
* Truth Values:: What is ``true'' and what is ``false''.
* Typing and Comparison:: How variables acquire types, and how this
affects comparison of numbers and strings with
@samp{<}, etc.
* Boolean Ops:: Combining comparison expressions using boolean
operators @samp{||} (``or''), @samp{&&}
(``and'') and @samp{!} (``not'').
* Conditional Exp:: Conditional expressions select between two
subexpressions under control of a third
subexpression.
* Function Calls:: A function call is an expression.
* Precedence:: How various operators nest.
@end menu
@node Constants, Using Constant Regexps, Expressions, Expressions
@section Constant Expressions
@cindex constants, types of
@cindex string constants
The simplest type of expression is the @dfn{constant}, which always has
the same value. There are three types of constants: numeric constants,
string constants, and regular expression constants.
@menu
* Scalar Constants:: Numeric and string constants.
* Regexp Constants:: Regular Expression constants.
@end menu
@node Scalar Constants, Regexp Constants, Constants, Constants
@subsection Numeric and String Constants
@cindex numeric constant
@cindex numeric value
A @dfn{numeric constant} stands for a number. This number can be an
integer, a decimal fraction, or a number in scientific (exponential)
notation.@footnote{The internal representation uses double-precision
floating point numbers. If you don't know what that means, then don't
worry about it.} Here are some examples of numeric constants, which all
have the same value:
@example
105
1.05e+2
1050e-1
@end example
A string constant consists of a sequence of characters enclosed in
double-quote marks. For example:
@example
"parrot"
@end example
@noindent
@cindex differences between @code{gawk} and @code{awk}
represents the string whose contents are @samp{parrot}. Strings in
@code{gawk} can be of any length and they can contain any of the possible
eight-bit ASCII characters including ASCII NUL (character code zero).
Other @code{awk}
implementations may have difficulty with some character codes.
@node Regexp Constants, , Scalar Constants, Constants
@subsection Regular Expression Constants
@cindex @code{~} operator
@cindex @code{!~} operator
A regexp constant is a regular expression description enclosed in
slashes, such as @code{@w{/^beginning and end$/}}. Most regexps used in
@code{awk} programs are constant, but the @samp{~} and @samp{!~}
matching operators can also match computed or ``dynamic'' regexps
(which are just ordinary strings or variables that contain a regexp).
@node Using Constant Regexps, Variables, Constants, Expressions
@section Using Regular Expression Constants
When used on the right hand side of the @samp{~} or @samp{!~}
operators, a regexp constant merely stands for the regexp that is to be
matched.
@cindex dark corner
Regexp constants (such as @code{/foo/}) may be used like simple expressions.
When a
regexp constant appears by itself, it has the same meaning as if it appeared
in a pattern, i.e.@: @samp{($0 ~ /foo/)} (d.c.)
(@pxref{Expression Patterns, ,Expressions as Patterns}).
This means that the two code segments,
@example
if ($0 ~ /barfly/ || $0 ~ /camelot/)
print "found"
@end example
@noindent
and
@example
if (/barfly/ || /camelot/)
print "found"
@end example
@noindent
are exactly equivalent.
One rather bizarre consequence of this rule is that the following
boolean expression is valid, but does not do what the user probably
intended:
@example
# note that /foo/ is on the left of the ~
if (/foo/ ~ $1) print "found foo"
@end example
@noindent
This code is ``obviously'' testing @code{$1} for a match against the regexp
@code{/foo/}. But in fact, the expression @samp{/foo/ ~ $1} actually means
@samp{($0 ~ /foo/) ~ $1}. In other words, first match the input record
against the regexp @code{/foo/}. The result will be either zero or one,
depending upon the success or failure of the match. Then match that result
against the first field in the record.
Since it is unlikely that you would ever really wish to make this kind of
test, @code{gawk} will issue a warning when it sees this construct in
a program.
Another consequence of this rule is that the assignment statement
@example
matches = /foo/
@end example
@noindent
will assign either zero or one to the variable @code{matches}, depending
upon the contents of the current input record.
This feature of the language was never well documented until the
POSIX specification.
@cindex differences between @code{gawk} and @code{awk}
@cindex dark corner
Constant regular expressions are also used as the first argument for
the @code{gensub}, @code{sub} and @code{gsub} functions, and as the
second argument of the @code{match} function
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
Modern implementations of @code{awk}, including @code{gawk}, allow
the third argument of @code{split} to be a regexp constant, while some
older implementations do not (d.c.).
This can lead to confusion when attempting to use regexp constants
as arguments to user defined functions
(@pxref{User-defined, , User-defined Functions}).
For example:
@example
@group
function mysub(pat, repl, str, global)
@{
if (global)
gsub(pat, repl, str)
else
sub(pat, repl, str)
return str
@}
@end group
@group
@{
@dots{}
text = "hi! hi yourself!"
mysub(/hi/, "howdy", text, 1)
@dots{}
@}
@end group
@end example
In this example, the programmer wishes to pass a regexp constant to the
user-defined function @code{mysub}, which will in turn pass it on to
either @code{sub} or @code{gsub}. However, what really happens is that
the @code{pat} parameter will be either one or zero, depending upon whether
or not @code{$0} matches @code{/hi/}.
As it is unlikely that you would ever really wish to pass a truth value
in this way, @code{gawk} will issue a warning when it sees a regexp
constant used as a parameter to a user-defined function.
@node Variables, Conversion, Using Constant Regexps, Expressions
@section Variables
Variables are ways of storing values at one point in your program for
use later in another part of your program. You can manipulate them
entirely within your program text, and you can also assign values to
them on the @code{awk} command line.
@menu
* Using Variables:: Using variables in your programs.
* Assignment Options:: Setting variables on the command line and a
summary of command line syntax. This is an
advanced method of input.
@end menu
@node Using Variables, Assignment Options, Variables, Variables
@subsection Using Variables in a Program
@cindex variables, user-defined
@cindex user-defined variables
Variables let you give names to values and refer to them later. You have
already seen variables in many of the examples. The name of a variable
must be a sequence of letters, digits and underscores, but it may not begin
with a digit. Case is significant in variable names; @code{a} and @code{A}
are distinct variables.
A variable name is a valid expression by itself; it represents the
variable's current value. Variables are given new values with
@dfn{assignment operators}, @dfn{increment operators} and
@dfn{decrement operators}.
@xref{Assignment Ops, ,Assignment Expressions}.
A few variables have special built-in meanings, such as @code{FS}, the
field separator, and @code{NF}, the number of fields in the current
input record. @xref{Built-in Variables}, for a list of them. These
built-in variables can be used and assigned just like all other
variables, but their values are also used or changed automatically by
@code{awk}. All built-in variables names are entirely upper-case.
Variables in @code{awk} can be assigned either numeric or string
values. By default, variables are initialized to the empty string, which
is zero if converted to a number. There is no need to
``initialize'' each variable explicitly in @code{awk},
the way you would in C and in most other traditional languages.
@node Assignment Options, , Using Variables, Variables
@subsection Assigning Variables on the Command Line
You can set any @code{awk} variable by including a @dfn{variable assignment}
among the arguments on the command line when you invoke @code{awk}
(@pxref{Other Arguments, ,Other Command Line Arguments}). Such an assignment has
this form:
@example
@var{variable}=@var{text}
@end example
@noindent
With it, you can set a variable either at the beginning of the
@code{awk} run or in between input files.
If you precede the assignment with the @samp{-v} option, like this:
@example
-v @var{variable}=@var{text}
@end example
@noindent
then the variable is set at the very beginning, before even the
@code{BEGIN} rules are run. The @samp{-v} option and its assignment
must precede all the file name arguments, as well as the program text.
(@xref{Options, ,Command Line Options}, for more information about
the @samp{-v} option.)
Otherwise, the variable assignment is performed at a time determined by
its position among the input file arguments: after the processing of the
preceding input file argument. For example:
@example
awk '@{ print $n @}' n=4 inventory-shipped n=2 BBS-list
@end example
@noindent
prints the value of field number @code{n} for all input records. Before
the first file is read, the command line sets the variable @code{n}
equal to four. This causes the fourth field to be printed in lines from
the file @file{inventory-shipped}. After the first file has finished,
but before the second file is started, @code{n} is set to two, so that the
second field is printed in lines from @file{BBS-list}.
@example
@group
$ awk '@{ print $n @}' n=4 inventory-shipped n=2 BBS-list
@print{} 15
@print{} 24
@dots{}
@print{} 555-5553
@print{} 555-3412
@dots{}
@end group
@end example
Command line arguments are made available for explicit examination by
the @code{awk} program in an array named @code{ARGV}
(@pxref{ARGC and ARGV, ,Using @code{ARGC} and @code{ARGV}}).
@cindex dark corner
@code{awk} processes the values of command line assignments for escape
sequences (d.c.) (@pxref{Escape Sequences}).
@node Conversion, Arithmetic Ops, Variables, Expressions
@section Conversion of Strings and Numbers
@cindex conversion of strings and numbers
Strings are converted to numbers, and numbers to strings, if the context
of the @code{awk} program demands it. For example, if the value of
either @code{foo} or @code{bar} in the expression @samp{foo + bar}
happens to be a string, it is converted to a number before the addition
is performed. If numeric values appear in string concatenation, they
are converted to strings. Consider this:
@example
two = 2; three = 3
print (two three) + 4
@end example
@noindent
This prints the (numeric) value 27. The numeric values of
the variables @code{two} and @code{three} are converted to strings and
concatenated together, and the resulting string is converted back to the
number 23, to which four is then added.
@cindex null string
@cindex empty string
@cindex type conversion
If, for some reason, you need to force a number to be converted to a
string, concatenate the empty string, @code{""}, with that number.
To force a string to be converted to a number, add zero to that string.
A string is converted to a number by interpreting any numeric prefix
of the string as numerals:
@code{"2.5"} converts to 2.5, @code{"1e3"} converts to 1000, and @code{"25fix"}
has a numeric value of 25.
Strings that can't be interpreted as valid numbers are converted to
zero.
@vindex CONVFMT
The exact manner in which numbers are converted into strings is controlled
by the @code{awk} built-in variable @code{CONVFMT} (@pxref{Built-in Variables}).
Numbers are converted using the @code{sprintf} function
(@pxref{String Functions, ,Built-in Functions for String Manipulation})
with @code{CONVFMT} as the format
specifier.
@code{CONVFMT}'s default value is @code{"%.6g"}, which prints a value with
at least six significant digits. For some applications you will want to
change it to specify more precision. Double precision on most modern
machines gives you 16 or 17 decimal digits of precision.
Strange results can happen if you set @code{CONVFMT} to a string that doesn't
tell @code{sprintf} how to format floating point numbers in a useful way.
For example, if you forget the @samp{%} in the format, all numbers will be
converted to the same constant string.
@cindex dark corner
As a special case, if a number is an integer, then the result of converting
it to a string is @emph{always} an integer, no matter what the value of
@code{CONVFMT} may be. Given the following code fragment:
@example
CONVFMT = "%2.2f"
a = 12
b = a ""
@end example
@noindent
@code{b} has the value @code{"12"}, not @code{"12.00"} (d.c.).
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@vindex OFMT
Prior to the POSIX standard, @code{awk} specified that the value
of @code{OFMT} was used for converting numbers to strings. @code{OFMT}
specifies the output format to use when printing numbers with @code{print}.
@code{CONVFMT} was introduced in order to separate the semantics of
conversion from the semantics of printing. Both @code{CONVFMT} and
@code{OFMT} have the same default value: @code{"%.6g"}. In the vast majority
of cases, old @code{awk} programs will not change their behavior.
However, this use of @code{OFMT} is something to keep in mind if you must
port your program to other implementations of @code{awk}; we recommend
that instead of changing your programs, you just port @code{gawk} itself!
@xref{Print, ,The @code{print} Statement},
for more information on the @code{print} statement.
@node Arithmetic Ops, Concatenation, Conversion, Expressions
@section Arithmetic Operators
@cindex arithmetic operators
@cindex operators, arithmetic
@cindex addition
@cindex subtraction
@cindex multiplication
@cindex division
@cindex remainder
@cindex quotient
@cindex exponentiation
The @code{awk} language uses the common arithmetic operators when
evaluating expressions. All of these arithmetic operators follow normal
precedence rules, and work as you would expect them to.
Here is a file @file{grades} containing a list of student names and
three test scores per student (it's a small class):
@example
Pat 100 97 58
Sandy 84 72 93
Chris 72 92 89
@end example
@noindent
This programs takes the file @file{grades}, and prints the average
of the scores.
@example
$ awk '@{ sum = $2 + $3 + $4 ; avg = sum / 3
> print $1, avg @}' grades
@print{} Pat 85
@print{} Sandy 83
@print{} Chris 84.3333
@end example
This table lists the arithmetic operators in @code{awk}, in order from
highest precedence to lowest:
@c @cartouche
@table @code
@item - @var{x}
Negation.
@item + @var{x}
Unary plus. The expression is converted to a number.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@item @var{x} ^ @var{y}
@itemx @var{x} ** @var{y}
Exponentiation: @var{x} raised to the @var{y} power. @samp{2 ^ 3} has
the value eight. The character sequence @samp{**} is equivalent to
@samp{^}. (The POSIX standard only specifies the use of @samp{^}
for exponentiation.)
@item @var{x} * @var{y}
Multiplication.
@item @var{x} / @var{y}
Division. Since all numbers in @code{awk} are
real numbers, the result is not rounded to an integer: @samp{3 / 4}
has the value 0.75.
@item @var{x} % @var{y}
@cindex differences between @code{gawk} and @code{awk}
Remainder. The quotient is rounded toward zero to an integer,
multiplied by @var{y} and this result is subtracted from @var{x}.
This operation is sometimes known as ``trunc-mod.'' The following
relation always holds:
@example
b * int(a / b) + (a % b) == a
@end example
One possibly undesirable effect of this definition of remainder is that
@code{@var{x} % @var{y}} is negative if @var{x} is negative. Thus,
@example
-17 % 8 = -1
@end example
In other @code{awk} implementations, the signedness of the remainder
may be machine dependent.
@c !!! what does posix say?
@item @var{x} + @var{y}
Addition.
@item @var{x} - @var{y}
Subtraction.
@end table
@c @end cartouche
For maximum portability, do not use the @samp{**} operator.
Unary plus and minus have the same precedence,
the multiplication operators all have the same precedence, and
addition and subtraction have the same precedence.
@node Concatenation, Assignment Ops, Arithmetic Ops, Expressions
@section String Concatenation
@cindex Kernighan, Brian
@display
@i{It seemed like a good idea at the time.}
Brian Kernighan
@end display
@sp 1
@cindex string operators
@cindex operators, string
@cindex concatenation
There is only one string operation: concatenation. It does not have a
specific operator to represent it. Instead, concatenation is performed by
writing expressions next to one another, with no operator. For example:
@example
@group
$ awk '@{ print "Field number one: " $1 @}' BBS-list
@print{} Field number one: aardvark
@print{} Field number one: alpo-net
@dots{}
@end group
@end example
Without the space in the string constant after the @samp{:}, the line
would run together. For example:
@example
@group
$ awk '@{ print "Field number one:" $1 @}' BBS-list
@print{} Field number one:aardvark
@print{} Field number one:alpo-net
@dots{}
@end group
@end example
Since string concatenation does not have an explicit operator, it is
often necessary to insure that it happens where you want it to by
using parentheses to enclose
the items to be concatenated. For example, the
following code fragment does not concatenate @code{file} and @code{name}
as you might expect:
@example
@group
file = "file"
name = "name"
print "something meaningful" > file name
@end group
@end example
@noindent
It is necessary to use the following:
@example
print "something meaningful" > (file name)
@end example
We recommend that you use parentheses around concatenation in all but the
most common contexts (such as on the right-hand side of @samp{=}).
@node Assignment Ops, Increment Ops, Concatenation, Expressions
@section Assignment Expressions
@cindex assignment operators
@cindex operators, assignment
@cindex expression, assignment
An @dfn{assignment} is an expression that stores a new value into a
variable. For example, let's assign the value one to the variable
@code{z}:
@example
z = 1
@end example
After this expression is executed, the variable @code{z} has the value one.
Whatever old value @code{z} had before the assignment is forgotten.
Assignments can store string values also. For example, this would store
the value @code{"this food is good"} in the variable @code{message}:
@example
thing = "food"
predicate = "good"
message = "this " thing " is " predicate
@end example
@noindent
(This also illustrates string concatenation.)
The @samp{=} sign is called an @dfn{assignment operator}. It is the
simplest assignment operator because the value of the right-hand
operand is stored unchanged.
@cindex side effect
Most operators (addition, concatenation, and so on) have no effect
except to compute a value. If you ignore the value, you might as well
not use the operator. An assignment operator is different; it does
produce a value, but even if you ignore the value, the assignment still
makes itself felt through the alteration of the variable. We call this
a @dfn{side effect}.
@cindex lvalue
@cindex rvalue
The left-hand operand of an assignment need not be a variable
(@pxref{Variables}); it can also be a field
(@pxref{Changing Fields, ,Changing the Contents of a Field}) or
an array element (@pxref{Arrays, ,Arrays in @code{awk}}).
These are all called @dfn{lvalues},
which means they can appear on the left-hand side of an assignment operator.
The right-hand operand may be any expression; it produces the new value
which the assignment stores in the specified variable, field or array
element. (Such values are called @dfn{rvalues}).
@cindex types of variables
It is important to note that variables do @emph{not} have permanent types.
The type of a variable is simply the type of whatever value it happens
to hold at the moment. In the following program fragment, the variable
@code{foo} has a numeric value at first, and a string value later on:
@example
@group
foo = 1
print foo
foo = "bar"
print foo
@end group
@end example
@noindent
When the second assignment gives @code{foo} a string value, the fact that
it previously had a numeric value is forgotten.
String values that do not begin with a digit have a numeric value of
zero. After executing this code, the value of @code{foo} is five:
@example
foo = "a string"
foo = foo + 5
@end example
@noindent
(Note that using a variable as a number and then later as a string can
be confusing and is poor programming style. The above examples illustrate how
@code{awk} works, @emph{not} how you should write your own programs!)
An assignment is an expression, so it has a value: the same value that
is assigned. Thus, @samp{z = 1} as an expression has the value one.
One consequence of this is that you can write multiple assignments together:
@example
x = y = z = 0
@end example
@noindent
stores the value zero in all three variables. It does this because the
value of @samp{z = 0}, which is zero, is stored into @code{y}, and then
the value of @samp{y = z = 0}, which is zero, is stored into @code{x}.
You can use an assignment anywhere an expression is called for. For
example, it is valid to write @samp{x != (y = 1)} to set @code{y} to one
and then test whether @code{x} equals one. But this style tends to make
programs hard to read; except in a one-shot program, you should
not use such nesting of assignments.
Aside from @samp{=}, there are several other assignment operators that
do arithmetic with the old value of the variable. For example, the
operator @samp{+=} computes a new value by adding the right-hand value
to the old value of the variable. Thus, the following assignment adds
five to the value of @code{foo}:
@example
foo += 5
@end example
@noindent
This is equivalent to the following:
@example
foo = foo + 5
@end example
@noindent
Use whichever one makes the meaning of your program clearer.
There are situations where using @samp{+=} (or any assignment operator)
is @emph{not} the same as simply repeating the left-hand operand in the
right-hand expression. For example:
@cindex Rankin, Pat
@example
@group
# Thanks to Pat Rankin for this example
BEGIN @{
foo[rand()] += 5
for (x in foo)
print x, foo[x]
bar[rand()] = bar[rand()] + 5
for (x in bar)
print x, bar[x]
@}
@end group
@end example
@noindent
The indices of @code{bar} are guaranteed to be different, because
@code{rand} will return different values each time it is called.
(Arrays and the @code{rand} function haven't been covered yet.
@xref{Arrays, ,Arrays in @code{awk}},
and see @ref{Numeric Functions, ,Numeric Built-in Functions}, for more information).
This example illustrates an important fact about the assignment
operators: the left-hand expression is only evaluated @emph{once}.
It is also up to the implementation as to which expression is evaluated
first, the left-hand one or the right-hand one.
Consider this example:
@example
i = 1
a[i += 2] = i + 1
@end example
@noindent
The value of @code{a[3]} could be either two or four.
Here is a table of the arithmetic assignment operators. In each
case, the right-hand operand is an expression whose value is converted
to a number.
@c @cartouche
@table @code
@item @var{lvalue} += @var{increment}
Adds @var{increment} to the value of @var{lvalue} to make the new value
of @var{lvalue}.
@item @var{lvalue} -= @var{decrement}
Subtracts @var{decrement} from the value of @var{lvalue}.
@item @var{lvalue} *= @var{coefficient}
Multiplies the value of @var{lvalue} by @var{coefficient}.
@item @var{lvalue} /= @var{divisor}
Divides the value of @var{lvalue} by @var{divisor}.
@item @var{lvalue} %= @var{modulus}
Sets @var{lvalue} to its remainder by @var{modulus}.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@item @var{lvalue} ^= @var{power}
@itemx @var{lvalue} **= @var{power}
Raises @var{lvalue} to the power @var{power}.
(Only the @samp{^=} operator is specified by POSIX.)
@end table
@c @end cartouche
For maximum portability, do not use the @samp{**=} operator.
@node Increment Ops, Truth Values, Assignment Ops, Expressions
@section Increment and Decrement Operators
@cindex increment operators
@cindex operators, increment
@dfn{Increment} and @dfn{decrement operators} increase or decrease the value of
a variable by one. You could do the same thing with an assignment operator, so
the increment operators add no power to the @code{awk} language; but they
are convenient abbreviations for very common operations.
The operator to add one is written @samp{++}. It can be used to increment
a variable either before or after taking its value.
To pre-increment a variable @var{v}, write @samp{++@var{v}}. This adds
one to the value of @var{v} and that new value is also the value of this
expression. The assignment expression @samp{@var{v} += 1} is completely
equivalent.
Writing the @samp{++} after the variable specifies post-increment. This
increments the variable value just the same; the difference is that the
value of the increment expression itself is the variable's @emph{old}
value. Thus, if @code{foo} has the value four, then the expression @samp{foo++}
has the value four, but it changes the value of @code{foo} to five.
The post-increment @samp{foo++} is nearly equivalent to writing @samp{(foo
+= 1) - 1}. It is not perfectly equivalent because all numbers in
@code{awk} are floating point: in floating point, @samp{foo + 1 - 1} does
not necessarily equal @code{foo}. But the difference is minute as
long as you stick to numbers that are fairly small (less than 10e12).
Any lvalue can be incremented. Fields and array elements are incremented
just like variables. (Use @samp{$(i++)} when you wish to do a field reference
and a variable increment at the same time. The parentheses are necessary
because of the precedence of the field reference operator, @samp{$}.)
@cindex decrement operators
@cindex operators, decrement
The decrement operator @samp{--} works just like @samp{++} except that
it subtracts one instead of adding. Like @samp{++}, it can be used before
the lvalue to pre-decrement or after it to post-decrement.
Here is a summary of increment and decrement expressions.
@c @cartouche
@table @code
@item ++@var{lvalue}
This expression increments @var{lvalue} and the new value becomes the
value of the expression.
@item @var{lvalue}++
This expression increments @var{lvalue}, but
the value of the expression is the @emph{old} value of @var{lvalue}.
@item --@var{lvalue}
Like @samp{++@var{lvalue}}, but instead of adding, it subtracts. It
decrements @var{lvalue} and delivers the value that results.
@item @var{lvalue}--
Like @samp{@var{lvalue}++}, but instead of adding, it subtracts. It
decrements @var{lvalue}. The value of the expression is the @emph{old}
value of @var{lvalue}.
@end table
@c @end cartouche
@node Truth Values, Typing and Comparison, Increment Ops, Expressions
@section True and False in @code{awk}
@cindex truth values
@cindex logical true
@cindex logical false
Many programming languages have a special representation for the concepts
of ``true'' and ``false.'' Such languages usually use the special
constants @code{true} and @code{false}, or perhaps their upper-case
equivalents.
@cindex null string
@cindex empty string
@code{awk} is different. It borrows a very simple concept of true and
false from C. In @code{awk}, any non-zero numeric value, @emph{or} any
non-empty string value is true. Any other value (zero or the null
string, @code{""}) is false. The following program will print @samp{A strange
truth value} three times:
@example
@group
BEGIN @{
if (3.1415927)
print "A strange truth value"
if ("Four Score And Seven Years Ago")
print "A strange truth value"
if (j = 57)
print "A strange truth value"
@}
@end group
@end example
@cindex dark corner
There is a surprising consequence of the ``non-zero or non-null'' rule:
The string constant @code{"0"} is actually true, since it is non-null (d.c.).
@node Typing and Comparison, Boolean Ops, Truth Values, Expressions
@section Variable Typing and Comparison Expressions
@cindex comparison expressions
@cindex expression, comparison
@cindex expression, matching
@cindex relational operators
@cindex operators, relational
@cindex regexp match/non-match operators
@cindex variable typing
@cindex types of variables
@c 2e: consider splitting this section into subsections
@display
@i{The Guide is definitive. Reality is frequently inaccurate.}
The Hitchhiker's Guide to the Galaxy
@end display
@sp 1
Unlike other programming languages, @code{awk} variables do not have a
fixed type. Instead, they can be either a number or a string, depending
upon the value that is assigned to them.
@cindex numeric string
The 1992 POSIX standard introduced
the concept of a @dfn{numeric string}, which is simply a string that looks
like a number, for example, @code{@w{" +2"}}. This concept is used
for determining the type of a variable.
The type of the variable is important, since the types of two variables
determine how they are compared.
In @code{gawk}, variable typing follows these rules.
@enumerate 1
@item
A numeric literal or the result of a numeric operation has the @var{numeric}
attribute.
@item
A string literal or the result of a string operation has the @var{string}
attribute.
@item
Fields, @code{getline} input, @code{FILENAME}, @code{ARGV} elements,
@code{ENVIRON} elements and the
elements of an array created by @code{split} that are numeric strings
have the @var{strnum} attribute. Otherwise, they have the @var{string}
attribute.
Uninitialized variables also have the @var{strnum} attribute.
@item
Attributes propagate across assignments, but are not changed by
any use.
@c (Although a use may cause the entity to acquire an additional
@c value such that it has both a numeric and string value -- this leaves the
@c attribute unchanged.)
@c This is important but not relevant
@end enumerate
The last rule is particularly important. In the following program,
@code{a} has numeric type, even though it is later used in a string
operation.
@example
BEGIN @{
a = 12.345
b = a " is a cute number"
print b
@}
@end example
When two operands are compared, either string comparison or numeric comparison
may be used, depending on the attributes of the operands, according to the
following, symmetric, matrix:
@c thanks to Karl Berry, kb@cs.umb.edu, for major help with TeX tables
@tex
\centerline{
\vbox{\bigskip % space above the table (about 1 linespace)
% Because we have vertical rules, we can't let TeX insert interline space
% in its usual way.
\offinterlineskip
%
% Define the table template. & separates columns, and \cr ends the
% template (and each row). # is replaced by the text of that entry on
% each row. The template for the first column breaks down like this:
% \strut -- a way to make each line have the height and depth
% of a normal line of type, since we turned off interline spacing.
% \hfil -- infinite glue; has the effect of right-justifying in this case.
% # -- replaced by the text (for instance, `STRNUM', in the last row).
% \quad -- about the width of an `M'. Just separates the columns.
%
% The second column (\vrule#) is what generates the vertical rule that
% spans table rows.
%
% The doubled && before the next entry means `repeat the following
% template as many times as necessary on each line' -- in our case, twice.
%
% The template itself, \quad#\hfil, left-justifies with a little space before.
%
\halign{\strut\hfil#\quad&\vrule#&&\quad#\hfil\cr
&&STRING &NUMERIC &STRNUM\cr
% The \omit tells TeX to skip inserting the template for this column on
% this particular row. In this case, we only want a little extra space
% to separate the heading row from the rule below it. the depth 2pt --
% `\vrule depth 2pt' is that little space.
\omit &depth 2pt\cr
% This is the horizontal rule below the heading. Since it has nothing to
% do with the columns of the table, we use \noalign to get it in there.
\noalign{\hrule}
% Like above, this time a little more space.
\omit &depth 4pt\cr
% The remaining rows have nothing special about them.
STRING &&string &string &string\cr
NUMERIC &&string &numeric &numeric\cr
STRNUM &&string &numeric &numeric\cr
}}}
@end tex
@ifinfo
@display
+----------------------------------------------
| STRING NUMERIC STRNUM
--------+----------------------------------------------
|
STRING | string string string
|
NUMERIC | string numeric numeric
|
STRNUM | string numeric numeric
--------+----------------------------------------------
@end display
@end ifinfo
The basic idea is that user input that looks numeric, and @emph{only}
user input, should be treated as numeric, even though it is actually
made of characters, and is therefore also a string.
@dfn{Comparison expressions} compare strings or numbers for
relationships such as equality. They are written using @dfn{relational
operators}, which are a superset of those in C. Here is a table of
them:
@cindex relational operators
@cindex operators, relational
@cindex @code{<} operator
@cindex @code{<=} operator
@cindex @code{>} operator
@cindex @code{>=} operator
@cindex @code{==} operator
@cindex @code{!=} operator
@cindex @code{~} operator
@cindex @code{!~} operator
@cindex @code{in} operator
@c @cartouche
@table @code
@item @var{x} < @var{y}
True if @var{x} is less than @var{y}.
@item @var{x} <= @var{y}
True if @var{x} is less than or equal to @var{y}.
@item @var{x} > @var{y}
True if @var{x} is greater than @var{y}.
@item @var{x} >= @var{y}
True if @var{x} is greater than or equal to @var{y}.
@item @var{x} == @var{y}
True if @var{x} is equal to @var{y}.
@item @var{x} != @var{y}
True if @var{x} is not equal to @var{y}.
@item @var{x} ~ @var{y}
True if the string @var{x} matches the regexp denoted by @var{y}.
@item @var{x} !~ @var{y}
True if the string @var{x} does not match the regexp denoted by @var{y}.
@item @var{subscript} in @var{array}
True if the array @var{array} has an element with the subscript @var{subscript}.
@end table
@c @end cartouche
Comparison expressions have the value one if true and zero if false.
When comparing operands of mixed types, numeric operands are converted
to strings using the value of @code{CONVFMT}
(@pxref{Conversion, ,Conversion of Strings and Numbers}).
Strings are compared
by comparing the first character of each, then the second character of each,
and so on. Thus @code{"10"} is less than @code{"9"}. If there are two
strings where one is a prefix of the other, the shorter string is less than
the longer one. Thus @code{"abc"} is less than @code{"abcd"}.
@cindex common mistakes
@cindex mistakes, common
@cindex errors, common
It is very easy to accidentally mistype the @samp{==} operator, and
leave off one of the @samp{=}s. The result is still valid @code{awk}
code, but the program will not do what you mean:
@example
if (a = b) # oops! should be a == b
@dots{}
else
@dots{}
@end example
@noindent
Unless @code{b} happens to be zero or the null string, the @code{if}
part of the test will always succeed. Because the operators are
so similar, this kind of error is very difficult to spot when
scanning the source code.
Here are some sample expressions, how @code{gawk} compares them, and what
the result of the comparison is.
@table @code
@item 1.5 <= 2.0
numeric comparison (true)
@item "abc" >= "xyz"
string comparison (false)
@item 1.5 != " +2"
string comparison (true)
@item "1e2" < "3"
string comparison (true)
@item a = 2; b = "2"
@itemx a == b
string comparison (true)
@item a = 2; b = " +2"
@itemx a == b
string comparison (false)
@end table
In this example,
@example
@group
$ echo 1e2 3 | awk '@{ print ($1 < $2) ? "true" : "false" @}'
@print{} false
@end group
@end example
@noindent
the result is @samp{false} since both @code{$1} and @code{$2} are numeric
strings and thus both have the @var{strnum} attribute,
dictating a numeric comparison.
The purpose of the comparison rules and the use of numeric strings is
to attempt to produce the behavior that is ``least surprising,'' while
still ``doing the right thing.''
@cindex comparisons, string vs. regexp
@cindex string comparison vs. regexp comparison
@cindex regexp comparison vs. string comparison
String comparisons and regular expression comparisons are very different.
For example,
@example
x == "foo"
@end example
@noindent
has the value of one, or is true, if the variable @code{x}
is precisely @samp{foo}. By contrast,
@example
x ~ /foo/
@end example
@noindent
has the value one if @code{x} contains @samp{foo}, such as
@code{"Oh, what a fool am I!"}.
The right hand operand of the @samp{~} and @samp{!~} operators may be
either a regexp constant (@code{/@dots{}/}), or an ordinary
expression, in which case the value of the expression as a string is used as a
dynamic regexp (@pxref{Regexp Usage, ,How to Use Regular Expressions}; also
@pxref{Computed Regexps, ,Using Dynamic Regexps}).
@cindex regexp as expression
In recent implementations of @code{awk}, a constant regular
expression in slashes by itself is also an expression. The regexp
@code{/@var{regexp}/} is an abbreviation for this comparison expression:
@example
$0 ~ /@var{regexp}/
@end example
One special place where @code{/foo/} is @emph{not} an abbreviation for
@samp{$0 ~ /foo/} is when it is the right-hand operand of @samp{~} or
@samp{!~}!
@xref{Using Constant Regexps, ,Using Regular Expression Constants},
where this is discussed in more detail.
@c This paragraph has been here since day 1, and has always bothered
@c me, especially since the expression doesn't really make a lot of
@c sense. So, just take it out.
@ignore
In some contexts it may be necessary to write parentheses around the
regexp to avoid confusing the @code{gawk} parser. For example,
@samp{(/x/ - /y/) > threshold} is not allowed, but @samp{((/x/) - (/y/))
> threshold} parses properly.
@end ignore
@node Boolean Ops, Conditional Exp, Typing and Comparison, Expressions
@section Boolean Expressions
@cindex expression, boolean
@cindex boolean expressions
@cindex operators, boolean
@cindex boolean operators
@cindex logical operations
@cindex operations, logical
@cindex short-circuit operators
@cindex operators, short-circuit
@cindex and operator
@cindex or operator
@cindex not operator
@cindex @code{&&} operator
@cindex @code{||} operator
@cindex @code{!} operator
A @dfn{boolean expression} is a combination of comparison expressions or
matching expressions, using the boolean operators ``or''
(@samp{||}), ``and'' (@samp{&&}), and ``not'' (@samp{!}), along with
parentheses to control nesting. The truth value of the boolean expression is
computed by combining the truth values of the component expressions.
Boolean expressions are also referred to as @dfn{logical expressions}.
The terms are equivalent.
Boolean expressions can be used wherever comparison and matching
expressions can be used. They can be used in @code{if}, @code{while},
@code{do} and @code{for} statements
(@pxref{Statements, ,Control Statements in Actions}).
They have numeric values (one if true, zero if false), which come into play
if the result of the boolean expression is stored in a variable, or
used in arithmetic.
In addition, every boolean expression is also a valid pattern, so
you can use one as a pattern to control the execution of rules.
Here are descriptions of the three boolean operators, with examples.
@c @cartouche
@table @code
@item @var{boolean1} && @var{boolean2}
True if both @var{boolean1} and @var{boolean2} are true. For example,
the following statement prints the current input record if it contains
both @samp{2400} and @samp{foo}.
@example
if ($0 ~ /2400/ && $0 ~ /foo/) print
@end example
The subexpression @var{boolean2} is evaluated only if @var{boolean1}
is true. This can make a difference when @var{boolean2} contains
expressions that have side effects: in the case of @samp{$0 ~ /foo/ &&
($2 == bar++)}, the variable @code{bar} is not incremented if there is
no @samp{foo} in the record.
@item @var{boolean1} || @var{boolean2}
True if at least one of @var{boolean1} or @var{boolean2} is true.
For example, the following statement prints all records in the input
that contain @emph{either} @samp{2400} or
@samp{foo}, or both.
@example
if ($0 ~ /2400/ || $0 ~ /foo/) print
@end example
The subexpression @var{boolean2} is evaluated only if @var{boolean1}
is false. This can make a difference when @var{boolean2} contains
expressions that have side effects.
@item ! @var{boolean}
True if @var{boolean} is false. For example, the following program prints
all records in the input file @file{BBS-list} that do @emph{not} contain the
string @samp{foo}.
@c A better example would be `if (! (subscript in array)) ...' but we
@c haven't done anything with arrays or `in' yet. Sigh.
@example
awk '@{ if (! ($0 ~ /foo/)) print @}' BBS-list
@end example
@end table
@c @end cartouche
The @samp{&&} and @samp{||} operators are called @dfn{short-circuit}
operators because of the way they work. Evaluation of the full expression
is ``short-circuited'' if the result can be determined part way through
its evaluation.
@cindex line continuation
You can continue a statement that uses @samp{&&} or @samp{||} simply
by putting a newline after them. But you cannot put a newline in front
of either of these operators without using backslash continuation
(@pxref{Statements/Lines, ,@code{awk} Statements Versus Lines}).
The actual value of an expression using the @samp{!} operator will be
either one or zero, depending upon the truth value of the expression it
is applied to.
The @samp{!} operator is often useful for changing the sense of a flag
variable from false to true and back again. For example, the following
program is one way to print lines in between special bracketing lines:
@example
$1 == "START" @{ interested = ! interested @}
interested == 1 @{ print @}
$1 == "END" @{ interested = ! interested @}
@end example
@noindent
The variable @code{interested}, like all @code{awk} variables, starts
out initialized to zero, which is also false. When a line is seen whose
first field is @samp{START}, the value of @code{interested} is toggled
to true, using @samp{!}. The next rule prints lines as long as
@code{interested} is true. When a line is seen whose first field is
@samp{END}, @code{interested} is toggled back to false.
@ignore
We should discuss using `next' in the two rules that toggle the
variable, to avoid printing the bracketing lines, but that's more
distraction than really needed.
@end ignore
@node Conditional Exp, Function Calls, Boolean Ops, Expressions
@section Conditional Expressions
@cindex conditional expression
@cindex expression, conditional
A @dfn{conditional expression} is a special kind of expression with
three operands. It allows you to use one expression's value to select
one of two other expressions.
The conditional expression is the same as in the C language:
@example
@var{selector} ? @var{if-true-exp} : @var{if-false-exp}
@end example
@noindent
There are three subexpressions. The first, @var{selector}, is always
computed first. If it is ``true'' (not zero and not null) then
@var{if-true-exp} is computed next and its value becomes the value of
the whole expression. Otherwise, @var{if-false-exp} is computed next
and its value becomes the value of the whole expression.
For example, this expression produces the absolute value of @code{x}:
@example
x > 0 ? x : -x
@end example
Each time the conditional expression is computed, exactly one of
@var{if-true-exp} and @var{if-false-exp} is computed; the other is ignored.
This is important when the expressions contain side effects. For example,
this conditional expression examines element @code{i} of either array
@code{a} or array @code{b}, and increments @code{i}.
@example
x == y ? a[i++] : b[i++]
@end example
@noindent
This is guaranteed to increment @code{i} exactly once, because each time
only one of the two increment expressions is executed,
and the other is not.
@xref{Arrays, ,Arrays in @code{awk}},
for more information about arrays.
@cindex differences between @code{gawk} and @code{awk}
@cindex line continuation
As a minor @code{gawk} extension,
you can continue a statement that uses @samp{?:} simply
by putting a newline after either character.
However, you cannot put a newline in front
of either character without using backslash continuation
(@pxref{Statements/Lines, ,@code{awk} Statements Versus Lines}).
@node Function Calls, Precedence, Conditional Exp, Expressions
@section Function Calls
@cindex function call
@cindex calling a function
A @dfn{function} is a name for a particular calculation. Because it has
a name, you can ask for it by name at any point in the program. For
example, the function @code{sqrt} computes the square root of a number.
A fixed set of functions are @dfn{built-in}, which means they are
available in every @code{awk} program. The @code{sqrt} function is one
of these. @xref{Built-in, ,Built-in Functions}, for a list of built-in
functions and their descriptions. In addition, you can define your own
functions for use in your program.
@xref{User-defined, ,User-defined Functions}, for how to do this.
@cindex arguments in function call
The way to use a function is with a @dfn{function call} expression,
which consists of the function name followed immediately by a list of
@dfn{arguments} in parentheses. The arguments are expressions which
provide the raw materials for the function's calculations.
When there is more than one argument, they are separated by commas. If
there are no arguments, write just @samp{()} after the function name.
Here are some examples:
@example
sqrt(x^2 + y^2) @i{one argument}
atan2(y, x) @i{two arguments}
rand() @i{no arguments}
@end example
@strong{Do not put any space between the function name and the
open-parenthesis!} A user-defined function name looks just like the name of
a variable, and space would make the expression look like concatenation
of a variable with an expression inside parentheses. Space before the
parenthesis is harmless with built-in functions, but it is best not to get
into the habit of using space to avoid mistakes with user-defined
functions.
Each function expects a particular number of arguments. For example, the
@code{sqrt} function must be called with a single argument, the number
to take the square root of:
@example
sqrt(@var{argument})
@end example
Some of the built-in functions allow you to omit the final argument.
If you do so, they use a reasonable default.
@xref{Built-in, ,Built-in Functions}, for full details. If arguments
are omitted in calls to user-defined functions, then those arguments are
treated as local variables, initialized to the empty string
(@pxref{User-defined, ,User-defined Functions}).
Like every other expression, the function call has a value, which is
computed by the function based on the arguments you give it. In this
example, the value of @samp{sqrt(@var{argument})} is the square root of
@var{argument}. A function can also have side effects, such as assigning
values to certain variables or doing I/O.
Here is a command to read numbers, one number per line, and print the
square root of each one:
@example
@group
$ awk '@{ print "The square root of", $1, "is", sqrt($1) @}'
1
@print{} The square root of 1 is 1
3
@print{} The square root of 3 is 1.73205
5
@print{} The square root of 5 is 2.23607
@kbd{Control-d}
@end group
@end example
@node Precedence, , Function Calls, Expressions
@section Operator Precedence (How Operators Nest)
@cindex precedence
@cindex operator precedence
@dfn{Operator precedence} determines how operators are grouped, when
different operators appear close by in one expression. For example,
@samp{*} has higher precedence than @samp{+}; thus, @samp{a + b * c}
means to multiply @code{b} and @code{c}, and then add @code{a} to the
product (i.e.@: @samp{a + (b * c)}).
You can overrule the precedence of the operators by using parentheses.
You can think of the precedence rules as saying where the
parentheses are assumed to be if you do not write parentheses yourself. In
fact, it is wise to always use parentheses whenever you have an unusual
combination of operators, because other people who read the program may
not remember what the precedence is in this case. You might forget,
too; then you could make a mistake. Explicit parentheses will help prevent
any such mistake.
When operators of equal precedence are used together, the leftmost
operator groups first, except for the assignment, conditional and
exponentiation operators, which group in the opposite order.
Thus, @samp{a - b + c} groups as @samp{(a - b) + c}, and
@samp{a = b = c} groups as @samp{a = (b = c)}.
The precedence of prefix unary operators does not matter as long as only
unary operators are involved, because there is only one way to interpret
them---innermost first. Thus, @samp{$++i} means @samp{$(++i)} and
@samp{++$x} means @samp{++($x)}. However, when another operator follows
the operand, then the precedence of the unary operators can matter.
Thus, @samp{$x^2} means @samp{($x)^2}, but @samp{-x^2} means
@samp{-(x^2)}, because @samp{-} has lower precedence than @samp{^}
while @samp{$} has higher precedence.
Here is a table of @code{awk}'s operators, in order from highest
precedence to lowest:
@c use @code in the items, looks better in TeX w/o all the quotes
@table @code
@item (@dots{})
Grouping.
@item $
Field.
@item ++ --
Increment, decrement.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@item ^ **
Exponentiation. These operators group right-to-left.
(The @samp{**} operator is not specified by POSIX.)
@item + - !
Unary plus, minus, logical ``not''.
@item * / %
Multiplication, division, modulus.
@item + -
Addition, subtraction.
@item @r{Concatenation}
No special token is used to indicate concatenation.
The operands are simply written side by side.
@item < <= == !=
@itemx > >= >> |
Relational, and redirection.
The relational operators and the redirections have the same precedence
level. Characters such as @samp{>} serve both as relationals and as
redirections; the context distinguishes between the two meanings.
Note that the I/O redirection operators in @code{print} and @code{printf}
statements belong to the statement level, not to expressions. The
redirection does not produce an expression which could be the operand of
another operator. As a result, it does not make sense to use a
redirection operator near another operator of lower precedence, without
parentheses. Such combinations, for example @samp{print foo > a ? b : c},
result in syntax errors.
The correct way to write this statement is @samp{print foo > (a ? b : c)}.
@item ~ !~
Matching, non-matching.
@item in
Array membership.
@item &&
Logical ``and''.
@item ||
Logical ``or''.
@item ?:
Conditional. This operator groups right-to-left.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@item = += -= *=
@itemx /= %= ^= **=
Assignment. These operators group right-to-left.
(The @samp{**=} operator is not specified by POSIX.)
@end table
@node Patterns and Actions, Statements, Expressions, Top
@chapter Patterns and Actions
@cindex pattern, definition of
As you have already seen, each @code{awk} statement consists of
a pattern with an associated action. This chapter describes how
you build patterns and actions.
@menu
* Pattern Overview:: What goes into a pattern.
* Action Overview:: What goes into an action.
@end menu
@node Pattern Overview, Action Overview, Patterns and Actions, Patterns and Actions
@section Pattern Elements
Patterns in @code{awk} control the execution of rules: a rule is
executed when its pattern matches the current input record. This
section explains all about how to write patterns.
@menu
* Kinds of Patterns:: A list of all kinds of patterns.
* Regexp Patterns:: Using regexps as patterns.
* Expression Patterns:: Any expression can be used as a pattern.
* Ranges:: Pairs of patterns specify record ranges.
* BEGIN/END:: Specifying initialization and cleanup rules.
* Empty:: The empty pattern, which matches every record.
@end menu
@node Kinds of Patterns, Regexp Patterns, Pattern Overview, Pattern Overview
@subsection Kinds of Patterns
@cindex patterns, types of
Here is a summary of the types of patterns supported in @code{awk}.
@table @code
@item /@var{regular expression}/
A regular expression as a pattern. It matches when the text of the
input record fits the regular expression.
(@xref{Regexp, ,Regular Expressions}.)
@item @var{expression}
A single expression. It matches when its value
is non-zero (if a number) or non-null (if a string).
(@xref{Expression Patterns, ,Expressions as Patterns}.)
@item @var{pat1}, @var{pat2}
A pair of patterns separated by a comma, specifying a range of records.
The range includes both the initial record that matches @var{pat1}, and
the final record that matches @var{pat2}.
(@xref{Ranges, ,Specifying Record Ranges with Patterns}.)
@item BEGIN
@itemx END
Special patterns for you to supply start-up or clean-up actions for your
@code{awk} program.
(@xref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}.)
@item @var{empty}
The empty pattern matches every input record.
(@xref{Empty, ,The Empty Pattern}.)
@end table
@node Regexp Patterns, Expression Patterns, Kinds of Patterns, Pattern Overview
@subsection Regular Expressions as Patterns
We have been using regular expressions as patterns since our early examples.
This kind of pattern is simply a regexp constant in the pattern part of
a rule. Its meaning is @samp{$0 ~ /@var{pattern}/}.
The pattern matches when the input record matches the regexp.
For example:
@example
/foo|bar|baz/ @{ buzzwords++ @}
END @{ print buzzwords, "buzzwords seen" @}
@end example
@node Expression Patterns, Ranges, Regexp Patterns, Pattern Overview
@subsection Expressions as Patterns
Any @code{awk} expression is valid as an @code{awk} pattern.
Then the pattern matches if the expression's value is non-zero (if a
number) or non-null (if a string).
The expression is reevaluated each time the rule is tested against a new
input record. If the expression uses fields such as @code{$1}, the
value depends directly on the new input record's text; otherwise, it
depends only on what has happened so far in the execution of the
@code{awk} program, but that may still be useful.
A very common kind of expression used as a pattern is the comparison
expression, using the comparison operators described in
@ref{Typing and Comparison, ,Variable Typing and Comparison Expressions}.
Regexp matching and non-matching are also very common expressions.
The left operand of the @samp{~} and @samp{!~} operators is a string.
The right operand is either a constant regular expression enclosed in
slashes (@code{/@var{regexp}/}), or any expression, whose string value
is used as a dynamic regular expression
(@pxref{Computed Regexps, , Using Dynamic Regexps}).
The following example prints the second field of each input record
whose first field is precisely @samp{foo}.
@example
$ awk '$1 == "foo" @{ print $2 @}' BBS-list
@end example
@noindent
(There is no output, since there is no BBS site named ``foo''.)
Contrast this with the following regular expression match, which would
accept any record with a first field that contains @samp{foo}:
@example
@group
$ awk '$1 ~ /foo/ @{ print $2 @}' BBS-list
@print{} 555-1234
@print{} 555-6699
@print{} 555-6480
@print{} 555-2127
@end group
@end example
Boolean expressions are also commonly used as patterns.
Whether the pattern
matches an input record depends on whether its subexpressions match.
For example, the following command prints all records in
@file{BBS-list} that contain both @samp{2400} and @samp{foo}.
@example
$ awk '/2400/ && /foo/' BBS-list
@print{} fooey 555-1234 2400/1200/300 B
@end example
The following command prints all records in
@file{BBS-list} that contain @emph{either} @samp{2400} or @samp{foo}, or
both.
@example
@group
$ awk '/2400/ || /foo/' BBS-list
@print{} alpo-net 555-3412 2400/1200/300 A
@print{} bites 555-1675 2400/1200/300 A
@print{} fooey 555-1234 2400/1200/300 B
@print{} foot 555-6699 1200/300 B
@print{} macfoo 555-6480 1200/300 A
@print{} sdace 555-3430 2400/1200/300 A
@print{} sabafoo 555-2127 1200/300 C
@end group
@end example
The following command prints all records in
@file{BBS-list} that do @emph{not} contain the string @samp{foo}.
@example
@group
$ awk '! /foo/' BBS-list
@print{} aardvark 555-5553 1200/300 B
@print{} alpo-net 555-3412 2400/1200/300 A
@print{} barfly 555-7685 1200/300 A
@print{} bites 555-1675 2400/1200/300 A
@print{} camelot 555-0542 300 C
@print{} core 555-2912 1200/300 C
@print{} sdace 555-3430 2400/1200/300 A
@end group
@end example
The subexpressions of a boolean operator in a pattern can be constant regular
expressions, comparisons, or any other @code{awk} expressions. Range
patterns are not expressions, so they cannot appear inside boolean
patterns. Likewise, the special patterns @code{BEGIN} and @code{END},
which never match any input record, are not expressions and cannot
appear inside boolean patterns.
A regexp constant as a pattern is also a special case of an expression
pattern. @code{/foo/} as an expression has the value one if @samp{foo}
appears in the current input record; thus, as a pattern, @code{/foo/}
matches any record containing @samp{foo}.
@node Ranges, BEGIN/END, Expression Patterns, Pattern Overview
@subsection Specifying Record Ranges with Patterns
@cindex range pattern
@cindex pattern, range
@cindex matching ranges of lines
A @dfn{range pattern} is made of two patterns separated by a comma, of
the form @samp{@var{begpat}, @var{endpat}}. It matches ranges of
consecutive input records. The first pattern, @var{begpat}, controls
where the range begins, and the second one, @var{endpat}, controls where
it ends. For example,
@example
awk '$1 == "on", $1 == "off"'
@end example
@noindent
prints every record between @samp{on}/@samp{off} pairs, inclusive.
A range pattern starts out by matching @var{begpat}
against every input record; when a record matches @var{begpat}, the
range pattern becomes @dfn{turned on}. The range pattern matches this
record. As long as it stays turned on, it automatically matches every
input record read. It also matches @var{endpat} against
every input record; when that succeeds, the range pattern is turned
off again for the following record. Then it goes back to checking
@var{begpat} against each record.
The record that turns on the range pattern and the one that turns it
off both match the range pattern. If you don't want to operate on
these records, you can write @code{if} statements in the rule's action
to distinguish them from the records you are interested in.
It is possible for a pattern to be turned both on and off by the same
record, if the record satisfies both conditions. Then the action is
executed for just that record.
For example, suppose you have text between two identical markers (say
the @samp{%} symbol) that you wish to ignore. You might try to
combine a range pattern that describes the delimited text with the
@code{next} statement
(not discussed yet, @pxref{Next Statement, , The @code{next} Statement}),
which causes @code{awk} to skip any further processing of the current
record and start over again with the next input record. Such a program
would look like this:
@example
/^%$/,/^%$/ @{ next @}
@{ print @}
@end example
@noindent
@cindex skipping lines between markers
This program fails because the range pattern is both turned on and turned off
by the first line with just a @samp{%} on it. To accomplish this task, you
must write the program this way, using a flag:
@example
/^%$/ @{ skip = ! skip; next @}
skip == 1 @{ next @} # skip lines with `skip' set
@end example
Note that in a range pattern, the @samp{,} has the lowest precedence
(is evaluated last) of all the operators. Thus, for example, the
following program attempts to combine a range pattern with another,
simpler test.
@example
echo Yes | awk '/1/,/2/ || /Yes/'
@end example
The author of this program intended it to mean @samp{(/1/,/2/) || /Yes/}.
However, @code{awk} interprets this as @samp{/1/, (/2/ || /Yes/)}.
This cannot be changed or worked around; range patterns do not combine
with other patterns.
@node BEGIN/END, Empty, Ranges, Pattern Overview
@subsection The @code{BEGIN} and @code{END} Special Patterns
@cindex @code{BEGIN} special pattern
@cindex pattern, @code{BEGIN}
@cindex @code{END} special pattern
@cindex pattern, @code{END}
@code{BEGIN} and @code{END} are special patterns. They are not used to
match input records. Rather, they supply start-up or
clean-up actions for your @code{awk} script.
@menu
* Using BEGIN/END:: How and why to use BEGIN/END rules.
* I/O And BEGIN/END:: I/O issues in BEGIN/END rules.
@end menu
@node Using BEGIN/END, I/O And BEGIN/END, BEGIN/END, BEGIN/END
@subsubsection Startup and Cleanup Actions
A @code{BEGIN} rule is executed, once, before the first input record
has been read. An @code{END} rule is executed, once, after all the
input has been read. For example:
@example
@group
$ awk '
> BEGIN @{ print "Analysis of \"foo\"" @}
> /foo/ @{ ++n @}
> END @{ print "\"foo\" appears " n " times." @}' BBS-list
@print{} Analysis of "foo"
@print{} "foo" appears 4 times.
@end group
@end example
This program finds the number of records in the input file @file{BBS-list}
that contain the string @samp{foo}. The @code{BEGIN} rule prints a title
for the report. There is no need to use the @code{BEGIN} rule to
initialize the counter @code{n} to zero, as @code{awk} does this
automatically (@pxref{Variables}).
The second rule increments the variable @code{n} every time a
record containing the pattern @samp{foo} is read. The @code{END} rule
prints the value of @code{n} at the end of the run.
The special patterns @code{BEGIN} and @code{END} cannot be used in ranges
or with boolean operators (indeed, they cannot be used with any operators).
An @code{awk} program may have multiple @code{BEGIN} and/or @code{END}
rules. They are executed in the order they appear, all the @code{BEGIN}
rules at start-up and all the @code{END} rules at termination.
@code{BEGIN} and @code{END} rules may be intermixed with other rules.
This feature was added in the 1987 version of @code{awk}, and is included
in the POSIX standard. The original (1978) version of @code{awk}
required you to put the @code{BEGIN} rule at the beginning of the
program, and the @code{END} rule at the end, and only allowed one of
each. This is no longer required, but it is a good idea in terms of
program organization and readability.
Multiple @code{BEGIN} and @code{END} rules are useful for writing
library functions, since each library file can have its own @code{BEGIN} and/or
@code{END} rule to do its own initialization and/or cleanup. Note that
the order in which library functions are named on the command line
controls the order in which their @code{BEGIN} and @code{END} rules are
executed. Therefore you have to be careful to write such rules in
library files so that the order in which they are executed doesn't matter.
@xref{Options, ,Command Line Options}, for more information on
using library functions.
@xref{Library Functions, ,A Library of @code{awk} Functions},
for a number of useful library functions.
@cindex dark corner
If an @code{awk} program only has a @code{BEGIN} rule, and no other
rules, then the program exits after the @code{BEGIN} rule has been run.
(The original version of @code{awk} used to keep reading and ignoring input
until end of file was seen.) However, if an @code{END} rule exists,
then the input will be read, even if there are no other rules in
the program. This is necessary in case the @code{END} rule checks the
@code{FNR} and @code{NR} variables (d.c.).
@code{BEGIN} and @code{END} rules must have actions; there is no default
action for these rules since there is no current record when they run.
@node I/O And BEGIN/END, , Using BEGIN/END, BEGIN/END
@subsubsection Input/Output from @code{BEGIN} and @code{END} Rules
@cindex I/O from @code{BEGIN} and @code{END}
There are several (sometimes subtle) issues involved when doing I/O
from a @code{BEGIN} or @code{END} rule.
The first has to do with the value of @code{$0} in a @code{BEGIN}
rule. Since @code{BEGIN} rules are executed before any input is read,
there simply is no input record, and therefore no fields, when
executing @code{BEGIN} rules. References to @code{$0} and the fields
yield a null string or zero, depending upon the context. One way
to give @code{$0} a real value is to execute a @code{getline} command
without a variable (@pxref{Getline, ,Explicit Input with @code{getline}}).
Another way is to simply assign a value to it.
@cindex differences between @code{gawk} and @code{awk}
The second point is similar to the first, but from the other direction.
Inside an @code{END} rule, what is the value of @code{$0} and @code{NF}?
Traditionally, due largely to implementation issues, @code{$0} and
@code{NF} were @emph{undefined} inside an @code{END} rule.
The POSIX standard specified that @code{NF} was available in an @code{END}
rule, containing the number of fields from the last input record.
Due most probably to an oversight, the standard does not say that @code{$0}
is also preserved, although logically one would think that it should be.
In fact, @code{gawk} does preserve the value of @code{$0} for use in
@code{END} rules. Be aware, however, that Unix @code{awk}, and possibly
other implementations, do not.
The third point follows from the first two. What is the meaning of
@samp{print} inside a @code{BEGIN} or @code{END} rule? The meaning is
the same as always, @samp{print $0}. If @code{$0} is the null string,
then this prints an empty line. Many long time @code{awk} programmers
use @samp{print} in @code{BEGIN} and @code{END} rules, to mean
@samp{@w{print ""}}, relying on @code{$0} being null. While you might
generally get away with this in @code{BEGIN} rules, in @code{gawk} at
least, it is a very bad idea in @code{END} rules. It is also poor
style, since if you want an empty line in the output, you
should say so explicitly in your program.
@node Empty, , BEGIN/END, Pattern Overview
@subsection The Empty Pattern
@cindex empty pattern
@cindex pattern, empty
An empty (i.e.@: non-existent) pattern is considered to match @emph{every}
input record. For example, the program:
@example
awk '@{ print $1 @}' BBS-list
@end example
@noindent
prints the first field of every record.
@node Action Overview, , Pattern Overview, Patterns and Actions
@section Overview of Actions
@cindex action, definition of
@cindex curly braces
@cindex action, curly braces
@cindex action, separating statements
An @code{awk} program or script consists of a series of
rules and function definitions, interspersed. (Functions are
described later. @xref{User-defined, ,User-defined Functions}.)
A rule contains a pattern and an action, either of which (but not
both) may be
omitted. The purpose of the @dfn{action} is to tell @code{awk} what to do
once a match for the pattern is found. Thus, in outline, an @code{awk}
program generally looks like this:
@example
@r{[}@var{pattern}@r{]} @r{[}@{ @var{action} @}@r{]}
@r{[}@var{pattern}@r{]} @r{[}@{ @var{action} @}@r{]}
@dots{}
function @var{name}(@var{args}) @{ @dots{} @}
@dots{}
@end example
An action consists of one or more @code{awk} @dfn{statements}, enclosed
in curly braces (@samp{@{} and @samp{@}}). Each statement specifies one
thing to be done. The statements are separated by newlines or
semicolons.
The curly braces around an action must be used even if the action
contains only one statement, or even if it contains no statements at
all. However, if you omit the action entirely, omit the curly braces as
well. An omitted action is equivalent to @samp{@{ print $0 @}}.
@example
/foo/ @{ @} # match foo, do nothing - empty action
/foo/ # match foo, print the record - omitted action
@end example
Here are the kinds of statements supported in @code{awk}:
@itemize @bullet
@item
Expressions, which can call functions or assign values to variables
(@pxref{Expressions}). Executing
this kind of statement simply computes the value of the expression.
This is useful when the expression has side effects
(@pxref{Assignment Ops, ,Assignment Expressions}).
@item
Control statements, which specify the control flow of @code{awk}
programs. The @code{awk} language gives you C-like constructs
(@code{if}, @code{for}, @code{while}, and @code{do}) as well as a few
special ones (@pxref{Statements, ,Control Statements in Actions}).
@item
Compound statements, which consist of one or more statements enclosed in
curly braces. A compound statement is used in order to put several
statements together in the body of an @code{if}, @code{while}, @code{do}
or @code{for} statement.
@item
Input statements, using the @code{getline} command
(@pxref{Getline, ,Explicit Input with @code{getline}}), the @code{next}
statement (@pxref{Next Statement, ,The @code{next} Statement}),
and the @code{nextfile} statement
(@pxref{Nextfile Statement, ,The @code{nextfile} Statement}).
@item
Output statements, @code{print} and @code{printf}.
@xref{Printing, ,Printing Output}.
@item
Deletion statements, for deleting array elements.
@xref{Delete, ,The @code{delete} Statement}.
@end itemize
@iftex
The next chapter covers control statements in detail.
@end iftex
@node Statements, Built-in Variables, Patterns and Actions, Top
@chapter Control Statements in Actions
@cindex control statement
@dfn{Control statements} such as @code{if}, @code{while}, and so on
control the flow of execution in @code{awk} programs. Most of the
control statements in @code{awk} are patterned on similar statements in
C.
All the control statements start with special keywords such as @code{if}
and @code{while}, to distinguish them from simple expressions.
@cindex compound statement
@cindex statement, compound
Many control statements contain other statements; for example, the
@code{if} statement contains another statement which may or may not be
executed. The contained statement is called the @dfn{body}. If you
want to include more than one statement in the body, group them into a
single @dfn{compound statement} with curly braces, separating them with
newlines or semicolons.
@menu
* If Statement:: Conditionally execute some @code{awk}
statements.
* While Statement:: Loop until some condition is satisfied.
* Do Statement:: Do specified action while looping until some
condition is satisfied.
* For Statement:: Another looping statement, that provides
initialization and increment clauses.
* Break Statement:: Immediately exit the innermost enclosing loop.
* Continue Statement:: Skip to the end of the innermost enclosing
loop.
* Next Statement:: Stop processing the current input record.
* Nextfile Statement:: Stop processing the current file.
* Exit Statement:: Stop execution of @code{awk}.
@end menu
@node If Statement, While Statement, Statements, Statements
@section The @code{if}-@code{else} Statement
@cindex @code{if}-@code{else} statement
The @code{if}-@code{else} statement is @code{awk}'s decision-making
statement. It looks like this:
@example
if (@var{condition}) @var{then-body} @r{[}else @var{else-body}@r{]}
@end example
@noindent
The @var{condition} is an expression that controls what the rest of the
statement will do. If @var{condition} is true, @var{then-body} is
executed; otherwise, @var{else-body} is executed.
The @code{else} part of the statement is
optional. The condition is considered false if its value is zero or
the null string, and true otherwise.
Here is an example:
@example
if (x % 2 == 0)
print "x is even"
else
print "x is odd"
@end example
In this example, if the expression @samp{x % 2 == 0} is true (that is,
the value of @code{x} is evenly divisible by two), then the first @code{print}
statement is executed, otherwise the second @code{print} statement is
executed.
If the @code{else} appears on the same line as @var{then-body}, and
@var{then-body} is not a compound statement (i.e.@: not surrounded by
curly braces), then a semicolon must separate @var{then-body} from
@code{else}. To illustrate this, let's rewrite the previous example:
@example
if (x % 2 == 0) print "x is even"; else
print "x is odd"
@end example
@noindent
If you forget the @samp{;}, @code{awk} won't be able to interpret the
statement, and you will get a syntax error.
We would not actually write this example this way, because a human
reader might fail to see the @code{else} if it were not the first thing
on its line.
@node While Statement, Do Statement, If Statement, Statements
@section The @code{while} Statement
@cindex @code{while} statement
@cindex loop
@cindex body of a loop
In programming, a @dfn{loop} means a part of a program that can
be executed two or more times in succession.
The @code{while} statement is the simplest looping statement in
@code{awk}. It repeatedly executes a statement as long as a condition is
true. It looks like this:
@example
while (@var{condition})
@var{body}
@end example
@noindent
Here @var{body} is a statement that we call the @dfn{body} of the loop,
and @var{condition} is an expression that controls how long the loop
keeps running.
The first thing the @code{while} statement does is test @var{condition}.
If @var{condition} is true, it executes the statement @var{body}.
@ifinfo
(The @var{condition} is true when the value
is not zero and not a null string.)
@end ifinfo
After @var{body} has been executed,
@var{condition} is tested again, and if it is still true, @var{body} is
executed again. This process repeats until @var{condition} is no longer
true. If @var{condition} is initially false, the body of the loop is
never executed, and @code{awk} continues with the statement following
the loop.
This example prints the first three fields of each record, one per line.
@example
awk '@{ i = 1
while (i <= 3) @{
print $i
i++
@}
@}' inventory-shipped
@end example
@noindent
Here the body of the loop is a compound statement enclosed in braces,
containing two statements.
The loop works like this: first, the value of @code{i} is set to one.
Then, the @code{while} tests whether @code{i} is less than or equal to
three. This is true when @code{i} equals one, so the @code{i}-th
field is printed. Then the @samp{i++} increments the value of @code{i}
and the loop repeats. The loop terminates when @code{i} reaches four.
As you can see, a newline is not required between the condition and the
body; but using one makes the program clearer unless the body is a
compound statement or is very simple. The newline after the open-brace
that begins the compound statement is not required either, but the
program would be harder to read without it.
@node Do Statement, For Statement, While Statement, Statements
@section The @code{do}-@code{while} Statement
The @code{do} loop is a variation of the @code{while} looping statement.
The @code{do} loop executes the @var{body} once, and then repeats @var{body}
as long as @var{condition} is true. It looks like this:
@example
@group
do
@var{body}
while (@var{condition})
@end group
@end example
Even if @var{condition} is false at the start, @var{body} is executed at
least once (and only once, unless executing @var{body} makes
@var{condition} true). Contrast this with the corresponding
@code{while} statement:
@example
while (@var{condition})
@var{body}
@end example
@noindent
This statement does not execute @var{body} even once if @var{condition}
is false to begin with.
Here is an example of a @code{do} statement:
@example
awk '@{ i = 1
do @{
print $0
i++
@} while (i <= 10)
@}'
@end example
@noindent
This program prints each input record ten times. It isn't a very
realistic example, since in this case an ordinary @code{while} would do
just as well. But this reflects actual experience; there is only
occasionally a real use for a @code{do} statement.
@node For Statement, Break Statement, Do Statement, Statements
@section The @code{for} Statement
@cindex @code{for} statement
The @code{for} statement makes it more convenient to count iterations of a
loop. The general form of the @code{for} statement looks like this:
@example
for (@var{initialization}; @var{condition}; @var{increment})
@var{body}
@end example
@noindent
The @var{initialization}, @var{condition} and @var{increment} parts are
arbitrary @code{awk} expressions, and @var{body} stands for any
@code{awk} statement.
The @code{for} statement starts by executing @var{initialization}.
Then, as long
as @var{condition} is true, it repeatedly executes @var{body} and then
@var{increment}. Typically @var{initialization} sets a variable to
either zero or one, @var{increment} adds one to it, and @var{condition}
compares it against the desired number of iterations.
Here is an example of a @code{for} statement:
@example
@group
awk '@{ for (i = 1; i <= 3; i++)
print $i
@}' inventory-shipped
@end group
@end example
@noindent
This prints the first three fields of each input record, one field per
line.
You cannot set more than one variable in the
@var{initialization} part unless you use a multiple assignment statement
such as @samp{x = y = 0}, which is possible only if all the initial values
are equal. (But you can initialize additional variables by writing
their assignments as separate statements preceding the @code{for} loop.)
The same is true of the @var{increment} part; to increment additional
variables, you must write separate statements at the end of the loop.
The C compound expression, using C's comma operator, would be useful in
this context, but it is not supported in @code{awk}.
Most often, @var{increment} is an increment expression, as in the
example above. But this is not required; it can be any expression
whatever. For example, this statement prints all the powers of two
between one and 100:
@example
for (i = 1; i <= 100; i *= 2)
print i
@end example
Any of the three expressions in the parentheses following the @code{for} may
be omitted if there is nothing to be done there. Thus, @w{@samp{for (; x
> 0;)}} is equivalent to @w{@samp{while (x > 0)}}. If the
@var{condition} is omitted, it is treated as @var{true}, effectively
yielding an @dfn{infinite loop} (i.e.@: a loop that will never
terminate).
In most cases, a @code{for} loop is an abbreviation for a @code{while}
loop, as shown here:
@example
@var{initialization}
while (@var{condition}) @{
@var{body}
@var{increment}
@}
@end example
@noindent
The only exception is when the @code{continue} statement
(@pxref{Continue Statement, ,The @code{continue} Statement}) is used
inside the loop; changing a @code{for} statement to a @code{while}
statement in this way can change the effect of the @code{continue}
statement inside the loop.
There is an alternate version of the @code{for} loop, for iterating over
all the indices of an array:
@example
for (i in array)
@var{do something with} array[i]
@end example
@noindent
@xref{Scanning an Array, ,Scanning All Elements of an Array},
for more information on this version of the @code{for} loop.
The @code{awk} language has a @code{for} statement in addition to a
@code{while} statement because often a @code{for} loop is both less work to
type and more natural to think of. Counting the number of iterations is
very common in loops. It can be easier to think of this counting as part
of looping rather than as something to do inside the loop.
The next section has more complicated examples of @code{for} loops.
@node Break Statement, Continue Statement, For Statement, Statements
@section The @code{break} Statement
@cindex @code{break} statement
@cindex loops, exiting
The @code{break} statement jumps out of the innermost @code{for},
@code{while}, or @code{do} loop that encloses it. The
following example finds the smallest divisor of any integer, and also
identifies prime numbers:
@example
awk '# find smallest divisor of num
@{ num = $1
for (div = 2; div*div <= num; div++)
if (num % div == 0)
break
if (num % div == 0)
printf "Smallest divisor of %d is %d\n", num, div
else
printf "%d is prime\n", num
@}'
@end example
When the remainder is zero in the first @code{if} statement, @code{awk}
immediately @dfn{breaks out} of the containing @code{for} loop. This means
that @code{awk} proceeds immediately to the statement following the loop
and continues processing. (This is very different from the @code{exit}
statement which stops the entire @code{awk} program.
@xref{Exit Statement, ,The @code{exit} Statement}.)
Here is another program equivalent to the previous one. It illustrates how
the @var{condition} of a @code{for} or @code{while} could just as well be
replaced with a @code{break} inside an @code{if}:
@example
@group
awk '# find smallest divisor of num
@{ num = $1
for (div = 2; ; div++) @{
if (num % div == 0) @{
printf "Smallest divisor of %d is %d\n", num, div
break
@}
if (div*div > num) @{
printf "%d is prime\n", num
break
@}
@}
@}'
@end group
@end example
@cindex @code{break}, outside of loops
@cindex historical features
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@cindex dark corner
As described above, the @code{break} statement has no meaning when
used outside the body of a loop. However, although it was never documented,
historical implementations of @code{awk} have treated the @code{break}
statement outside of a loop as if it were a @code{next} statement
(@pxref{Next Statement, ,The @code{next} Statement}).
Recent versions of Unix @code{awk} no longer allow this usage.
@code{gawk} will support this use of @code{break} only if @samp{--traditional}
has been specified on the command line
(@pxref{Options, ,Command Line Options}).
Otherwise, it will be treated as an error, since the POSIX standard
specifies that @code{break} should only be used inside the body of a
loop (d.c.).
@node Continue Statement, Next Statement, Break Statement, Statements
@section The @code{continue} Statement
@cindex @code{continue} statement
The @code{continue} statement, like @code{break}, is used only inside
@code{for}, @code{while}, and @code{do} loops. It skips
over the rest of the loop body, causing the next cycle around the loop
to begin immediately. Contrast this with @code{break}, which jumps out
of the loop altogether.
@c The point of this program was to illustrate the use of continue with
@c a while loop. But Karl Berry points out that that is done adequately
@c below, and that this example is very un-awk-like. So for now, we'll
@c omit it.
@ignore
In Texinfo source files, text that the author wishes to ignore can be
enclosed between lines that start with @samp{@@ignore} and end with
@samp{@@end ignore}. Here is a program that strips out lines between
@samp{@@ignore} and @samp{@@end ignore} pairs.
@example
BEGIN @{
while (getline > 0) @{
if (/^@@ignore/)
ignoring = 1
else if (/^@@end[ \t]+ignore/) @{
ignoring = 0
continue
@}
if (ignoring)
continue
print
@}
@}
@end example
When an @samp{@@ignore} is seen, the @code{ignoring} flag is set to one (true).
When @samp{@@end ignore} is seen, the flag is reset to zero (false). As long
as the flag is true, the input record is not printed, because the
@code{continue} restarts the @code{while} loop, skipping over the @code{print}
statement.
@c Exercise!!!
@c How could this program be written to make better use of the awk language?
@end ignore
The @code{continue} statement in a @code{for} loop directs @code{awk} to
skip the rest of the body of the loop, and resume execution with the
increment-expression of the @code{for} statement. The following program
illustrates this fact:
@example
awk 'BEGIN @{
for (x = 0; x <= 20; x++) @{
if (x == 5)
continue
printf "%d ", x
@}
print ""
@}'
@end example
@noindent
This program prints all the numbers from zero to 20, except for five, for
which the @code{printf} is skipped. Since the increment @samp{x++}
is not skipped, @code{x} does not remain stuck at five. Contrast the
@code{for} loop above with this @code{while} loop:
@example
awk 'BEGIN @{
x = 0
while (x <= 20) @{
if (x == 5)
continue
printf "%d ", x
x++
@}
print ""
@}'
@end example
@noindent
This program loops forever once @code{x} gets to five.
@cindex @code{continue}, outside of loops
@cindex historical features
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@cindex dark corner
As described above, the @code{continue} statement has no meaning when
used outside the body of a loop. However, although it was never documented,
historical implementations of @code{awk} have treated the @code{continue}
statement outside of a loop as if it were a @code{next} statement
(@pxref{Next Statement, ,The @code{next} Statement}).
Recent versions of Unix @code{awk} no longer allow this usage.
@code{gawk} will support this use of @code{continue} only if
@samp{--traditional} has been specified on the command line
(@pxref{Options, ,Command Line Options}).
Otherwise, it will be treated as an error, since the POSIX standard
specifies that @code{continue} should only be used inside the body of a
loop (d.c.).
@node Next Statement, Nextfile Statement, Continue Statement, Statements
@section The @code{next} Statement
@cindex @code{next} statement
The @code{next} statement forces @code{awk} to immediately stop processing
the current record and go on to the next record. This means that no
further rules are executed for the current record. The rest of the
current rule's action is not executed either.
Contrast this with the effect of the @code{getline} function
(@pxref{Getline, ,Explicit Input with @code{getline}}). That too causes
@code{awk} to read the next record immediately, but it does not alter the
flow of control in any way. So the rest of the current action executes
with a new input record.
At the highest level, @code{awk} program execution is a loop that reads
an input record and then tests each rule's pattern against it. If you
think of this loop as a @code{for} statement whose body contains the
rules, then the @code{next} statement is analogous to a @code{continue}
statement: it skips to the end of the body of this implicit loop, and
executes the increment (which reads another record).
For example, if your @code{awk} program works only on records with four
fields, and you don't want it to fail when given bad input, you might
use this rule near the beginning of the program:
@example
@group
NF != 4 @{
err = sprintf("%s:%d: skipped: NF != 4\n", FILENAME, FNR)
print err > "/dev/stderr"
next
@}
@end group
@end example
@noindent
so that the following rules will not see the bad record. The error
message is redirected to the standard error output stream, as error
messages should be. @xref{Special Files, ,Special File Names in @code{gawk}}.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
According to the POSIX standard, the behavior is undefined if
the @code{next} statement is used in a @code{BEGIN} or @code{END} rule.
@code{gawk} will treat it as a syntax error.
Although POSIX permits it,
some other @code{awk} implementations don't allow the @code{next}
statement inside function bodies
(@pxref{User-defined, ,User-defined Functions}).
Just as any other @code{next} statement, a @code{next} inside a
function body reads the next record and starts processing it with the
first rule in the program.
If the @code{next} statement causes the end of the input to be reached,
then the code in any @code{END} rules will be executed.
@xref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}.
@cindex @code{next}, inside a user-defined function
@strong{Caution:} Some @code{awk} implementations generate a run-time
error if you use the @code{next} statement inside a user-defined function
(@pxref{User-defined, , User-defined Functions}).
@code{gawk} does not have this problem.
@node Nextfile Statement, Exit Statement, Next Statement, Statements
@section The @code{nextfile} Statement
@cindex @code{nextfile} statement
@cindex differences between @code{gawk} and @code{awk}
@code{gawk} provides the @code{nextfile} statement,
which is similar to the @code{next} statement.
However, instead of abandoning processing of the current record, the
@code{nextfile} statement instructs @code{gawk} to stop processing the
current data file.
Upon execution of the @code{nextfile} statement, @code{FILENAME} is
updated to the name of the next data file listed on the command line,
@code{FNR} is reset to one, @code{ARGIND} is incremented, and processing
starts over with the first rule in the progam. @xref{Built-in Variables}.
If the @code{nextfile} statement causes the end of the input to be reached,
then the code in any @code{END} rules will be executed.
@xref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}.
The @code{nextfile} statement is a @code{gawk} extension; it is not
(currently) available in any other @code{awk} implementation.
@xref{Nextfile Function, ,Implementing @code{nextfile} as a Function},
for a user-defined function you can use to simulate the @code{nextfile}
statement.
The @code{nextfile} statement would be useful if you have many data
files to process, and you expect that you
would not want to process every record in every file.
Normally, in order to move on to
the next data file, you would have to continue scanning the unwanted
records. The @code{nextfile} statement accomplishes this much more
efficiently.
@cindex @code{next file} statement
@strong{Caution:} Versions of @code{gawk} prior to 3.0 used two
words (@samp{next file}) for the @code{nextfile} statement. This was
changed in 3.0 to one word, since the treatment of @samp{file} was
inconsistent. When it appeared after @code{next}, it was a keyword.
Otherwise, it was a regular identifier. The old usage is still
accepted. However, @code{gawk} will generate a warning message, and
support for @code{next file} will eventually be discontinued in a
future version of @code{gawk}.
@node Exit Statement, , Nextfile Statement, Statements
@section The @code{exit} Statement
@cindex @code{exit} statement
The @code{exit} statement causes @code{awk} to immediately stop
executing the current rule and to stop processing input; any remaining input
is ignored. It looks like this:
@example
exit @r{[}@var{return code}@r{]}
@end example
If an @code{exit} statement is executed from a @code{BEGIN} rule the
program stops processing everything immediately. No input records are
read. However, if an @code{END} rule is present, it is executed
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}).
If @code{exit} is used as part of an @code{END} rule, it causes
the program to stop immediately.
An @code{exit} statement that is not part
of a @code{BEGIN} or @code{END} rule stops the execution of any further
automatic rules for the current record, skips reading any remaining input
records, and executes
the @code{END} rule if there is one.
If you do not want the @code{END} rule to do its job in this case, you
can set a variable to non-zero before the @code{exit} statement, and check
that variable in the @code{END} rule.
@xref{Assert Function, ,Assertions},
for an example that does this.
@cindex dark corner
If an argument is supplied to @code{exit}, its value is used as the exit
status code for the @code{awk} process. If no argument is supplied,
@code{exit} returns status zero (success). In the case where an argument
is supplied to a first @code{exit} statement, and then @code{exit} is
called a second time with no argument, the previously supplied exit value
is used (d.c.).
For example, let's say you've discovered an error condition you really
don't know how to handle. Conventionally, programs report this by
exiting with a non-zero status. Your @code{awk} program can do this
using an @code{exit} statement with a non-zero argument. Here is an
example:
@example
@group
BEGIN @{
if (("date" | getline date_now) < 0) @{
print "Can't get system date" > "/dev/stderr"
exit 1
@}
print "current date is", date_now
close("date")
@}
@end group
@end example
@node Built-in Variables, Arrays, Statements, Top
@chapter Built-in Variables
@cindex built-in variables
Most @code{awk} variables are available for you to use for your own
purposes; they never change except when your program assigns values to
them, and never affect anything except when your program examines them.
However, a few variables in @code{awk} have special built-in meanings.
Some of them @code{awk} examines automatically, so that they enable you
to tell @code{awk} how to do certain things. Others are set
automatically by @code{awk}, so that they carry information from the
internal workings of @code{awk} to your program.
This chapter documents all the built-in variables of @code{gawk}. Most
of them are also documented in the chapters describing their areas of
activity.
@menu
* User-modified:: Built-in variables that you change to control
@code{awk}.
* Auto-set:: Built-in variables where @code{awk} gives you
information.
* ARGC and ARGV:: Ways to use @code{ARGC} and @code{ARGV}.
@end menu
@node User-modified, Auto-set, Built-in Variables, Built-in Variables
@section Built-in Variables that Control @code{awk}
@cindex built-in variables, user modifiable
This is an alphabetical list of the variables which you can change to
control how @code{awk} does certain things. Those variables that are
specific to @code{gawk} are marked with an asterisk, @samp{*}.
@table @code
@vindex CONVFMT
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
@item CONVFMT
This string controls conversion of numbers to
strings (@pxref{Conversion, ,Conversion of Strings and Numbers}).
It works by being passed, in effect, as the first argument to the
@code{sprintf} function
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
Its default value is @code{"%.6g"}.
@code{CONVFMT} was introduced by the POSIX standard.
@vindex FIELDWIDTHS
@item FIELDWIDTHS *
This is a space separated list of columns that tells @code{gawk}
how to split input with fixed, columnar boundaries. It is an
experimental feature. Assigning to @code{FIELDWIDTHS}
overrides the use of @code{FS} for field splitting.
@xref{Constant Size, ,Reading Fixed-width Data}, for more information.
If @code{gawk} is in compatibility mode
(@pxref{Options, ,Command Line Options}), then @code{FIELDWIDTHS}
has no special meaning, and field splitting operations are done based
exclusively on the value of @code{FS}.
@vindex FS
@item FS
@code{FS} is the input field separator
(@pxref{Field Separators, ,Specifying How Fields are Separated}).
The value is a single-character string or a multi-character regular
expression that matches the separations between fields in an input
record. If the value is the null string (@code{""}), then each
character in the record becomes a separate field.
The default value is @w{@code{" "}}, a string consisting of a single
space. As a special exception, this value means that any
sequence of spaces, tabs, and/or newlines is a single separator.@footnote{In
POSIX @code{awk}, newline does not count as whitespace.} It also causes
spaces, tabs, and newlines at the beginning and end of a record to be ignored.
You can set the value of @code{FS} on the command line using the
@samp{-F} option:
@example
awk -F, '@var{program}' @var{input-files}
@end example
If @code{gawk} is using @code{FIELDWIDTHS} for field-splitting,
assigning a value to @code{FS} will cause @code{gawk} to return to
the normal, @code{FS}-based, field splitting. An easy way to do this
is to simply say @samp{FS = FS}, perhaps with an explanatory comment.
@vindex IGNORECASE
@item IGNORECASE *
If @code{IGNORECASE} is non-zero or non-null, then all string comparisons,
and all regular expression matching are case-independent. Thus, regexp
matching with @samp{~} and @samp{!~}, and the @code{gensub},
@code{gsub}, @code{index}, @code{match}, @code{split} and @code{sub}
functions, record termination with @code{RS}, and field splitting with
@code{FS} all ignore case when doing their particular regexp operations.
The value of @code{IGNORECASE} does @emph{not} affect array subscripting.
@xref{Case-sensitivity, ,Case-sensitivity in Matching}.
If @code{gawk} is in compatibility mode
(@pxref{Options, ,Command Line Options}),
then @code{IGNORECASE} has no special meaning, and string
and regexp operations are always case-sensitive.
@vindex OFMT
@item OFMT
This string controls conversion of numbers to
strings (@pxref{Conversion, ,Conversion of Strings and Numbers}) for
printing with the @code{print} statement. It works by being passed, in
effect, as the first argument to the @code{sprintf} function
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
Its default value is @code{"%.6g"}. Earlier versions of @code{awk}
also used @code{OFMT} to specify the format for converting numbers to
strings in general expressions; this is now done by @code{CONVFMT}.
@vindex OFS
@item OFS
This is the output field separator (@pxref{Output Separators}). It is
output between the fields output by a @code{print} statement. Its
default value is @w{@code{" "}}, a string consisting of a single space.
@vindex ORS
@item ORS
This is the output record separator. It is output at the end of every
@code{print} statement. Its default value is @code{"\n"}.
(@xref{Output Separators}.)
@vindex RS
@item RS
This is @code{awk}'s input record separator. Its default value is a string
containing a single newline character, which means that an input record
consists of a single line of text.
It can also be the null string, in which case records are separated by
runs of blank lines, or a regexp, in which case records are separated by
matches of the regexp in the input text.
(@xref{Records, ,How Input is Split into Records}.)
@vindex SUBSEP
@item SUBSEP
@code{SUBSEP} is the subscript separator. It has the default value of
@code{"\034"}, and is used to separate the parts of the indices of a
multi-dimensional array. Thus, the expression @code{@w{foo["A", "B"]}}
really accesses @code{foo["A\034B"]}
(@pxref{Multi-dimensional, ,Multi-dimensional Arrays}).
@end table
@node Auto-set, ARGC and ARGV, User-modified, Built-in Variables
@section Built-in Variables that Convey Information
@cindex built-in variables, convey information
This is an alphabetical list of the variables that are set
automatically by @code{awk} on certain occasions in order to provide
information to your program. Those variables that are specific to
@code{gawk} are marked with an asterisk, @samp{*}.
@table @code
@vindex ARGC
@vindex ARGV
@item ARGC
@itemx ARGV
The command-line arguments available to @code{awk} programs are stored in
an array called @code{ARGV}. @code{ARGC} is the number of command-line
arguments present. @xref{Other Arguments, ,Other Command Line Arguments}.
Unlike most @code{awk} arrays,
@code{ARGV} is indexed from zero to @code{ARGC} @minus{} 1. For example:
@example
@group
$ awk 'BEGIN @{
> for (i = 0; i < ARGC; i++)
> print ARGV[i]
> @}' inventory-shipped BBS-list
@print{} awk
@print{} inventory-shipped
@print{} BBS-list
@end group
@end example
@noindent
In this example, @code{ARGV[0]} contains @code{"awk"}, @code{ARGV[1]}
contains @code{"inventory-shipped"}, and @code{ARGV[2]} contains
@code{"BBS-list"}. The value of @code{ARGC} is three, one more than the
index of the last element in @code{ARGV}, since the elements are numbered
from zero.
The names @code{ARGC} and @code{ARGV}, as well as the convention of indexing
the array from zero to @code{ARGC} @minus{} 1, are derived from the C language's
method of accessing command line arguments.
@xref{ARGC and ARGV, , Using @code{ARGC} and @code{ARGV}}, for information
about how @code{awk} uses these variables.
@vindex ARGIND
@item ARGIND *
The index in @code{ARGV} of the current file being processed.
Every time @code{gawk} opens a new data file for processing, it sets
@code{ARGIND} to the index in @code{ARGV} of the file name.
When @code{gawk} is processing the input files, it is always
true that @samp{FILENAME == ARGV[ARGIND]}.
This variable is useful in file processing; it allows you to tell how far
along you are in the list of data files, and to distinguish between
successive instances of the same filename on the command line.
While you can change the value of @code{ARGIND} within your @code{awk}
program, @code{gawk} will automatically set it to a new value when the
next file is opened.
This variable is a @code{gawk} extension. In other @code{awk} implementations,
or if @code{gawk} is in compatibility mode
(@pxref{Options, ,Command Line Options}),
it is not special.
@vindex ENVIRON
@item ENVIRON
An associative array that contains the values of the environment. The array
indices are the environment variable names; the values are the values of
the particular environment variables. For example,
@code{ENVIRON["HOME"]} might be @file{/home/arnold}. Changing this array
does not affect the environment passed on to any programs that
@code{awk} may spawn via redirection or the @code{system} function.
(In a future version of @code{gawk}, it may do so.)
Some operating systems may not have environment variables.
On such systems, the @code{ENVIRON} array is empty (except for
@w{@code{ENVIRON["AWKPATH"]}}).
@vindex ERRNO
@item ERRNO *
If a system error occurs either doing a redirection for @code{getline},
during a read for @code{getline}, or during a @code{close} operation,
then @code{ERRNO} will contain a string describing the error.
This variable is a @code{gawk} extension. In other @code{awk} implementations,
or if @code{gawk} is in compatibility mode
(@pxref{Options, ,Command Line Options}),
it is not special.
@cindex dark corner
@vindex FILENAME
@item FILENAME
This is the name of the file that @code{awk} is currently reading.
When no data files are listed on the command line, @code{awk} reads
from the standard input, and @code{FILENAME} is set to @code{"-"}.
@code{FILENAME} is changed each time a new file is read
(@pxref{Reading Files, ,Reading Input Files}).
Inside a @code{BEGIN} rule, the value of @code{FILENAME} is
@code{""}, since there are no input files being processed
yet.@footnote{Some early implementations of Unix @code{awk} initialized
@code{FILENAME} to @code{"-"}, even if there were data files to be
processed. This behavior was incorrect, and should not be relied
upon in your programs.} (d.c.)
@vindex FNR
@item FNR
@code{FNR} is the current record number in the current file. @code{FNR} is
incremented each time a new record is read
(@pxref{Getline, ,Explicit Input with @code{getline}}). It is reinitialized
to zero each time a new input file is started.
@vindex NF
@item NF
@code{NF} is the number of fields in the current input record.
@code{NF} is set each time a new record is read, when a new field is
created, or when @code{$0} changes (@pxref{Fields, ,Examining Fields}).
@vindex NR
@item NR
This is the number of input records @code{awk} has processed since
the beginning of the program's execution
(@pxref{Records, ,How Input is Split into Records}).
@code{NR} is set each time a new record is read.
@vindex RLENGTH
@item RLENGTH
@code{RLENGTH} is the length of the substring matched by the
@code{match} function
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
@code{RLENGTH} is set by invoking the @code{match} function. Its value
is the length of the matched string, or @minus{}1 if no match was found.
@vindex RSTART
@item RSTART
@code{RSTART} is the start-index in characters of the substring matched by the
@code{match} function
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
@code{RSTART} is set by invoking the @code{match} function. Its value
is the position of the string where the matched substring starts, or zero
if no match was found.
@vindex RT
@item RT *
@code{RT} is set each time a record is read. It contains the input text
that matched the text denoted by @code{RS}, the record separator.
This variable is a @code{gawk} extension. In other @code{awk} implementations,
or if @code{gawk} is in compatibility mode
(@pxref{Options, ,Command Line Options}),
it is not special.
@end table
@cindex dark corner
A side note about @code{NR} and @code{FNR}.
@code{awk} simply increments both of these variables
each time it reads a record, instead of setting them to the absolute
value of the number of records read. This means that your program can
change these variables, and their new values will be incremented for
each record (d.c.). For example:
@example
@group
$ echo '1
> 2
> 3
> 4' | awk 'NR == 2 @{ NR = 17 @}
> @{ print NR @}'
@print{} 1
@print{} 17
@print{} 18
@print{} 19
@end group
@end example
@noindent
Before @code{FNR} was added to the @code{awk} language
(@pxref{V7/SVR3.1, ,Major Changes between V7 and SVR3.1}),
many @code{awk} programs used this feature to track the number of
records in a file by resetting @code{NR} to zero when @code{FILENAME}
changed.
@node ARGC and ARGV, , Auto-set, Built-in Variables
@section Using @code{ARGC} and @code{ARGV}
In @ref{Auto-set, , Built-in Variables that Convey Information},
you saw this program describing the information contained in @code{ARGC}
and @code{ARGV}:
@example
@group
$ awk 'BEGIN @{
> for (i = 0; i < ARGC; i++)
> print ARGV[i]
> @}' inventory-shipped BBS-list
@print{} awk
@print{} inventory-shipped
@print{} BBS-list
@end group
@end example
@noindent
In this example, @code{ARGV[0]} contains @code{"awk"}, @code{ARGV[1]}
contains @code{"inventory-shipped"}, and @code{ARGV[2]} contains
@code{"BBS-list"}.
Notice that the @code{awk} program is not entered in @code{ARGV}. The
other special command line options, with their arguments, are also not
entered. But variable assignments on the command line @emph{are}
treated as arguments, and do show up in the @code{ARGV} array.
Your program can alter @code{ARGC} and the elements of @code{ARGV}.
Each time @code{awk} reaches the end of an input file, it uses the next
element of @code{ARGV} as the name of the next input file. By storing a
different string there, your program can change which files are read.
You can use @code{"-"} to represent the standard input. By storing
additional elements and incrementing @code{ARGC} you can cause
additional files to be read.
If you decrease the value of @code{ARGC}, that eliminates input files
from the end of the list. By recording the old value of @code{ARGC}
elsewhere, your program can treat the eliminated arguments as
something other than file names.
To eliminate a file from the middle of the list, store the null string
(@code{""}) into @code{ARGV} in place of the file's name. As a
special feature, @code{awk} ignores file names that have been
replaced with the null string.
You may also use the @code{delete} statement to remove elements from
@code{ARGV} (@pxref{Delete, ,The @code{delete} Statement}).
All of these actions are typically done from the @code{BEGIN} rule,
before actual processing of the input begins.
@xref{Split Program, ,Splitting a Large File Into Pieces}, and see
@ref{Tee Program, ,Duplicating Output Into Multiple Files}, for an example
of each way of removing elements from @code{ARGV}.
The following fragment processes @code{ARGV} in order to examine, and
then remove, command line options.
@example
@group
BEGIN @{
for (i = 1; i < ARGC; i++) @{
if (ARGV[i] == "-v")
verbose = 1
else if (ARGV[i] == "-d")
debug = 1
@end group
@group
else if (ARGV[i] ~ /^-?/) @{
e = sprintf("%s: unrecognized option -- %c",
ARGV[0], substr(ARGV[i], 1, ,1))
print e > "/dev/stderr"
@} else
break
delete ARGV[i]
@}
@}
@end group
@end example
To actually get the options into the @code{awk} program, you have to
end the @code{awk} options with @samp{--}, and then supply your options,
like so:
@example
awk -f myprog -- -v -d file1 file2 @dots{}
@end example
@cindex differences between @code{gawk} and @code{awk}
This is not necessary in @code{gawk}: Unless @samp{--posix} has been
specified, @code{gawk} silently puts any unrecognized options into
@code{ARGV} for the @code{awk} program to deal with.
As soon as it
sees an unknown option, @code{gawk} stops looking for other options it might
otherwise recognize. The above example with @code{gawk} would be:
@example
gawk -f myprog -d -v file1 file2 @dots{}
@end example
@noindent
Since @samp{-d} is not a valid @code{gawk} option, the following @samp{-v}
is passed on to the @code{awk} program.
@node Arrays, Built-in, Built-in Variables, Top
@chapter Arrays in @code{awk}
An @dfn{array} is a table of values, called @dfn{elements}. The
elements of an array are distinguished by their indices. @dfn{Indices}
may be either numbers or strings. @code{awk} maintains a single set
of names that may be used for naming variables, arrays and functions
(@pxref{User-defined, ,User-defined Functions}).
Thus, you cannot have a variable and an array with the same name in the
same @code{awk} program.
@menu
* Array Intro:: Introduction to Arrays
* Reference to Elements:: How to examine one element of an array.
* Assigning Elements:: How to change an element of an array.
* Array Example:: Basic Example of an Array
* Scanning an Array:: A variation of the @code{for} statement. It
loops through the indices of an array's
existing elements.
* Delete:: The @code{delete} statement removes an element
from an array.
* Numeric Array Subscripts:: How to use numbers as subscripts in
@code{awk}.
* Uninitialized Subscripts:: Using Uninitialized variables as subscripts.
* Multi-dimensional:: Emulating multi-dimensional arrays in
@code{awk}.
* Multi-scanning:: Scanning multi-dimensional arrays.
@end menu
@node Array Intro, Reference to Elements, Arrays, Arrays
@section Introduction to Arrays
@cindex arrays
The @code{awk} language provides one-dimensional @dfn{arrays} for storing groups
of related strings or numbers.
Every @code{awk} array must have a name. Array names have the same
syntax as variable names; any valid variable name would also be a valid
array name. But you cannot use one name in both ways (as an array and
as a variable) in one @code{awk} program.
Arrays in @code{awk} superficially resemble arrays in other programming
languages; but there are fundamental differences. In @code{awk}, you
don't need to specify the size of an array before you start to use it.
Additionally, any number or string in @code{awk} may be used as an
array index, not just consecutive integers.
In most other languages, you have to @dfn{declare} an array and specify
how many elements or components it contains. In such languages, the
declaration causes a contiguous block of memory to be allocated for that
many elements. An index in the array usually must be a positive integer; for
example, the index zero specifies the first element in the array, which is
actually stored at the beginning of the block of memory. Index one
specifies the second element, which is stored in memory right after the
first element, and so on. It is impossible to add more elements to the
array, because it has room for only as many elements as you declared.
(Some languages allow arbitrary starting and ending indices,
e.g., @samp{15 .. 27}, but the size of the array is still fixed when
the array is declared.)
A contiguous array of four elements might look like this,
conceptually, if the element values are eight, @code{"foo"},
@code{""} and 30:
@iftex
@c from Karl Berry, much thanks for the help.
@tex
\bigskip % space above the table (about 1 linespace)
\offinterlineskip
\newdimen\width \width = 1.5cm
\newdimen\hwidth \hwidth = 4\width \advance\hwidth by 2pt % 5 * 0.4pt
\centerline{\vbox{
\halign{\strut\hfil\ignorespaces#&&\vrule#&\hbox to\width{\hfil#\unskip\hfil}\cr
\noalign{\hrule width\hwidth}
&&{\tt 8} &&{\tt "foo"} &&{\tt ""} &&{\tt 30} &&\quad value\cr
\noalign{\hrule width\hwidth}
\noalign{\smallskip}
&\omit&0&\omit &1 &\omit&2 &\omit&3 &\omit&\quad index\cr
}
}}
@end tex
@end iftex
@ifinfo
@example
+---------+---------+--------+---------+
| 8 | "foo" | "" | 30 | @r{value}
+---------+---------+--------+---------+
0 1 2 3 @r{index}
@end example
@end ifinfo
@noindent
Only the values are stored; the indices are implicit from the order of
the values. Eight is the value at index zero, because eight appears in the
position with zero elements before it.
@cindex arrays, definition of
@cindex associative arrays
@cindex arrays, associative
Arrays in @code{awk} are different: they are @dfn{associative}. This means
that each array is a collection of pairs: an index, and its corresponding
array element value:
@example
@r{Element} 4 @r{Value} 30
@r{Element} 2 @r{Value} "foo"
@r{Element} 1 @r{Value} 8
@r{Element} 3 @r{Value} ""
@end example
@noindent
We have shown the pairs in jumbled order because their order is irrelevant.
One advantage of associative arrays is that new pairs can be added
at any time. For example, suppose we add to the above array a tenth element
whose value is @w{@code{"number ten"}}. The result is this:
@example
@r{Element} 10 @r{Value} "number ten"
@r{Element} 4 @r{Value} 30
@r{Element} 2 @r{Value} "foo"
@r{Element} 1 @r{Value} 8
@r{Element} 3 @r{Value} ""
@end example
@noindent
@cindex sparse arrays
@cindex arrays, sparse
Now the array is @dfn{sparse}, which just means some indices are missing:
it has elements 1--4 and 10, but doesn't have elements 5, 6, 7, 8, or 9.
@c ok, I should spell out the above, but ...
Another consequence of associative arrays is that the indices don't
have to be positive integers. Any number, or even a string, can be
an index. For example, here is an array which translates words from
English into French:
@example
@r{Element} "dog" @r{Value} "chien"
@r{Element} "cat" @r{Value} "chat"
@r{Element} "one" @r{Value} "un"
@r{Element} 1 @r{Value} "un"
@end example
@noindent
Here we decided to translate the number one in both spelled-out and
numeric form---thus illustrating that a single array can have both
numbers and strings as indices.
(In fact, array subscripts are always strings; this is discussed
in more detail in
@ref{Numeric Array Subscripts, ,Using Numbers to Subscript Arrays}.)
@cindex Array subscripts and @code{IGNORECASE}
@cindex @code{IGNORECASE} and array subscripts
@vindex IGNORECASE
The value of @code{IGNORECASE} has no effect upon array subscripting.
You must use the exact same string value to retrieve an array element
as you used to store it.
When @code{awk} creates an array for you, e.g., with the @code{split}
built-in function,
that array's indices are consecutive integers starting at one.
(@xref{String Functions, ,Built-in Functions for String Manipulation}.)
@node Reference to Elements, Assigning Elements, Array Intro, Arrays
@section Referring to an Array Element
@cindex array reference
@cindex element of array
@cindex reference to array
The principal way of using an array is to refer to one of its elements.
An array reference is an expression which looks like this:
@example
@var{array}[@var{index}]
@end example
@noindent
Here, @var{array} is the name of an array. The expression @var{index} is
the index of the element of the array that you want.
The value of the array reference is the current value of that array
element. For example, @code{foo[4.3]} is an expression for the element
of array @code{foo} at index @samp{4.3}.
If you refer to an array element that has no recorded value, the value
of the reference is @code{""}, the null string. This includes elements
to which you have not assigned any value, and elements that have been
deleted (@pxref{Delete, ,The @code{delete} Statement}). Such a reference
automatically creates that array element, with the null string as its value.
(In some cases, this is unfortunate, because it might waste memory inside
@code{awk}.)
@cindex arrays, presence of elements
@cindex arrays, the @code{in} operator
You can find out if an element exists in an array at a certain index with
the expression:
@example
@var{index} in @var{array}
@end example
@noindent
This expression tests whether or not the particular index exists,
without the side effect of creating that element if it is not present.
The expression has the value one (true) if @code{@var{array}[@var{index}]}
exists, and zero (false) if it does not exist.
For example, to test whether the array @code{frequencies} contains the
index @samp{2}, you could write this statement:
@example
if (2 in frequencies)
print "Subscript 2 is present."
@end example
Note that this is @emph{not} a test of whether or not the array
@code{frequencies} contains an element whose @emph{value} is two.
(There is no way to do that except to scan all the elements.) Also, this
@emph{does not} create @code{frequencies[2]}, while the following
(incorrect) alternative would do so:
@example
if (frequencies[2] != "")
print "Subscript 2 is present."
@end example
@node Assigning Elements, Array Example, Reference to Elements, Arrays
@section Assigning Array Elements
@cindex array assignment
@cindex element assignment
Array elements are lvalues: they can be assigned values just like
@code{awk} variables:
@example
@var{array}[@var{subscript}] = @var{value}
@end example
@noindent
Here @var{array} is the name of your array. The expression
@var{subscript} is the index of the element of the array that you want
to assign a value. The expression @var{value} is the value you are
assigning to that element of the array.
@node Array Example, Scanning an Array, Assigning Elements, Arrays
@section Basic Array Example
The following program takes a list of lines, each beginning with a line
number, and prints them out in order of line number. The line numbers are
not in order, however, when they are first read: they are scrambled. This
program sorts the lines by making an array using the line numbers as
subscripts. It then prints out the lines in sorted order of their numbers.
It is a very simple program, and gets confused if it encounters repeated
numbers, gaps, or lines that don't begin with a number.
@example
@c file eg/misc/arraymax.awk
@{
if ($1 > max)
max = $1
arr[$1] = $0
@}
END @{
for (x = 1; x <= max; x++)
print arr[x]
@}
@c endfile
@end example
The first rule keeps track of the largest line number seen so far;
it also stores each line into the array @code{arr}, at an index that
is the line's number.
The second rule runs after all the input has been read, to print out
all the lines.
When this program is run with the following input:
@example
@group
@c file eg/misc/arraymax.data
5 I am the Five man
2 Who are you? The new number two!
4 . . . And four on the floor
1 Who is number one?
3 I three you.
@c endfile
@end group
@end example
@noindent
its output is this:
@example
1 Who is number one?
2 Who are you? The new number two!
3 I three you.
4 . . . And four on the floor
5 I am the Five man
@end example
If a line number is repeated, the last line with a given number overrides
the others.
Gaps in the line numbers can be handled with an easy improvement to the
program's @code{END} rule:
@example
END @{
for (x = 1; x <= max; x++)
if (x in arr)
print arr[x]
@}
@end example
@node Scanning an Array, Delete, Array Example, Arrays
@section Scanning All Elements of an Array
@cindex @code{for (x in @dots{})}
@cindex arrays, special @code{for} statement
@cindex scanning an array
In programs that use arrays, you often need a loop that executes
once for each element of an array. In other languages, where arrays are
contiguous and indices are limited to positive integers, this is
easy: you can
find all the valid indices by counting from the lowest index
up to the highest. This
technique won't do the job in @code{awk}, since any number or string
can be an array index. So @code{awk} has a special kind of @code{for}
statement for scanning an array:
@example
for (@var{var} in @var{array})
@var{body}
@end example
@noindent
This loop executes @var{body} once for each index in @var{array} that your
program has previously used, with the
variable @var{var} set to that index.
Here is a program that uses this form of the @code{for} statement. The
first rule scans the input records and notes which words appear (at
least once) in the input, by storing a one into the array @code{used} with
the word as index. The second rule scans the elements of @code{used} to
find all the distinct words that appear in the input. It prints each
word that is more than 10 characters long, and also prints the number of
such words. @xref{String Functions, ,Built-in Functions for String Manipulation}, for more information
on the built-in function @code{length}.
@example
# Record a 1 for each word that is used at least once.
@{
for (i = 1; i <= NF; i++)
used[$i] = 1
@}
# Find number of distinct words more than 10 characters long.
END @{
for (x in used)
if (length(x) > 10) @{
++num_long_words
print x
@}
print num_long_words, "words longer than 10 characters"
@}
@end example
@noindent
@xref{Word Sorting, ,Generating Word Usage Counts},
for a more detailed example of this type.
The order in which elements of the array are accessed by this statement
is determined by the internal arrangement of the array elements within
@code{awk} and cannot be controlled or changed. This can lead to
problems if new elements are added to @var{array} by statements in
the loop body; you cannot predict whether or not the @code{for} loop will
reach them. Similarly, changing @var{var} inside the loop may produce
strange results. It is best to avoid such things.
@node Delete, Numeric Array Subscripts, Scanning an Array, Arrays
@section The @code{delete} Statement
@cindex @code{delete} statement
@cindex deleting elements of arrays
@cindex removing elements of arrays
@cindex arrays, deleting an element
You can remove an individual element of an array using the @code{delete}
statement:
@example
delete @var{array}[@var{index}]
@end example
Once you have deleted an array element, you can no longer obtain any
value the element once had. It is as if you had never referred
to it and had never given it any value.
Here is an example of deleting elements in an array:
@example
for (i in frequencies)
delete frequencies[i]
@end example
@noindent
This example removes all the elements from the array @code{frequencies}.
If you delete an element, a subsequent @code{for} statement to scan the array
will not report that element, and the @code{in} operator to check for
the presence of that element will return zero (i.e.@: false):
@example
delete foo[4]
if (4 in foo)
print "This will never be printed"
@end example
It is important to note that deleting an element is @emph{not} the
same as assigning it a null value (the empty string, @code{""}).
@example
foo[4] = ""
if (4 in foo)
print "This is printed, even though foo[4] is empty"
@end example
It is not an error to delete an element that does not exist.
@cindex arrays, deleting entire contents
@cindex deleting entire arrays
@cindex differences between @code{gawk} and @code{awk}
You can delete all the elements of an array with a single statement,
by leaving off the subscript in the @code{delete} statement.
@example
delete @var{array}
@end example
This ability is a @code{gawk} extension; it is not available in
compatibility mode (@pxref{Options, ,Command Line Options}).
Using this version of the @code{delete} statement is about three times
more efficient than the equivalent loop that deletes each element one
at a time.
@cindex portability issues
The following statement provides a portable, but non-obvious way to clear
out an array.
@cindex Brennan, Michael
@example
@group
# thanks to Michael Brennan for pointing this out
split("", array)
@end group
@end example
The @code{split} function
(@pxref{String Functions, ,Built-in Functions for String Manipulation})
clears out the target array first. This call asks it to split
apart the null string. Since there is no data to split out, the
function simply clears the array and then returns.
@node Numeric Array Subscripts, Uninitialized Subscripts, Delete, Arrays
@section Using Numbers to Subscript Arrays
An important aspect of arrays to remember is that @emph{array subscripts
are always strings}. If you use a numeric value as a subscript,
it will be converted to a string value before it is used for subscripting
(@pxref{Conversion, ,Conversion of Strings and Numbers}).
@cindex conversions, during subscripting
@cindex numbers, used as subscripts
@vindex CONVFMT
This means that the value of the built-in variable @code{CONVFMT} can potentially
affect how your program accesses elements of an array. For example:
@example
xyz = 12.153
data[xyz] = 1
CONVFMT = "%2.2f"
@group
if (xyz in data)
printf "%s is in data\n", xyz
else
printf "%s is not in data\n", xyz
@end group
@end example
@noindent
This prints @samp{12.15 is not in data}. The first statement gives
@code{xyz} a numeric value. Assigning to
@code{data[xyz]} subscripts @code{data} with the string value @code{"12.153"}
(using the default conversion value of @code{CONVFMT}, @code{"%.6g"}),
and assigns one to @code{data["12.153"]}. The program then changes
the value of @code{CONVFMT}. The test @samp{(xyz in data)} generates a new
string value from @code{xyz}, this time @code{"12.15"}, since the value of
@code{CONVFMT} only allows two significant digits. This test fails,
since @code{"12.15"} is a different string from @code{"12.153"}.
According to the rules for conversions
(@pxref{Conversion, ,Conversion of Strings and Numbers}), integer
values are always converted to strings as integers, no matter what the
value of @code{CONVFMT} may happen to be. So the usual case of:
@example
for (i = 1; i <= maxsub; i++)
@i{do something with} array[i]
@end example
@noindent
will work, no matter what the value of @code{CONVFMT}.
Like many things in @code{awk}, the majority of the time things work
as you would expect them to work. But it is useful to have a precise
knowledge of the actual rules, since sometimes they can have a subtle
effect on your programs.
@node Uninitialized Subscripts, Multi-dimensional, Numeric Array Subscripts, Arrays
@section Using Uninitialized Variables as Subscripts
@cindex uninitialized variables, as array subscripts
@cindex array subscripts, uninitialized variables
Suppose you want to print your input data in reverse order.
A reasonable attempt at a program to do so (with some test
data) might look like this:
@example
@group
$ echo 'line 1
> line 2
> line 3' | awk '@{ l[lines] = $0; ++lines @}
> END @{
> for (i = lines-1; i >= 0; --i)
> print l[i]
> @}'
@print{} line 3
@print{} line 2
@end group
@end example
Unfortunately, the very first line of input data did not come out in the
output!
At first glance, this program should have worked. The variable @code{lines}
is uninitialized, and uninitialized variables have the numeric value zero.
So, the value of @code{l[0]} should have been printed.
The issue here is that subscripts for @code{awk} arrays are @strong{always}
strings. And uninitialized variables, when used as strings, have the
value @code{""}, not zero. Thus, @samp{line 1} ended up stored in
@code{l[""]}.
The following version of the program works correctly:
@example
@{ l[lines++] = $0 @}
END @{
for (i = lines - 1; i >= 0; --i)
print l[i]
@}
@end example
Here, the @samp{++} forces @code{lines} to be numeric, thus making
the ``old value'' numeric zero, which is then converted to @code{"0"}
as the array subscript.
@cindex null string, as array subscript
@cindex dark corner
As we have just seen, even though it is somewhat unusual, the null string
(@code{""}) is a valid array subscript (d.c.). If @samp{--lint} is provided
on the command line (@pxref{Options, ,Command Line Options}),
@code{gawk} will warn about the use of the null string as a subscript.
@node Multi-dimensional, Multi-scanning, Uninitialized Subscripts, Arrays
@section Multi-dimensional Arrays
@cindex subscripts in arrays
@cindex arrays, multi-dimensional subscripts
@cindex multi-dimensional subscripts
A multi-dimensional array is an array in which an element is identified
by a sequence of indices, instead of a single index. For example, a
two-dimensional array requires two indices. The usual way (in most
languages, including @code{awk}) to refer to an element of a
two-dimensional array named @code{grid} is with
@code{grid[@var{x},@var{y}]}.
@vindex SUBSEP
Multi-dimensional arrays are supported in @code{awk} through
concatenation of indices into one string. What happens is that
@code{awk} converts the indices into strings
(@pxref{Conversion, ,Conversion of Strings and Numbers}) and
concatenates them together, with a separator between them. This creates
a single string that describes the values of the separate indices. The
combined string is used as a single index into an ordinary,
one-dimensional array. The separator used is the value of the built-in
variable @code{SUBSEP}.
For example, suppose we evaluate the expression @samp{foo[5,12] = "value"}
when the value of @code{SUBSEP} is @code{"@@"}. The numbers five and 12 are
converted to strings and
concatenated with an @samp{@@} between them, yielding @code{"5@@12"}; thus,
the array element @code{foo["5@@12"]} is set to @code{"value"}.
Once the element's value is stored, @code{awk} has no record of whether
it was stored with a single index or a sequence of indices. The two
expressions @samp{foo[5,12]} and @w{@samp{foo[5 SUBSEP 12]}} are always
equivalent.
The default value of @code{SUBSEP} is the string @code{"\034"},
which contains a non-printing character that is unlikely to appear in an
@code{awk} program or in most input data.
The usefulness of choosing an unlikely character comes from the fact
that index values that contain a string matching @code{SUBSEP} lead to
combined strings that are ambiguous. Suppose that @code{SUBSEP} were
@code{"@@"}; then @w{@samp{foo["a@@b", "c"]}} and @w{@samp{foo["a",
"b@@c"]}} would be indistinguishable because both would actually be
stored as @samp{foo["a@@b@@c"]}.
You can test whether a particular index-sequence exists in a
``multi-dimensional'' array with the same operator @samp{in} used for single
dimensional arrays. Instead of a single index as the left-hand operand,
write the whole sequence of indices, separated by commas, in
parentheses:
@example
(@var{subscript1}, @var{subscript2}, @dots{}) in @var{array}
@end example
The following example treats its input as a two-dimensional array of
fields; it rotates this array 90 degrees clockwise and prints the
result. It assumes that all lines have the same number of
elements.
@example
@group
awk '@{
if (max_nf < NF)
max_nf = NF
max_nr = NR
for (x = 1; x <= NF; x++)
vector[x, NR] = $x
@}
@end group
@group
END @{
for (x = 1; x <= max_nf; x++) @{
for (y = max_nr; y >= 1; --y)
printf("%s ", vector[x, y])
printf("\n")
@}
@}'
@end group
@end example
@noindent
When given the input:
@example
@group
1 2 3 4 5 6
2 3 4 5 6 1
3 4 5 6 1 2
4 5 6 1 2 3
@end group
@end example
@noindent
it produces:
@example
@group
4 3 2 1
5 4 3 2
6 5 4 3
1 6 5 4
2 1 6 5
3 2 1 6
@end group
@end example
@node Multi-scanning, , Multi-dimensional, Arrays
@section Scanning Multi-dimensional Arrays
There is no special @code{for} statement for scanning a
``multi-dimensional'' array; there cannot be one, because in truth there
are no multi-dimensional arrays or elements; there is only a
multi-dimensional @emph{way of accessing} an array.
However, if your program has an array that is always accessed as
multi-dimensional, you can get the effect of scanning it by combining
the scanning @code{for} statement
(@pxref{Scanning an Array, ,Scanning All Elements of an Array}) with the
@code{split} built-in function
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
It works like this:
@example
for (combined in array) @{
split(combined, separate, SUBSEP)
@dots{}
@}
@end example
@noindent
This sets @code{combined} to
each concatenated, combined index in the array, and splits it
into the individual indices by breaking it apart where the value of
@code{SUBSEP} appears. The split-out indices become the elements of
the array @code{separate}.
Thus, suppose you have previously stored a value in @code{array[1, "foo"]};
then an element with index @code{"1\034foo"} exists in
@code{array}. (Recall that the default value of @code{SUBSEP} is
the character with code 034.) Sooner or later the @code{for} statement
will find that index and do an iteration with @code{combined} set to
@code{"1\034foo"}. Then the @code{split} function is called as
follows:
@example
split("1\034foo", separate, "\034")
@end example
@noindent
The result of this is to set @code{separate[1]} to @code{"1"} and
@code{separate[2]} to @code{"foo"}. Presto, the original sequence of
separate indices has been recovered.
@node Built-in, User-defined, Arrays, Top
@chapter Built-in Functions
@c 2e: USE TEXINFO-2 FUNCTION DEFINITION STUFF!!!!!!!!!!!!!
@cindex built-in functions
@dfn{Built-in} functions are functions that are always available for
your @code{awk} program to call. This chapter defines all the built-in
functions in @code{awk}; some of them are mentioned in other sections,
but they are summarized here for your convenience. (You can also define
new functions yourself. @xref{User-defined, ,User-defined Functions}.)
@menu
* Calling Built-in:: How to call built-in functions.
* Numeric Functions:: Functions that work with numbers, including
@code{int}, @code{sin} and @code{rand}.
* String Functions:: Functions for string manipulation, such as
@code{split}, @code{match}, and
@code{sprintf}.
* I/O Functions:: Functions for files and shell commands.
* Time Functions:: Functions for dealing with time stamps.
@end menu
@node Calling Built-in, Numeric Functions, Built-in, Built-in
@section Calling Built-in Functions
To call a built-in function, write the name of the function followed
by arguments in parentheses. For example, @samp{atan2(y + z, 1)}
is a call to the function @code{atan2}, with two arguments.
Whitespace is ignored between the built-in function name and the
open-parenthesis, but we recommend that you avoid using whitespace
there. User-defined functions do not permit whitespace in this way, and
you will find it easier to avoid mistakes by following a simple
convention which always works: no whitespace after a function name.
@cindex differences between @code{gawk} and @code{awk}
Each built-in function accepts a certain number of arguments.
In some cases, arguments can be omitted. The defaults for omitted
arguments vary from function to function and are described under the
individual functions. In some @code{awk} implementations, extra
arguments given to built-in functions are ignored. However, in @code{gawk},
it is a fatal error to give extra arguments to a built-in function.
When a function is called, expressions that create the function's actual
parameters are evaluated completely before the function call is performed.
For example, in the code fragment:
@example
i = 4
j = sqrt(i++)
@end example
@noindent
the variable @code{i} is set to five before @code{sqrt} is called
with a value of four for its actual parameter.
@cindex evaluation, order of
@cindex order of evaluation
The order of evaluation of the expressions used for the function's
parameters is undefined. Thus, you should not write programs that
assume that parameters are evaluated from left to right or from
right to left. For example,
@example
i = 5
j = atan2(i++, i *= 2)
@end example
If the order of evaluation is left to right, then @code{i} first becomes
six, and then 12, and @code{atan2} is called with the two arguments six
and 12. But if the order of evaluation is right to left, @code{i}
first becomes 10, and then 11, and @code{atan2} is called with the
two arguments 11 and 10.
@node Numeric Functions, String Functions, Calling Built-in, Built-in
@section Numeric Built-in Functions
Here is a full list of built-in functions that work with numbers.
Optional parameters are enclosed in square brackets (``['' and ``]'').
@table @code
@item int(@var{x})
@findex int
This produces the nearest integer to @var{x}, located between @var{x} and zero,
truncated toward zero.
For example, @code{int(3)} is three, @code{int(3.9)} is three, @code{int(-3.9)}
is @minus{}3, and @code{int(-3)} is @minus{}3 as well.
@item sqrt(@var{x})
@findex sqrt
This gives you the positive square root of @var{x}. It reports an error
if @var{x} is negative. Thus, @code{sqrt(4)} is two.
@item exp(@var{x})
@findex exp
This gives you the exponential of @var{x} (@code{e ^ @var{x}}), or reports
an error if @var{x} is out of range. The range of values @var{x} can have
depends on your machine's floating point representation.
@item log(@var{x})
@findex log
This gives you the natural logarithm of @var{x}, if @var{x} is positive;
otherwise, it reports an error.
@item sin(@var{x})
@findex sin
This gives you the sine of @var{x}, with @var{x} in radians.
@item cos(@var{x})
@findex cos
This gives you the cosine of @var{x}, with @var{x} in radians.
@item atan2(@var{y}, @var{x})
@findex atan2
This gives you the arctangent of @code{@var{y} / @var{x}} in radians.
@item rand()
@findex rand
This gives you a random number. The values of @code{rand} are
uniformly-distributed between zero and one.
The value is never zero and never one.
Often you want random integers instead. Here is a user-defined function
you can use to obtain a random non-negative integer less than @var{n}:
@example
function randint(n) @{
return int(n * rand())
@}
@end example
@noindent
The multiplication produces a random real number greater than zero and less
than @code{n}. We then make it an integer (using @code{int}) between zero
and @code{n} @minus{} 1, inclusive.
Here is an example where a similar function is used to produce
random integers between one and @var{n}. This program
prints a new random number for each input record.
@example
@group
awk '
# Function to roll a simulated die.
function roll(n) @{ return 1 + int(rand() * n) @}
@end group
@group
# Roll 3 six-sided dice and
# print total number of points.
@{
printf("%d points\n",
roll(6)+roll(6)+roll(6))
@}'
@end group
@end example
@cindex seed for random numbers
@cindex random numbers, seed of
@comment MAWK uses a different seed each time.
@strong{Caution:} In most @code{awk} implementations, including @code{gawk},
@code{rand} starts generating numbers from the same
starting number, or @dfn{seed}, each time you run @code{awk}. Thus,
a program will generate the same results each time you run it.
The numbers are random within one @code{awk} run, but predictable
from run to run. This is convenient for debugging, but if you want
a program to do different things each time it is used, you must change
the seed to a value that will be different in each run. To do this,
use @code{srand}.
@item srand(@r{[}@var{x}@r{]})
@findex srand
The function @code{srand} sets the starting point, or seed,
for generating random numbers to the value @var{x}.
Each seed value leads to a particular sequence of random
numbers.@footnote{Computer generated random numbers really are not truly
random. They are technically known as ``pseudo-random.'' This means
that while the numbers in a sequence appear to be random, you can in
fact generate the same sequence of random numbers over and over again.}
Thus, if you set the seed to the same value a second time, you will get
the same sequence of random numbers again.
If you omit the argument @var{x}, as in @code{srand()}, then the current
date and time of day are used for a seed. This is the way to get random
numbers that are truly unpredictable.
The return value of @code{srand} is the previous seed. This makes it
easy to keep track of the seeds for use in consistently reproducing
sequences of random numbers.
@end table
@node String Functions, I/O Functions, Numeric Functions, Built-in
@section Built-in Functions for String Manipulation
The functions in this section look at or change the text of one or more
strings.
Optional parameters are enclosed in square brackets (``['' and ``]'').
@table @code
@item index(@var{in}, @var{find})
@findex index
This searches the string @var{in} for the first occurrence of the string
@var{find}, and returns the position in characters where that occurrence
begins in the string @var{in}. For example:
@example
$ awk 'BEGIN @{ print index("peanut", "an") @}'
@print{} 3
@end example
@noindent
If @var{find} is not found, @code{index} returns zero.
(Remember that string indices in @code{awk} start at one.)
@item length(@r{[}@var{string}@r{]})
@findex length
This gives you the number of characters in @var{string}. If
@var{string} is a number, the length of the digit string representing
that number is returned. For example, @code{length("abcde")} is five. By
contrast, @code{length(15 * 35)} works out to three. How? Well, 15 * 35 =
525, and 525 is then converted to the string @code{"525"}, which has
three characters.
If no argument is supplied, @code{length} returns the length of @code{$0}.
@cindex historical features
@cindex portability issues
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
In older versions of @code{awk}, you could call the @code{length} function
without any parentheses. Doing so is marked as ``deprecated'' in the
POSIX standard. This means that while you can do this in your
programs, it is a feature that can eventually be removed from a future
version of the standard. Therefore, for maximal portability of your
@code{awk} programs, you should always supply the parentheses.
@item match(@var{string}, @var{regexp})
@findex match
The @code{match} function searches the string, @var{string}, for the
longest, leftmost substring matched by the regular expression,
@var{regexp}. It returns the character position, or @dfn{index}, of
where that substring begins (one, if it starts at the beginning of
@var{string}). If no match is found, it returns zero.
@vindex RSTART
@vindex RLENGTH
The @code{match} function sets the built-in variable @code{RSTART} to
the index. It also sets the built-in variable @code{RLENGTH} to the
length in characters of the matched substring. If no match is found,
@code{RSTART} is set to zero, and @code{RLENGTH} to @minus{}1.
For example:
@example
@group
@c file eg/misc/findpat.sh
awk '@{
if ($1 == "FIND")
regex = $2
else @{
where = match($0, regex)
if (where != 0)
print "Match of", regex, "found at", \
where, "in", $0
@}
@}'
@c endfile
@end group
@end example
@noindent
This program looks for lines that match the regular expression stored in
the variable @code{regex}. This regular expression can be changed. If the
first word on a line is @samp{FIND}, @code{regex} is changed to be the
second word on that line. Therefore, given:
@example
@c file eg/misc/findpat.data
FIND ru+n
My program runs
but not very quickly
FIND Melvin
JF+KM
This line is property of Reality Engineering Co.
Melvin was here.
@c endfile
@end example
@noindent
@code{awk} prints:
@example
Match of ru+n found at 12 in My program runs
Match of Melvin found at 1 in Melvin was here.
@end example
@item split(@var{string}, @var{array} @r{[}, @var{fieldsep}@r{]})
@findex split
This divides @var{string} into pieces separated by @var{fieldsep},
and stores the pieces in @var{array}. The first piece is stored in
@code{@var{array}[1]}, the second piece in @code{@var{array}[2]}, and so
forth. The string value of the third argument, @var{fieldsep}, is
a regexp describing where to split @var{string} (much as @code{FS} can
be a regexp describing where to split input records). If
the @var{fieldsep} is omitted, the value of @code{FS} is used.
@code{split} returns the number of elements created.
The @code{split} function splits strings into pieces in a
manner similar to the way input lines are split into fields. For example:
@example
split("cul-de-sac", a, "-")
@end example
@noindent
splits the string @samp{cul-de-sac} into three fields using @samp{-} as the
separator. It sets the contents of the array @code{a} as follows:
@example
a[1] = "cul"
a[2] = "de"
a[3] = "sac"
@end example
@noindent
The value returned by this call to @code{split} is three.
As with input field-splitting, when the value of @var{fieldsep} is
@w{@code{" "}}, leading and trailing whitespace is ignored, and the elements
are separated by runs of whitespace.
@cindex differences between @code{gawk} and @code{awk}
Also as with input field-splitting, if @var{fieldsep} is the null string, each
individual character in the string is split into its own array element.
(This is a @code{gawk}-specific extension.)
@cindex dark corner
Recent implementations of @code{awk}, including @code{gawk}, allow
the third argument to be a regexp constant (@code{/abc/}), as well as a
string (d.c.). The POSIX standard allows this as well.
Before splitting the string, @code{split} deletes any previously existing
elements in the array @var{array} (d.c.).
@item sprintf(@var{format}, @var{expression1},@dots{})
@findex sprintf
This returns (without printing) the string that @code{printf} would
have printed out with the same arguments
(@pxref{Printf, ,Using @code{printf} Statements for Fancier Printing}).
For example:
@example
sprintf("pi = %.2f (approx.)", 22/7)
@end example
@noindent
returns the string @w{@code{"pi = 3.14 (approx.)"}}.
@ignore
2e: For sub, gsub, and gensub, either here or in the "how much matches"
section, we need some explanation that it is possible to match the
null string when using closures like *. E.g.,
$ echo abc | awk '{ gsub(/m*/, "X"); print }'
@print{} XaXbXcX
Although this makes a certain amount of sense, it can be very
suprising.
@end ignore
@item sub(@var{regexp}, @var{replacement} @r{[}, @var{target}@r{]})
@findex sub
The @code{sub} function alters the value of @var{target}.
It searches this value, which is treated as a string, for the
leftmost longest substring matched by the regular expression, @var{regexp},
extending this match as far as possible. Then the entire string is
changed by replacing the matched text with @var{replacement}.
The modified string becomes the new value of @var{target}.
This function is peculiar because @var{target} is not simply
used to compute a value, and not just any expression will do: it
must be a variable, field or array element, so that @code{sub} can
store a modified value there. If this argument is omitted, then the
default is to use and alter @code{$0}.
For example:
@example
str = "water, water, everywhere"
sub(/at/, "ith", str)
@end example
@noindent
sets @code{str} to @w{@code{"wither, water, everywhere"}}, by replacing the
leftmost, longest occurrence of @samp{at} with @samp{ith}.
The @code{sub} function returns the number of substitutions made (either
one or zero).
If the special character @samp{&} appears in @var{replacement}, it
stands for the precise substring that was matched by @var{regexp}. (If
the regexp can match more than one string, then this precise substring
may vary.) For example:
@example
awk '@{ sub(/candidate/, "& and his wife"); print @}'
@end example
@noindent
changes the first occurrence of @samp{candidate} to @samp{candidate
and his wife} on each input line.
Here is another example:
@example
awk 'BEGIN @{
str = "daabaaa"
sub(/a*/, "c&c", str)
print str
@}'
@print{} dcaacbaaa
@end example
@noindent
This shows how @samp{&} can represent a non-constant string, and also
illustrates the ``leftmost, longest'' rule in regexp matching
(@pxref{Leftmost Longest, ,How Much Text Matches?}).
The effect of this special character (@samp{&}) can be turned off by putting a
backslash before it in the string. As usual, to insert one backslash in
the string, you must write two backslashes. Therefore, write @samp{\\&}
in a string constant to include a literal @samp{&} in the replacement.
For example, here is how to replace the first @samp{|} on each line with
an @samp{&}:
@example
awk '@{ sub(/\|/, "\\&"); print @}'
@end example
@cindex @code{sub}, third argument of
@cindex @code{gsub}, third argument of
@strong{Note:} As mentioned above, the third argument to @code{sub} must
be a variable, field or array reference.
Some versions of @code{awk} allow the third argument to
be an expression which is not an lvalue. In such a case, @code{sub}
would still search for the pattern and return zero or one, but the result of
the substitution (if any) would be thrown away because there is no place
to put it. Such versions of @code{awk} accept expressions like
this:
@example
sub(/USA/, "United States", "the USA and Canada")
@end example
@noindent
For historical compatibility, @code{gawk} will accept erroneous code,
such as in the above example. However, using any other non-changeable
object as the third parameter will cause a fatal error, and your program
will not run.
@item gsub(@var{regexp}, @var{replacement} @r{[}, @var{target}@r{]})
@findex gsub
This is similar to the @code{sub} function, except @code{gsub} replaces
@emph{all} of the longest, leftmost, @emph{non-overlapping} matching
substrings it can find. The @samp{g} in @code{gsub} stands for
``global,'' which means replace everywhere. For example:
@example
awk '@{ gsub(/Britain/, "United Kingdom"); print @}'
@end example
@noindent
replaces all occurrences of the string @samp{Britain} with @samp{United
Kingdom} for all input records.
The @code{gsub} function returns the number of substitutions made. If
the variable to be searched and altered, @var{target}, is
omitted, then the entire input record, @code{$0}, is used.
As in @code{sub}, the characters @samp{&} and @samp{\} are special,
and the third argument must be an lvalue.
@end table
@table @code
@item gensub(@var{regexp}, @var{replacement}, @var{how} @r{[}, @var{target}@r{]})
@findex gensub
@code{gensub} is a general substitution function. Like @code{sub} and
@code{gsub}, it searches the target string @var{target} for matches of
the regular expression @var{regexp}. Unlike @code{sub} and
@code{gsub}, the modified string is returned as the result of the
function, and the original target string is @emph{not} changed. If
@var{how} is a string beginning with @samp{g} or @samp{G}, then it
replaces all matches of @var{regexp} with @var{replacement}.
Otherwise, @var{how} is a number indicating which match of @var{regexp}
to replace. If no @var{target} is supplied, @code{$0} is used instead.
@code{gensub} provides an additional feature that is not available
in @code{sub} or @code{gsub}: the ability to specify components of
a regexp in the replacement text. This is done by using parentheses
in the regexp to mark the components, and then specifying @samp{\@var{n}}
in the replacement text, where @var{n} is a digit from one to nine.
For example:
@example
@group
$ gawk '
> BEGIN @{
> a = "abc def"
> b = gensub(/(.+) (.+)/, "\\2 \\1", "g", a)
> print b
> @}'
@print{} def abc
@end group
@end example
@noindent
As described above for @code{sub}, you must type two backslashes in order
to get one into the string.
In the replacement text, the sequence @samp{\0} represents the entire
matched text, as does the character @samp{&}.
This example shows how you can use the third argument to control
which match of the regexp should be changed.
@example
$ echo a b c a b c |
> gawk '@{ print gensub(/a/, "AA", 2) @}'
@print{} a b c AA b c
@end example
In this case, @code{$0} is used as the default target string.
@code{gensub} returns the new string as its result, which is
passed directly to @code{print} for printing.
If the @var{how} argument is a string that does not begin with @samp{g} or
@samp{G}, or if it is a number that is less than zero, only one
substitution is performed.
@cindex differences between @code{gawk} and @code{awk}
@code{gensub} is a @code{gawk} extension; it is not available
in compatibility mode (@pxref{Options, ,Command Line Options}).
@item substr(@var{string}, @var{start} @r{[}, @var{length}@r{]})
@findex substr
This returns a @var{length}-character-long substring of @var{string},
starting at character number @var{start}. The first character of a
string is character number one. For example,
@code{substr("washington", 5, 3)} returns @code{"ing"}.
If @var{length} is not present, this function returns the whole suffix of
@var{string} that begins at character number @var{start}. For example,
@code{substr("washington", 5)} returns @code{"ington"}. The whole
suffix is also returned
if @var{length} is greater than the number of characters remaining
in the string, counting from character number @var{start}.
@strong{Note:} The string returned by @code{substr} @emph{cannot} be
assigned to. Thus, it is a mistake to attempt to change a portion of
a string, like this:
@example
string = "abcdef"
# try to get "abCDEf", won't work
substr(string, 3, 3) = "CDE"
@end example
@noindent
or to use @code{substr} as the third agument of @code{sub} or @code{gsub}:
@example
gsub(/xyz/, "pdq", substr($0, 5, 20)) # WRONG
@end example
@cindex case conversion
@cindex conversion of case
@item tolower(@var{string})
@findex tolower
This returns a copy of @var{string}, with each upper-case character
in the string replaced with its corresponding lower-case character.
Non-alphabetic characters are left unchanged. For example,
@code{tolower("MiXeD cAsE 123")} returns @code{"mixed case 123"}.
@item toupper(@var{string})
@findex toupper
This returns a copy of @var{string}, with each lower-case character
in the string replaced with its corresponding upper-case character.
Non-alphabetic characters are left unchanged. For example,
@code{toupper("MiXeD cAsE 123")} returns @code{"MIXED CASE 123"}.
@end table
@c fakenode --- for prepinfo
@subheading More About @samp{\} and @samp{&} with @code{sub}, @code{gsub} and @code{gensub}
@cindex escape processing, @code{sub} et. al.
When using @code{sub}, @code{gsub} or @code{gensub}, and trying to get literal
backslashes and ampersands into the replacement text, you need to remember
that there are several levels of @dfn{escape processing} going on.
First, there is the @dfn{lexical} level, which is when @code{awk} reads
your program, and builds an internal copy of your program that can
be executed.
Then there is the run-time level, when @code{awk} actually scans the
replacement string to determine what to generate.
At both levels, @code{awk} looks for a defined set of characters that
can come after a backslash. At the lexical level, it looks for the
escape sequences listed in @ref{Escape Sequences}.
Thus, for every @samp{\} that @code{awk} will process at the run-time
level, you type two @samp{\}s at the lexical level.
When a character that is not valid for an escape sequence follows the
@samp{\}, Unix @code{awk} and @code{gawk} both simply remove the initial
@samp{\}, and put the following character into the string. Thus, for
example, @code{"a\qb"} is treated as @code{"aqb"}.
At the run-time level, the various functions handle sequences of
@samp{\} and @samp{&} differently. The situation is (sadly) somewhat complex.
Historically, the @code{sub} and @code{gsub} functions treated the two
character sequence @samp{\&} specially; this sequence was replaced in
the generated text with a single @samp{&}. Any other @samp{\} within
the @var{replacement} string that did not precede an @samp{&} was passed
through unchanged. To illustrate with a table:
@c Thank to Karl Berry for help with the TeX stuff.
@tex
\vbox{\bigskip
% This table has lots of &'s and \'s, so unspecialize them.
\catcode`\& = \other \catcode`\\ = \other
% But then we need character for escape and tab.
@catcode`! = 4
@halign{@hfil#!@qquad@hfil#!@qquad#@hfil@cr
You type!@code{sub} sees!@code{sub} generates@cr
@hrulefill!@hrulefill!@hrulefill@cr
@code{\&}! @code{&}!the matched text@cr
@code{\\&}! @code{\&}!a literal @samp{&}@cr
@code{\\\&}! @code{\&}!a literal @samp{&}@cr
@code{\\\\&}! @code{\\&}!a literal @samp{\&}@cr
@code{\\\\\&}! @code{\\&}!a literal @samp{\&}@cr
@code{\\\\\\&}! @code{\\\&}!a literal @samp{\\&}@cr
@code{\\q}! @code{\q}!a literal @samp{\q}@cr
}
@bigskip}
@end tex
@ifinfo
@display
You type @code{sub} sees @code{sub} generates
-------- ---------- ---------------
@code{\&} @code{&} the matched text
@code{\\&} @code{\&} a literal @samp{&}
@code{\\\&} @code{\&} a literal @samp{&}
@code{\\\\&} @code{\\&} a literal @samp{\&}
@code{\\\\\&} @code{\\&} a literal @samp{\&}
@code{\\\\\\&} @code{\\\&} a literal @samp{\\&}
@code{\\q} @code{\q} a literal @samp{\q}
@end display
@end ifinfo
@noindent
This table shows both the lexical level processing, where
an odd number of backslashes becomes an even number at the run time level,
and the run-time processing done by @code{sub}.
(For the sake of simplicity, the rest of the tables below only show the
case of even numbers of @samp{\}s entered at the lexical level.)
The problem with the historical approach is that there is no way to get
a literal @samp{\} followed by the matched text.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
The 1992 POSIX standard attempted to fix this problem. The standard
says that @code{sub} and @code{gsub} look for either a @samp{\} or an @samp{&}
after the @samp{\}. If either one follows a @samp{\}, that character is
output literally. The interpretation of @samp{\} and @samp{&} then becomes
like this:
@c thanks to Karl Berry for formatting this table
@tex
\vbox{\bigskip
% This table has lots of &'s and \'s, so unspecialize them.
\catcode`\& = \other \catcode`\\ = \other
% But then we need character for escape and tab.
@catcode`! = 4
@halign{@hfil#!@qquad@hfil#!@qquad#@hfil@cr
You type!@code{sub} sees!@code{sub} generates@cr
@hrulefill!@hrulefill!@hrulefill@cr
@code{&}! @code{&}!the matched text@cr
@code{\\&}! @code{\&}!a literal @samp{&}@cr
@code{\\\\&}! @code{\\&}!a literal @samp{\}, then the matched text@cr
@code{\\\\\\&}! @code{\\\&}!a literal @samp{\&}@cr
}
@bigskip}
@end tex
@ifinfo
@display
You type @code{sub} sees @code{sub} generates
-------- ---------- ---------------
@code{&} @code{&} the matched text
@code{\\&} @code{\&} a literal @samp{&}
@code{\\\\&} @code{\\&} a literal @samp{\}, then the matched text
@code{\\\\\\&} @code{\\\&} a literal @samp{\&}
@end display
@end ifinfo
@noindent
This would appear to solve the problem.
Unfortunately, the phrasing of the standard is unusual. It
says, in effect, that @samp{\} turns off the special meaning of any
following character, but that for anything other than @samp{\} and @samp{&},
such special meaning is undefined. This wording leads to two problems.
@enumerate
@item
Backslashes must now be doubled in the @var{replacement} string, breaking
historical @code{awk} programs.
@item
To make sure that an @code{awk} program is portable, @emph{every} character
in the @var{replacement} string must be preceded with a
backslash.@footnote{This consequence was certainly unintended.}
@c I can say that, 'cause I was involved in making this change
@end enumerate
The POSIX standard is under revision.@footnote{As of @value{UPDATE-MONTH},
with final approval and publication hopefully sometime in 1997.}
Because of the above problems, proposed text for the revised standard
reverts to rules that correspond more closely to the original existing
practice. The proposed rules have special cases that make it possible
to produce a @samp{\} preceding the matched text.
@tex
\vbox{\bigskip
% This table has lots of &'s and \'s, so unspecialize them.
\catcode`\& = \other \catcode`\\ = \other
% But then we need character for escape and tab.
@catcode`! = 4
@halign{@hfil#!@qquad@hfil#!@qquad#@hfil@cr
You type!@code{sub} sees!@code{sub} generates@cr
@hrulefill!@hrulefill!@hrulefill@cr
@code{\\\\\\&}! @code{\\\&}!a literal @samp{\&}@cr
@code{\\\\&}! @code{\\&}!a literal @samp{\}, followed by the matched text@cr
@code{\\&}! @code{\&}!a literal @samp{&}@cr
@code{\\q}! @code{\q}!a literal @samp{\q}@cr
}
@bigskip}
@end tex
@ifinfo
@display
You type @code{sub} sees @code{sub} generates
-------- ---------- ---------------
@code{\\\\\\&} @code{\\\&} a literal @samp{\&}
@code{\\\\&} @code{\\&} a literal @samp{\}, followed by the matched text
@code{\\&} @code{\&} a literal @samp{&}
@code{\\q} @code{\q} a literal @samp{\q}
@end display
@end ifinfo
In a nutshell, at the run-time level, there are now three special sequences
of characters, @samp{\\\&}, @samp{\\&} and @samp{\&}, whereas historically,
there was only one. However, as in the historical case, any @samp{\} that
is not part of one of these three sequences is not special, and appears
in the output literally.
@code{gawk} 3.0 follows these proposed POSIX rules for @code{sub} and
@code{gsub}.
@c As much as we think it's a lousy idea. You win some, you lose some. Sigh.
Whether these proposed rules will actually become codified into the
standard is unknown at this point. Subsequent @code{gawk} releases will
track the standard and implement whatever the final version specifies;
this @value{DOCUMENT} will be updated as well.
The rules for @code{gensub} are considerably simpler. At the run-time
level, whenever @code{gawk} sees a @samp{\}, if the following character
is a digit, then the text that matched the corresponding parenthesized
subexpression is placed in the generated output. Otherwise,
no matter what the character after the @samp{\} is, that character will
appear in the generated text, and the @samp{\} will not.
@tex
\vbox{\bigskip
% This table has lots of &'s and \'s, so unspecialize them.
\catcode`\& = \other \catcode`\\ = \other
% But then we need character for escape and tab.
@catcode`! = 4
@halign{@hfil#!@qquad@hfil#!@qquad#@hfil@cr
You type!@code{gensub} sees!@code{gensub} generates@cr
@hrulefill!@hrulefill!@hrulefill@cr
@code{&}! @code{&}!the matched text@cr
@code{\\&}! @code{\&}!a literal @samp{&}@cr
@code{\\\\}! @code{\\}!a literal @samp{\}@cr
@code{\\\\&}! @code{\\&}!a literal @samp{\}, then the matched text@cr
@code{\\\\\\&}! @code{\\\&}!a literal @samp{\&}@cr
@code{\\q}! @code{\q}!a literal @samp{q}@cr
}
@bigskip}
@end tex
@ifinfo
@display
You type @code{gensub} sees @code{gensub} generates
-------- ------------- ------------------
@code{&} @code{&} the matched text
@code{\\&} @code{\&} a literal @samp{&}
@code{\\\\} @code{\\} a literal @samp{\}
@code{\\\\&} @code{\\&} a literal @samp{\}, then the matched text
@code{\\\\\\&} @code{\\\&} a literal @samp{\&}
@code{\\q} @code{\q} a literal @samp{q}
@end display
@end ifinfo
Because of the complexity of the lexical and run-time level processing,
and the special cases for @code{sub} and @code{gsub},
we recommend the use of @code{gawk} and @code{gensub} for when you have
to do substitutions.
@node I/O Functions, Time Functions, String Functions, Built-in
@section Built-in Functions for Input/Output
The following functions are related to Input/Output (I/O).
Optional parameters are enclosed in square brackets (``['' and ``]'').
@table @code
@item close(@var{filename})
@findex close
Close the file @var{filename}, for input or output. The argument may
alternatively be a shell command that was used for redirecting to or
from a pipe; then the pipe is closed.
@xref{Close Files And Pipes, ,Closing Input and Output Files and Pipes},
for more information.
@item fflush(@r{[}@var{filename}@r{]})
@findex fflush
@cindex portability issues
@cindex flushing buffers
@cindex buffers, flushing
@cindex buffering output
@cindex output, buffering
Flush any buffered output associated @var{filename}, which is either a
file opened for writing, or a shell command for redirecting output to
a pipe.
Many utility programs will @dfn{buffer} their output; they save information
to be written to a disk file or terminal in memory, until there is enough
for it to be worthwhile to send the data to the ouput device.
This is often more efficient than writing
every little bit of information as soon as it is ready. However, sometimes
it is necessary to force a program to @dfn{flush} its buffers; that is,
write the information to its destination, even if a buffer is not full.
This is the purpose of the @code{fflush} function; @code{gawk} too
buffers its output, and the @code{fflush} function can be used to force
@code{gawk} to flush its buffers.
@code{fflush} is a recent (1994) addition to the Bell Labs research
version of @code{awk}; it is not part of the POSIX standard, and will
not be available if @samp{--posix} has been specified on the command
line (@pxref{Options, ,Command Line Options}).
@code{gawk} extends the @code{fflush} function in two ways. The first
is to allow no argument at all. In this case, the buffer for the
standard output is flushed. The second way is to allow the null string
(@w{@code{""}}) as the argument. In this case, the buffers for
@emph{all} open output files and pipes are flushed.
@code{fflush} returns zero if the buffer was successfully flushed,
and nonzero otherwise.
@item system(@var{command})
@findex system
@cindex interaction, @code{awk} and other programs
The system function allows the user to execute operating system commands
and then return to the @code{awk} program. The @code{system} function
executes the command given by the string @var{command}. It returns, as
its value, the status returned by the command that was executed.
For example, if the following fragment of code is put in your @code{awk}
program:
@example
END @{
system("date | mail -s 'awk run done' root")
@}
@end example
@noindent
the system administrator will be sent mail when the @code{awk} program
finishes processing input and begins its end-of-input processing.
Note that redirecting @code{print} or @code{printf} into a pipe is often
enough to accomplish your task. However, if your @code{awk}
program is interactive, @code{system} is useful for cranking up large
self-contained programs, such as a shell or an editor.
Some operating systems cannot implement the @code{system} function.
@code{system} causes a fatal error if it is not supported.
@end table
@c fakenode --- for prepinfo
@subheading Interactive vs. Non-Interactive Buffering
@cindex buffering, interactive vs. non-interactive
@cindex buffering, non-interactive vs. interactive
@cindex interactive buffering vs. non-interactive
@cindex non-interactive buffering vs. interactive
As a side point, buffering issues can be even more confusing depending
upon whether or not your program is @dfn{interactive}, i.e., communicating
with a user sitting at a keyboard.@footnote{A program is interactive
if the standard output is connected
to a terminal device.}
Interactive programs generally @dfn{line buffer} their output; they
write out every line. Non-interactive programs wait until they have
a full buffer, which may be many lines of output.
@c Thanks to Walter.Mecky@dresdnerbank.de for this example, and for
@c motivating me to write this section.
Here is an example of the difference.
@example
$ awk '@{ print $1 + $2 @}'
1 1
@print{} 2
2 3
@print{} 5
@kbd{Control-d}
@end example
@noindent
Each line of output is printed immediately. Compare that behavior
with this example.
@example
$ awk '@{ print $1 + $2 @}' | cat
1 1
2 3
@kbd{Control-d}
@print{} 2
@print{} 5
@end example
@noindent
Here, no output is printed until after the @kbd{Control-d} is typed, since
it is all buffered, and sent down the pipe to @code{cat} in one shot.
@c fakenode --- for prepinfo
@subheading Controlling Output Buffering with @code{system}
@cindex flushing buffers
@cindex buffers, flushing
@cindex buffering output
@cindex output, buffering
The @code{fflush} function provides explicit control over output buffering for
individual files and pipes. However, its use is not portable to many other
@code{awk} implementations. An alternative method to flush output
buffers is by calling @code{system} with a null string as its argument:
@example
system("") # flush output
@end example
@noindent
@code{gawk} treats this use of the @code{system} function as a special
case, and is smart enough not to run a shell (or other command
interpreter) with the empty command. Therefore, with @code{gawk}, this
idiom is not only useful, it is efficient. While this method should work
with other @code{awk} implementations, it will not necessarily avoid
starting an unnecessary shell. (Other implementations may only
flush the buffer associated with the standard output, and not necessarily
all buffered output.)
If you think about what a programmer expects, it makes sense that
@code{system} should flush any pending output. The following program:
@example
BEGIN @{
print "first print"
system("echo system echo")
print "second print"
@}
@end example
@noindent
must print
@example
first print
system echo
second print
@end example
@noindent
and not
@example
system echo
first print
second print
@end example
If @code{awk} did not flush its buffers before calling @code{system}, the
latter (undesirable) output is what you would see.
@node Time Functions, , I/O Functions, Built-in
@section Functions for Dealing with Time Stamps
@cindex timestamps
@cindex time of day
A common use for @code{awk} programs is the processing of log files
containing time stamp information, indicating when a
particular log record was written. Many programs log their time stamp
in the form returned by the @code{time} system call, which is the
number of seconds since a particular epoch. On POSIX systems,
it is the number of seconds since Midnight, January 1, 1970, UTC.
In order to make it easier to process such log files, and to produce
useful reports, @code{gawk} provides two functions for working with time
stamps. Both of these are @code{gawk} extensions; they are not specified
in the POSIX standard, nor are they in any other known version
of @code{awk}.
Optional parameters are enclosed in square brackets (``['' and ``]'').
@table @code
@item systime()
@findex systime
This function returns the current time as the number of seconds since
the system epoch. On POSIX systems, this is the number of seconds
since Midnight, January 1, 1970, UTC. It may be a different number on
other systems.
@item strftime(@r{[}@var{format} @r{[}, @var{timestamp}@r{]]})
@findex strftime
This function returns a string. It is similar to the function of the
same name in ANSI C. The time specified by @var{timestamp} is used to
produce a string, based on the contents of the @var{format} string.
The @var{timestamp} is in the same format as the value returned by the
@code{systime} function. If no @var{timestamp} argument is supplied,
@code{gawk} will use the current time of day as the time stamp.
If no @var{format} argument is supplied, @code{strftime} uses
@code{@w{"%a %b %d %H:%M:%S %Z %Y"}}. This format string produces
output (almost) equivalent to that of the @code{date} utility.
(Versions of @code{gawk} prior to 3.0 require the @var{format} argument.)
@end table
The @code{systime} function allows you to compare a time stamp from a
log file with the current time of day. In particular, it is easy to
determine how long ago a particular record was logged. It also allows
you to produce log records using the ``seconds since the epoch'' format.
The @code{strftime} function allows you to easily turn a time stamp
into human-readable information. It is similar in nature to the @code{sprintf}
function
(@pxref{String Functions, ,Built-in Functions for String Manipulation}),
in that it copies non-format specification characters verbatim to the
returned string, while substituting date and time values for format
specifications in the @var{format} string.
@code{strftime} is guaranteed by the ANSI C standard to support
the following date format specifications:
@table @code
@item %a
The locale's abbreviated weekday name.
@item %A
The locale's full weekday name.
@item %b
The locale's abbreviated month name.
@item %B
The locale's full month name.
@item %c
The locale's ``appropriate'' date and time representation.
@item %d
The day of the month as a decimal number (01--31).
@item %H
The hour (24-hour clock) as a decimal number (00--23).
@item %I
The hour (12-hour clock) as a decimal number (01--12).
@item %j
The day of the year as a decimal number (001--366).
@item %m
The month as a decimal number (01--12).
@item %M
The minute as a decimal number (00--59).
@item %p
The locale's equivalent of the AM/PM designations associated
with a 12-hour clock.
@item %S
The second as a decimal number (00--60).@footnote{Occasionally there are
minutes in a year with a leap second, which is why the
seconds can go up to 60.}
@item %U
The week number of the year (the first Sunday as the first day of week one)
as a decimal number (00--53).
@item %w
The weekday as a decimal number (0--6). Sunday is day zero.
@item %W
The week number of the year (the first Monday as the first day of week one)
as a decimal number (00--53).
@item %x
The locale's ``appropriate'' date representation.
@item %X
The locale's ``appropriate'' time representation.
@item %y
The year without century as a decimal number (00--99).
@item %Y
The year with century as a decimal number (e.g., 1995).
@item %Z
The time zone name or abbreviation, or no characters if
no time zone is determinable.
@item %%
A literal @samp{%}.
@end table
If a conversion specifier is not one of the above, the behavior is
undefined.@footnote{This is because ANSI C leaves the
behavior of the C version of @code{strftime} undefined, and @code{gawk}
will use the system's version of @code{strftime} if it's there.
Typically, the conversion specifier will either not appear in the
returned string, or it will appear literally.}
@cindex locale, definition of
Informally, a @dfn{locale} is the geographic place in which a program
is meant to run. For example, a common way to abbreviate the date
September 4, 1991 in the United States would be ``9/4/91''.
In many countries in Europe, however, it would be abbreviated ``4.9.91''.
Thus, the @samp{%x} specification in a @code{"US"} locale might produce
@samp{9/4/91}, while in a @code{"EUROPE"} locale, it might produce
@samp{4.9.91}. The ANSI C standard defines a default @code{"C"}
locale, which is an environment that is typical of what most C programmers
are used to.
A public-domain C version of @code{strftime} is supplied with @code{gawk}
for systems that are not yet fully ANSI-compliant. If that version is
used to compile @code{gawk} (@pxref{Installation, ,Installing @code{gawk}}),
then the following additional format specifications are available:
@table @code
@item %D
Equivalent to specifying @samp{%m/%d/%y}.
@item %e
The day of the month, padded with a space if it is only one digit.
@item %h
Equivalent to @samp{%b}, above.
@item %n
A newline character (ASCII LF).
@item %r
Equivalent to specifying @samp{%I:%M:%S %p}.
@item %R
Equivalent to specifying @samp{%H:%M}.
@item %T
Equivalent to specifying @samp{%H:%M:%S}.
@item %t
A tab character.
@item %k
The hour (24-hour clock) as a decimal number (0-23).
Single digit numbers are padded with a space.
@item %l
The hour (12-hour clock) as a decimal number (1-12).
Single digit numbers are padded with a space.
@item %C
The century, as a number between 00 and 99.
@item %u
The weekday as a decimal number
[1 (Monday)--7].
@cindex ISO 8601
@item %V
The week number of the year (the first Monday as the first
day of week one) as a decimal number (01--53).
The method for determining the week number is as specified by ISO 8601
(to wit: if the week containing January 1 has four or more days in the
new year, then it is week one, otherwise it is week 53 of the previous year
and the next week is week one).
@item %G
The year with century of the ISO week number, as a decimal number.
For example, January 1, 1993, is in week 53 of 1992. Thus, the year
of its ISO week number is 1992, even though its year is 1993.
Similarly, December 31, 1973, is in week 1 of 1974. Thus, the year
of its ISO week number is 1974, even though its year is 1973.
@item %g
The year without century of the ISO week number, as a decimal number (00--99).
@item %Ec %EC %Ex %Ey %EY %Od %Oe %OH %OI
@itemx %Om %OM %OS %Ou %OU %OV %Ow %OW %Oy
These are ``alternate representations'' for the specifications
that use only the second letter (@samp{%c}, @samp{%C}, and so on).
They are recognized, but their normal representations are
used.@footnote{If you don't understand any of this, don't worry about
it; these facilities are meant to make it easier to ``internationalize''
programs.}
(These facilitate compliance with the POSIX @code{date} utility.)
@item %v
The date in VMS format (e.g., 20-JUN-1991).
@cindex RFC-822
@cindex RFC-1036
@item %z
The timezone offset in a +HHMM format (e.g., the format necessary to
produce RFC-822/RFC-1036 date headers).
@end table
This example is an @code{awk} implementation of the POSIX
@code{date} utility. Normally, the @code{date} utility prints the
current date and time of day in a well known format. However, if you
provide an argument to it that begins with a @samp{+}, @code{date}
will copy non-format specifier characters to the standard output, and
will interpret the current time according to the format specifiers in
the string. For example:
@example
$ date '+Today is %A, %B %d, %Y.'
@print{} Today is Thursday, July 11, 1991.
@end example
Here is the @code{gawk} version of the @code{date} utility.
It has a shell ``wrapper'', to handle the @samp{-u} option,
which requires that @code{date} run as if the time zone
was set to UTC.
@example
@group
#! /bin/sh
#
# date --- approximate the P1003.2 'date' command
case $1 in
-u) TZ=GMT0 # use UTC
export TZ
shift ;;
esac
@end group
@group
gawk 'BEGIN @{
format = "%a %b %d %H:%M:%S %Z %Y"
exitval = 0
@end group
@group
if (ARGC > 2)
exitval = 1
else if (ARGC == 2) @{
format = ARGV[1]
if (format ~ /^\+/)
format = substr(format, 2) # remove leading +
@}
print strftime(format)
exit exitval
@}' "$@@"
@end group
@end example
@node User-defined, Invoking Gawk, Built-in, Top
@chapter User-defined Functions
@cindex user-defined functions
@cindex functions, user-defined
Complicated @code{awk} programs can often be simplified by defining
your own functions. User-defined functions can be called just like
built-in ones (@pxref{Function Calls}), but it is up to you to define
them---to tell @code{awk} what they should do.
@menu
* Definition Syntax:: How to write definitions and what they mean.
* Function Example:: An example function definition and what it
does.
* Function Caveats:: Things to watch out for.
* Return Statement:: Specifying the value a function returns.
@end menu
@node Definition Syntax, Function Example, User-defined, User-defined
@section Function Definition Syntax
@cindex defining functions
@cindex function definition
Definitions of functions can appear anywhere between the rules of an
@code{awk} program. Thus, the general form of an @code{awk} program is
extended to include sequences of rules @emph{and} user-defined function
definitions.
There is no need in @code{awk} to put the definition of a function
before all uses of the function. This is because @code{awk} reads the
entire program before starting to execute any of it.
The definition of a function named @var{name} looks like this:
@example
function @var{name}(@var{parameter-list})
@{
@var{body-of-function}
@}
@end example
@cindex names, use of
@cindex namespaces
@noindent
@var{name} is the name of the function to be defined. A valid function
name is like a valid variable name: a sequence of letters, digits and
underscores, not starting with a digit.
Within a single @code{awk} program, any particular name can only be
used as a variable, array or function.
@var{parameter-list} is a list of the function's arguments and local
variable names, separated by commas. When the function is called,
the argument names are used to hold the argument values given in
the call. The local variables are initialized to the empty string.
A function cannot have two parameters with the same name.
The @var{body-of-function} consists of @code{awk} statements. It is the
most important part of the definition, because it says what the function
should actually @emph{do}. The argument names exist to give the body a
way to talk about the arguments; local variables, to give the body
places to keep temporary values.
Argument names are not distinguished syntactically from local variable
names; instead, the number of arguments supplied when the function is
called determines how many argument variables there are. Thus, if three
argument values are given, the first three names in @var{parameter-list}
are arguments, and the rest are local variables.
It follows that if the number of arguments is not the same in all calls
to the function, some of the names in @var{parameter-list} may be
arguments on some occasions and local variables on others. Another
way to think of this is that omitted arguments default to the
null string.
Usually when you write a function you know how many names you intend to
use for arguments and how many you intend to use as local variables. It is
conventional to place some extra space between the arguments and
the local variables, to document how your function is supposed to be used.
@cindex variable shadowing
During execution of the function body, the arguments and local variable
values hide or @dfn{shadow} any variables of the same names used in the
rest of the program. The shadowed variables are not accessible in the
function definition, because there is no way to name them while their
names have been taken away for the local variables. All other variables
used in the @code{awk} program can be referenced or set normally in the
function's body.
The arguments and local variables last only as long as the function body
is executing. Once the body finishes, you can once again access the
variables that were shadowed while the function was running.
@cindex recursive function
@cindex function, recursive
The function body can contain expressions which call functions. They
can even call this function, either directly or by way of another
function. When this happens, we say the function is @dfn{recursive}.
@cindex @code{awk} language, POSIX version
@cindex POSIX @code{awk}
In many @code{awk} implementations, including @code{gawk},
the keyword @code{function} may be
abbreviated @code{func}. However, POSIX only specifies the use of
the keyword @code{function}. This actually has some practical implications.
If @code{gawk} is in POSIX-compatibility mode
(@pxref{Options, ,Command Line Options}), then the following
statement will @emph{not} define a function:
@example
func foo() @{ a = sqrt($1) ; print a @}
@end example
@noindent
Instead it defines a rule that, for each record, concatenates the value
of the variable @samp{func} with the return value of the function @samp{foo}.
If the resulting string is non-null, the action is executed.
This is probably not what was desired. (@code{awk} accepts this input as
syntactically valid, since functions may be used before they are defined
in @code{awk} programs.)
@cindex portability issues
To ensure that your @code{awk} programs are portable, always use the
keyword @code{function} when defining a function.
@node Function Example, Function Caveats, Definition Syntax, User-defined
@section Function Definition Examples
Here is an example of a user-defined function, called @code{myprint}, that
takes a number and prints it in a specific format.
@example
function myprint(num)
@{
printf "%6.3g\n", num
@}
@end example
@noindent
To illustrate, here is an @code{awk} rule which uses our @code{myprint}
function:
@example
$3 > 0 @{ myprint($3) @}
@end example
@noindent
This program prints, in our special format, all the third fields that
contain a positive number in our input. Therefore, when given:
@example
@group
1.2 3.4 5.6 7.8
9.10 11.12 -13.14 15.16
17.18 19.20 21.22 23.24
@end group
@end example
@noindent
this program, using our function to format the results, prints:
@example
5.6
21.2
@end example
This function deletes all the elements in an array.
@example
function delarray(a, i)
@{
for (i in a)
delete a[i]
@}
@end example
When working with arrays, it is often necessary to delete all the elements
in an array and start over with a new list of elements
(@pxref{Delete, ,The @code{delete} Statement}).
Instead of having
to repeat this loop everywhere in your program that you need to clear out
an array, your program can just call @code{delarray}.
Here is an example of a recursive function. It takes a string
as an input parameter, and returns the string in backwards order.
@example
function rev(str, start)
@{
if (start == 0)
return ""
return (substr(str, start, 1) rev(str, start - 1))
@}
@end example
If this function is in a file named @file{rev.awk}, we can test it
this way:
@example
$ echo "Don't Panic!" |
> gawk --source '@{ print rev($0, length($0)) @}' -f rev.awk
@print{} !cinaP t'noD
@end example
Here is an example that uses the built-in function @code{strftime}.
(@xref{Time Functions, ,Functions for Dealing with Time Stamps},
for more information on @code{strftime}.)
The C @code{ctime} function takes a timestamp and returns it in a string,
formatted in a well known fashion. Here is an @code{awk} version:
@example
@c file eg/lib/ctime.awk
@group
# ctime.awk
#
# awk version of C ctime(3) function
function ctime(ts, format)
@{
format = "%a %b %d %H:%M:%S %Z %Y"
if (ts == 0)
ts = systime() # use current time as default
return strftime(format, ts)
@}
@c endfile
@end group
@end example
@node Function Caveats, Return Statement, Function Example, User-defined
@section Calling User-defined Functions
@cindex call by value
@cindex call by reference
@cindex calling a function
@cindex function call
@dfn{Calling a function} means causing the function to run and do its job.
A function call is an expression, and its value is the value returned by
the function.
A function call consists of the function name followed by the arguments
in parentheses. What you write in the call for the arguments are
@code{awk} expressions; each time the call is executed, these
expressions are evaluated, and the values are the actual arguments. For
example, here is a call to @code{foo} with three arguments (the first
being a string concatenation):
@example
foo(x y, "lose", 4 * z)
@end example
@strong{Caution:} whitespace characters (spaces and tabs) are not allowed
between the function name and the open-parenthesis of the argument list.
If you write whitespace by mistake, @code{awk} might think that you mean
to concatenate a variable with an expression in parentheses. However, it
notices that you used a function name and not a variable name, and reports
an error.
@cindex call by value
When a function is called, it is given a @emph{copy} of the values of
its arguments. This is known as @dfn{call by value}. The caller may use
a variable as the expression for the argument, but the called function
does not know this: it only knows what value the argument had. For
example, if you write this code:
@example
foo = "bar"
z = myfunc(foo)
@end example
@noindent
then you should not think of the argument to @code{myfunc} as being
``the variable @code{foo}.'' Instead, think of the argument as the
string value, @code{"bar"}.
If the function @code{myfunc} alters the values of its local variables,
this has no effect on any other variables. Thus, if @code{myfunc}
does this:
@example
@group
function myfunc(str)
@{
print str
str = "zzz"
print str
@}
@end group
@end example
@noindent
to change its first argument variable @code{str}, this @emph{does not}
change the value of @code{foo} in the caller. The role of @code{foo} in
calling @code{myfunc} ended when its value, @code{"bar"}, was computed.
If @code{str} also exists outside of @code{myfunc}, the function body
cannot alter this outer value, because it is shadowed during the
execution of @code{myfunc} and cannot be seen or changed from there.
@cindex call by reference
However, when arrays are the parameters to functions, they are @emph{not}
copied. Instead, the array itself is made available for direct manipulation
by the function. This is usually called @dfn{call by reference}.
Changes made to an array parameter inside the body of a function @emph{are}
visible outside that function.
@ifinfo
This can be @strong{very} dangerous if you do not watch what you are
doing. For example:
@end ifinfo
@iftex
@emph{This can be very dangerous if you do not watch what you are
doing.} For example:
@end iftex
@example
function changeit(array, ind, nvalue)
@{
array[ind] = nvalue
@}
BEGIN @{
a[1] = 1; a[2] = 2; a[3] = 3
changeit(a, 2, "two")
printf "a[1] = %s, a[2] = %s, a[3] = %s\n",
a[1], a[2], a[3]
@}
@end example
@noindent
This program prints @samp{a[1] = 1, a[2] = two, a[3] = 3}, because
@code{changeit} stores @code{"two"} in the second element of @code{a}.
@cindex undefined functions
@cindex functions, undefined
Some @code{awk} implementations allow you to call a function that
has not been defined, and only report a problem at run-time when the
program actually tries to call the function. For example:
@example
@group
BEGIN @{
if (0)
foo()
else
bar()
@}
function bar() @{ @dots{} @}
# note that `foo' is not defined
@end group
@end example
@noindent
Since the @samp{if} statement will never be true, it is not really a
problem that @code{foo} has not been defined. Usually though, it is a
problem if a program calls an undefined function.
@ignore
At one point, I had gawk dieing on this, but later decided that this might
break old programs and/or test suites.
@end ignore
If @samp{--lint} has been specified
(@pxref{Options, ,Command Line Options}),
@code{gawk} will report about calls to undefined functions.
Some @code{awk} implementations generate a run-time
error if you use the @code{next} statement
(@pxref{Next Statement, , The @code{next} Statement})
inside a user-defined function.
@code{gawk} does not have this problem.
@node Return Statement, , Function Caveats, User-defined
@section The @code{return} Statement
@cindex @code{return} statement
The body of a user-defined function can contain a @code{return} statement.
This statement returns control to the rest of the @code{awk} program. It
can also be used to return a value for use in the rest of the @code{awk}
program. It looks like this:
@example
return @r{[}@var{expression}@r{]}
@end example
The @var{expression} part is optional. If it is omitted, then the returned
value is undefined and, therefore, unpredictable.
A @code{return} statement with no value expression is assumed at the end of
every function definition. So if control reaches the end of the function
body, then the function returns an unpredictable value. @code{awk}
will @emph{not} warn you if you use the return value of such a function.
Sometimes, you want to write a function for what it does, not for
what it returns. Such a function corresponds to a @code{void} function
in C or to a @code{procedure} in Pascal. Thus, it may be appropriate to not
return any value; you should simply bear in mind that if you use the return
value of such a function, you do so at your own risk.
Here is an example of a user-defined function that returns a value
for the largest number among the elements of an array:
@example
@group
function maxelt(vec, i, ret)
@{
for (i in vec) @{
if (ret == "" || vec[i] > ret)
ret = vec[i]
@}
return ret
@}
@end group
@end example
@noindent
You call @code{maxelt} with one argument, which is an array name. The local
variables @code{i} and @code{ret} are not intended to be arguments;
while there is nothing to stop you from passing two or three arguments
to @code{maxelt}, the results would be strange. The extra space before
@code{i} in the function parameter list indicates that @code{i} and
@code{ret} are not supposed to be arguments. This is a convention that
you should follow when you define functions.
Here is a program that uses our @code{maxelt} function. It loads an
array, calls @code{maxelt}, and then reports the maximum number in that
array:
@example
@group
awk '
function maxelt(vec, i, ret)
@{
for (i in vec) @{
if (ret == "" || vec[i] > ret)
ret = vec[i]
@}
return ret
@}
@end group
@group
# Load all fields of each record into nums.
@{
for(i = 1; i <= NF; i++)
nums[NR, i] = $i
@}
END @{
print maxelt(nums)
@}'
@end group
@end example
Given the following input:
@example
@group
1 5 23 8 16
44 3 5 2 8 26
256 291 1396 2962 100
-6 467 998 1101
99385 11 0 225
@end group
@end example
@noindent
our program tells us (predictably) that @code{99385} is the largest number
in our array.
@node Invoking Gawk, Library Functions, User-defined, Top
@chapter Running @code{awk}
@cindex command line
@cindex invocation of @code{gawk}
@cindex arguments, command line
@cindex options, command line
@cindex long options
@cindex options, long
There are two ways to run @code{awk}: with an explicit program, or with
one or more program files. Here are templates for both of them; items
enclosed in @samp{@r{[}@dots{}@r{]}} in these templates are optional.
Besides traditional one-letter POSIX-style options, @code{gawk} also
supports GNU long options.
@example
awk @r{[@var{options}]} -f progfile @r{[@code{--}]} @var{file} @dots{}
awk @r{[@var{options}]} @r{[@code{--}]} '@var{program}' @var{file} @dots{}
@end example
@cindex empty program
@cindex dark corner
It is possible to invoke @code{awk} with an empty program:
@example
$ awk '' datafile1 datafile2
@end example
@noindent
Doing so makes little sense though; @code{awk} will simply exit
silently when given an empty program (d.c.). If @samp{--lint} has
been specified on the command line, @code{gawk} will issue a
warning that the program is empty.
@menu
* Options:: Command line options and their meanings.
* Other Arguments:: Input file names and variable assignments.
* AWKPATH Variable:: Searching directories for @code{awk} programs.
* Obsolete:: Obsolete Options and/or features.
* Undocumented:: Undocumented Options and Features.
* Known Bugs:: Known Bugs in @code{gawk}.
@end menu
@node Options, Other Arguments, Invoking Gawk, Invoking Gawk
@section Command Line Options
Options begin with a dash, and consist of a single character.
GNU style long options consist of two dashes and a keyword.
The keyword can be abbreviated, as long the abbreviation allows the option
to be uniquely identified. If the option takes an argument, then the
keyword is either immediately followed by an equals sign (@samp{=}) and the
argument's value, or the keyword and the argument's value are separated
by whitespace. For brevity, the discussion below only refers to the
traditional short options; however the long and short options are
interchangeable in all contexts.
Each long option for @code{gawk} has a corresponding
POSIX-style option. The options and their meanings are as follows:
@table @code
@item -F @var{fs}
@itemx --field-separator @var{fs}
@cindex @code{-F} option
@cindex @code{--field-separator} option
Sets the @code{FS} variable to @var{fs}
(@pxref{Field Separators, ,Specifying How Fields are Separated}).
@item -f @var{source-file}
@itemx --file @var{source-file}
@cindex @code{-f} option
@cindex @code{--file} option
Indicates that the @code{awk} program is to be found in @var{source-file}
instead of in the first non-option argument.
@item -v @var{var}=@var{val}
@itemx --assign @var{var}=@var{val}
@cindex @code{-v} option
@cindex @code{--assign} option
Sets the variable @var{var} to the value @var{val} @strong{before}
execution of the program begins. Such variable values are available
inside the @code{BEGIN} rule
(@pxref{Other Arguments, ,Other Command Line Arguments}).
The @samp{-v} option can only set one variable, but you can use
it more than once, setting another variable each time, like this:
@samp{awk @w{-v foo=1} @w{-v bar=2} @dots{}}.
@item -mf @var{NNN}
@itemx -mr @var{NNN}
Set various memory limits to the value @var{NNN}. The @samp{f} flag sets
the maximum number of fields, and the @samp{r} flag sets the maximum
record size. These two flags and the @samp{-m} option are from the
Bell Labs research version of Unix @code{awk}. They are provided
for compatibility, but otherwise ignored by
@code{gawk}, since @code{gawk} has no predefined limits.
@item -W @var{gawk-opt}
@cindex @code{-W} option
Following the POSIX standard, options that are implementation
specific are supplied as arguments to the @samp{-W} option. These options
also have corresponding GNU style long options.
See below.
@item --
Signals the end of the command line options. The following arguments
are not treated as options even if they begin with @samp{-}. This
interpretation of @samp{--} follows the POSIX argument parsing
conventions.
This is useful if you have file names that start with @samp{-},
or in shell scripts, if you have file names that will be specified
by the user which could start with @samp{-}.
@end table
The following @code{gawk}-specific options are available:
@table @code
@item -W traditional
@itemx -W compat
@itemx --traditional
@itemx --compat
@cindex @code{--compat} option
@cindex @code{--traditional} option
@cindex compatibility mode
Specifies @dfn{compatibility mode}, in which the GNU extensions to
the @code{awk} language are disabled, so that @code{gawk} behaves just
like the Bell Labs research version of Unix @code{awk}.
@samp{--traditional} is the preferred form of this option.
@xref{POSIX/GNU, ,Extensions in @code{gawk} Not in POSIX @code{awk}},
which summarizes the extensions. Also see
@ref{Compatibility Mode, ,Downward Compatibility and Debugging}.
@item -W copyleft
@itemx -W copyright
@itemx --copyleft
@itemx --copyright
@cindex @code{--copyleft} option
@cindex @code{--copyright} option
Print the short version of the General Public License, and then exit.
This option may disappear in a future version of @code{gawk}.
@item -W help
@itemx -W usage
@itemx --help
@itemx --usage
@cindex @code{--help} option
@cindex @code{--usage} option
Print a ``usage'' message summarizing the short and long style options
that @code{gawk} accepts, and then exit.
@item -W lint
@itemx --lint
@cindex @code{--lint} option
Warn about constructs that are dubious or non-portable to
other @code{awk} implementations.
Some warnings are issued when @code{gawk} first reads your program. Others
are issued at run-time, as your program executes.
@item -W lint-old
@itemx --lint-old
@cindex @code{--lint-old} option
Warn about constructs that are not available in
the original Version 7 Unix version of @code{awk}
(@pxref{V7/SVR3.1, , Major Changes between V7 and SVR3.1}).
@item -W posix
@itemx --posix
@cindex @code{--posix} option
@cindex POSIX mode
Operate in strict POSIX mode. This disables all @code{gawk}
extensions (just like @samp{--traditional}), and adds the following additional
restrictions:
@c IMPORTANT! Keep this list in sync with the one in node POSIX
@itemize @bullet
@item
@code{\x} escape sequences are not recognized
(@pxref{Escape Sequences}).
@item
Newlines do not act as whitespace to separate fields when @code{FS} is
equal to a single space.
@item
The synonym @code{func} for the keyword @code{function} is not
recognized (@pxref{Definition Syntax, ,Function Definition Syntax}).
@item
The operators @samp{**} and @samp{**=} cannot be used in
place of @samp{^} and @samp{^=} (@pxref{Arithmetic Ops, ,Arithmetic Operators},
and also @pxref{Assignment Ops, ,Assignment Expressions}).
@item
Specifying @samp{-Ft} on the command line does not set the value
of @code{FS} to be a single tab character
(@pxref{Field Separators, ,Specifying How Fields are Separated}).
@item
The @code{fflush} built-in function is not supported
(@pxref{I/O Functions, , Built-in Functions for Input/Output}).
@end itemize
If you supply both @samp{--traditional} and @samp{--posix} on the
command line, @samp{--posix} will take precedence. @code{gawk}
will also issue a warning if both options are supplied.
@item -W re-interval
@itemx --re-interval
Allow interval expressions
(@pxref{Regexp Operators, , Regular Expression Operators}),
in regexps.
Because interval expressions were traditionally not available in @code{awk},
@code{gawk} does not provide them by default. This prevents old @code{awk}
programs from breaking.
@item -W source @var{program-text}
@itemx --source @var{program-text}
@cindex @code{--source} option
Program source code is taken from the @var{program-text}. This option
allows you to mix source code in files with source
code that you enter on the command line. This is particularly useful
when you have library functions that you wish to use from your command line
programs (@pxref{AWKPATH Variable, ,The @code{AWKPATH} Environment Variable}).
@item -W version
@itemx --version
@cindex @code{--version} option
Prints version information for this particular copy of @code{gawk}.
This allows you to determine if your copy of @code{gawk} is up to date
with respect to whatever the Free Software Foundation is currently
distributing.
It is also useful for bug reports
(@pxref{Bugs, , Reporting Problems and Bugs}).
@end table
Any other options are flagged as invalid with a warning message, but
are otherwise ignored.
In compatibility mode, as a special case, if the value of @var{fs} supplied
to the @samp{-F} option is @samp{t}, then @code{FS} is set to the tab
character (@code{"\t"}). This is only true for @samp{--traditional}, and not
for @samp{--posix}
(@pxref{Field Separators, ,Specifying How Fields are Separated}).
The @samp{-f} option may be used more than once on the command line.
If it is, @code{awk} reads its program source from all of the named files, as
if they had been concatenated together into one big file. This is
useful for creating libraries of @code{awk} functions. Useful functions
can be written once, and then retrieved from a standard place, instead
of having to be included into each individual program.
You can type in a program at the terminal and still use library functions,
by specifying @samp{-f /dev/tty}. @code{awk} will read a file from the terminal
to use as part of the @code{awk} program. After typing your program,
type @kbd{Control-d} (the end-of-file character) to terminate it.
(You may also use @samp{-f -} to read program source from the standard
input, but then you will not be able to also use the standard input as a
source of data.)
Because it is clumsy using the standard @code{awk} mechanisms to mix source
file and command line @code{awk} programs, @code{gawk} provides the
@samp{--source} option. This does not require you to pre-empt the standard
input for your source code, and allows you to easily mix command line
and library source code
(@pxref{AWKPATH Variable, ,The @code{AWKPATH} Environment Variable}).
If no @samp{-f} or @samp{--source} option is specified, then @code{gawk}
will use the first non-option command line argument as the text of the
program source code.
@cindex @code{POSIXLY_CORRECT} environment variable
@cindex environment variable, @code{POSIXLY_CORRECT}
If the environment variable @code{POSIXLY_CORRECT} exists,
then @code{gawk} will behave in strict POSIX mode, exactly as if
you had supplied the @samp{--posix} command line option.
Many GNU programs look for this environment variable to turn on
strict POSIX mode. If you supply @samp{--lint} on the command line,
and @code{gawk} turns on POSIX mode because of @code{POSIXLY_CORRECT},
then it will print a warning message indicating that POSIX
mode is in effect.
You would typically set this variable in your shell's startup file.
For a Bourne compatible shell (such as Bash), you would add these
lines to the @file{.profile} file in your home directory.
@example
@group
POSIXLY_CORRECT=true
export POSIXLY_CORRECT
@end group
@end example
For a @code{csh} compatible shell,@footnote{Not recommended.}
you would add this line to the @file{.login} file in your home directory.
@example
setenv POSIXLY_CORRECT true
@end example
@node Other Arguments, AWKPATH Variable, Options, Invoking Gawk
@section Other Command Line Arguments
Any additional arguments on the command line are normally treated as
input files to be processed in the order specified. However, an
argument that has the form @code{@var{var}=@var{value}}, assigns
the value @var{value} to the variable @var{var}---it does not specify a
file at all.
@vindex ARGIND
@vindex ARGV
All these arguments are made available to your @code{awk} program in the
@code{ARGV} array (@pxref{Built-in Variables}). Command line options
and the program text (if present) are omitted from @code{ARGV}.
All other arguments, including variable assignments, are
included. As each element of @code{ARGV} is processed, @code{gawk}
sets the variable @code{ARGIND} to the index in @code{ARGV} of the
current element.
The distinction between file name arguments and variable-assignment
arguments is made when @code{awk} is about to open the next input file.
At that point in execution, it checks the ``file name'' to see whether
it is really a variable assignment; if so, @code{awk} sets the variable
instead of reading a file.
Therefore, the variables actually receive the given values after all
previously specified files have been read. In particular, the values of
variables assigned in this fashion are @emph{not} available inside a
@code{BEGIN} rule
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}),
since such rules are run before @code{awk} begins scanning the argument list.
@cindex dark corner
The variable values given on the command line are processed for escape
sequences (d.c.) (@pxref{Escape Sequences}).
In some earlier implementations of @code{awk}, when a variable assignment
occurred before any file names, the assignment would happen @emph{before}
the @code{BEGIN} rule was executed. @code{awk}'s behavior was thus
inconsistent; some command line assignments were available inside the
@code{BEGIN} rule, while others were not. However,
some applications came to depend
upon this ``feature.'' When @code{awk} was changed to be more consistent,
the @samp{-v} option was added to accommodate applications that depended
upon the old behavior.
The variable assignment feature is most useful for assigning to variables
such as @code{RS}, @code{OFS}, and @code{ORS}, which control input and
output formats, before scanning the data files. It is also useful for
controlling state if multiple passes are needed over a data file. For
example:
@cindex multiple passes over data
@cindex passes, multiple
@example
awk 'pass == 1 @{ @var{pass 1 stuff} @}
pass == 2 @{ @var{pass 2 stuff} @}' pass=1 mydata pass=2 mydata
@end example
Given the variable assignment feature, the @samp{-F} option for setting
the value of @code{FS} is not
strictly necessary. It remains for historical compatibility.
@node AWKPATH Variable, Obsolete, Other Arguments, Invoking Gawk
@section The @code{AWKPATH} Environment Variable
@cindex @code{AWKPATH} environment variable
@cindex environment variable, @code{AWKPATH}
@cindex search path
@cindex directory search
@cindex path, search
@cindex differences between @code{gawk} and @code{awk}
The previous section described how @code{awk} program files can be named
on the command line with the @samp{-f} option. In most @code{awk}
implementations, you must supply a precise path name for each program
file, unless the file is in the current directory.
@cindex search path, for source files
But in @code{gawk}, if the file name supplied to the @samp{-f} option
does not contain a @samp{/}, then @code{gawk} searches a list of
directories (called the @dfn{search path}), one by one, looking for a
file with the specified name.
The search path is a string consisting of directory names
separated by colons. @code{gawk} gets its search path from the
@code{AWKPATH} environment variable. If that variable does not exist,
@code{gawk} uses a default path, which is
@samp{.:/usr/local/share/awk}.@footnote{Your version of @code{gawk}
may use a directory that is different than @file{/usr/local/share/awk}; it
will depend upon how @code{gawk} was built and installed. The actual
directory will be the value of @samp{$(datadir)} generated when
@code{gawk} was configured. You probably don't need to worry about this
though.} (Programs written for use by
system administrators should use an @code{AWKPATH} variable that
does not include the current directory, @file{.}.)
The search path feature is particularly useful for building up libraries
of useful @code{awk} functions. The library files can be placed in a
standard directory that is in the default path, and then specified on
the command line with a short file name. Otherwise, the full file name
would have to be typed for each file.
By using both the @samp{--source} and @samp{-f} options, your command line
@code{awk} programs can use facilities in @code{awk} library files.
@xref{Library Functions, , A Library of @code{awk} Functions}.
Path searching is not done if @code{gawk} is in compatibility mode.
This is true for both @samp{--traditional} and @samp{--posix}.
@xref{Options, ,Command Line Options}.
@strong{Note:} if you want files in the current directory to be found,
you must include the current directory in the path, either by including
@file{.} explicitly in the path, or by writing a null entry in the
path. (A null entry is indicated by starting or ending the path with a
colon, or by placing two colons next to each other (@samp{::}).) If the
current directory is not included in the path, then files cannot be
found in the current directory. This path search mechanism is identical
to the shell's.
@c someday, @cite{The Bourne Again Shell}....
Starting with version 3.0, if @code{AWKPATH} is not defined in the
environment, @code{gawk} will place its default search path into
@code{ENVIRON["AWKPATH"]}. This makes it easy to determine
the actual search path @code{gawk} will use.
@node Obsolete, Undocumented, AWKPATH Variable, Invoking Gawk
@section Obsolete Options and/or Features
@cindex deprecated options
@cindex obsolete options
@cindex deprecated features
@cindex obsolete features
This section describes features and/or command line options from
previous releases of @code{gawk} that are either not available in the
current version, or that are still supported but deprecated (meaning that
they will @emph{not} be in the next release).
@c update this section for each release!
For version @value{VERSION}.@value{PATCHLEVEL} of @code{gawk}, there are no
command line options
or other deprecated features from the previous version of @code{gawk}.
@iftex
This section
@end iftex
@ifinfo
This node
@end ifinfo
is thus essentially a place holder,
in case some option becomes obsolete in a future version of @code{gawk}.
@ignore
@c This is pretty old news...
The public-domain version of @code{strftime} that is distributed with
@code{gawk} changed for the 2.14 release. The @samp{%V} conversion specifier
that used to generate the date in VMS format was changed to @samp{%v}.
This is because the POSIX standard for the @code{date} utility now
specifies a @samp{%V} conversion specifier.
@xref{Time Functions, ,Functions for Dealing with Time Stamps}, for details.
@end ignore
@node Undocumented, Known Bugs, Obsolete, Invoking Gawk
@section Undocumented Options and Features
@cindex undocumented features
@display
@i{Use the Source, Luke!}
Obi-Wan
@end display
@sp 1
This section intentionally left blank.
@c Read The Source, Luke!
@ignore
@c If these came out in the Info file or TeX document, then they wouldn't
@c be undocumented, would they?
@code{gawk} has one undocumented option:
@table @code
@item -W nostalgia
@itemx --nostalgia
Print the message @code{"awk: bailing out near line 1"} and dump core.
This option was inspired by the common behavior of very early versions of
Unix @code{awk}, and by a t--shirt.
@end table
Early versions of @code{awk} used to not require any separator (either
a newline or @samp{;}) between the rules in @code{awk} programs. Thus,
it was common to see one-line programs like:
@example
awk '@{ sum += $1 @} END @{ print sum @}'
@end example
@code{gawk} actually supports this, but it is purposely undocumented
since it is considered bad style. The correct way to write such a program
is either
@example
awk '@{ sum += $1 @} ; END @{ print sum @}'
@end example
@noindent
or
@example
awk '@{ sum += $1 @}
END @{ print sum @}' data
@end example
@noindent
@xref{Statements/Lines, ,@code{awk} Statements Versus Lines}, for a fuller
explanation.
@end ignore
@node Known Bugs, , Undocumented, Invoking Gawk
@section Known Bugs in @code{gawk}
@cindex bugs, known in @code{gawk}
@cindex known bugs
@itemize @bullet
@item
The @samp{-F} option for changing the value of @code{FS}
(@pxref{Options, ,Command Line Options})
is not necessary given the command line variable
assignment feature; it remains only for backwards compatibility.
@item
If your system actually has support for @file{/dev/fd} and the
associated @file{/dev/stdin}, @file{/dev/stdout}, and
@file{/dev/stderr} files, you may get different output from @code{gawk}
than you would get on a system without those files. When @code{gawk}
interprets these files internally, it synchronizes output to the
standard output with output to @file{/dev/stdout}, while on a system
with those files, the output is actually to different open files
(@pxref{Special Files, ,Special File Names in @code{gawk}}).
@item
Syntactically invalid single character programs tend to overflow
the parse stack, generating a rather unhelpful message. Such programs
are surprisingly difficult to diagnose in the completely general case,
and the effort to do so really is not worth it.
@end itemize
@node Library Functions, Sample Programs, Invoking Gawk, Top
@chapter A Library of @code{awk} Functions
@c 2e: USE TEXINFO-2 FUNCTION DEFINITION STUFF!!!!!!!!!!!!!
This chapter presents a library of useful @code{awk} functions. The
sample programs presented later
(@pxref{Sample Programs, ,Practical @code{awk} Programs})
use these functions.
The functions are presented here in a progression from simple to complex.
@ref{Extract Program, ,Extracting Programs from Texinfo Source Files},
presents a program that you can use to extract the source code for
these example library functions and programs from the Texinfo source
for this @value{DOCUMENT}.
(This has already been done as part of the @code{gawk} distribution.)
If you have written one or more useful, general purpose @code{awk} functions,
and would like to contribute them for a subsequent edition of this @value{DOCUMENT},
please contact the author. @xref{Bugs, ,Reporting Problems and Bugs},
for information on doing this. Don't just send code, as you will be
required to either place your code in the public domain,
publish it under the GPL (@pxref{Copying, ,GNU GENERAL PUBLIC LICENSE}),
or assign the copyright in it to the Free Software Foundation.
@menu
* Portability Notes:: What to do if you don't have @code{gawk}.
* Nextfile Function:: Two implementations of a @code{nextfile}
function.
* Assert Function:: A function for assertions in @code{awk}
programs.
* Round Function:: A function for rounding if @code{sprintf} does
not do it correctly.
* Ordinal Functions:: Functions for using characters as numbers and
vice versa.
* Join Function:: A function to join an array into a string.
* Mktime Function:: A function to turn a date into a timestamp.
* Gettimeofday Function:: A function to get formatted times.
* Filetrans Function:: A function for handling data file transitions.
* Getopt Function:: A function for processing command line
arguments.
* Passwd Functions:: Functions for getting user information.
* Group Functions:: Functions for getting group information.
* Library Names:: How to best name private global variables in
library functions.
@end menu
@node Portability Notes, Nextfile Function, Library Functions, Library Functions
@section Simulating @code{gawk}-specific Features
@cindex portability issues
The programs in this chapter and in
@ref{Sample Programs, ,Practical @code{awk} Programs},
freely use features that are specific to @code{gawk}.
This section briefly discusses how you can rewrite these programs for
different implementations of @code{awk}.
Diagnostic error messages are sent to @file{/dev/stderr}.
Use @samp{| "cat 1>&2"} instead of @samp{> "/dev/stderr"}, if your system
does not have a @file{/dev/stderr}, or if you cannot use @code{gawk}.
A number of programs use @code{nextfile}
(@pxref{Nextfile Statement, ,The @code{nextfile} Statement}),
to skip any remaining input in the input file.
@ref{Nextfile Function, ,Implementing @code{nextfile} as a Function},
shows you how to write a function that will do the same thing.
Finally, some of the programs choose to ignore upper-case and lower-case
distinctions in their input. They do this by assigning one to @code{IGNORECASE}.
You can achieve the same effect by adding the following rule to the
beginning of the program:
@example
# ignore case
@{ $0 = tolower($0) @}
@end example
@noindent
Also, verify that all regexp and string constants used in
comparisons only use lower-case letters.
@node Nextfile Function, Assert Function, Portability Notes, Library Functions
@section Implementing @code{nextfile} as a Function
@cindex skipping input files
@cindex input files, skipping
The @code{nextfile} statement presented in
@ref{Nextfile Statement, ,The @code{nextfile} Statement},
is a @code{gawk}-specific extension. It is not available in other
implementations of @code{awk}. This section shows two versions of a
@code{nextfile} function that you can use to simulate @code{gawk}'s
@code{nextfile} statement if you cannot use @code{gawk}.
Here is a first attempt at writing a @code{nextfile} function.
@example
@group
# nextfile --- skip remaining records in current file
# this should be read in before the "main" awk program
function nextfile() @{ _abandon_ = FILENAME; next @}
_abandon_ == FILENAME @{ next @}
@end group
@end example
This file should be included before the main program, because it supplies
a rule that must be executed first. This rule compares the current data
file's name (which is always in the @code{FILENAME} variable) to a private
variable named @code{_abandon_}. If the file name matches, then the action
part of the rule executes a @code{next} statement, to go on to the next
record. (The use of @samp{_} in the variable name is a convention.
It is discussed more fully in
@ref{Library Names, , Naming Library Function Global Variables}.)
The use of the @code{next} statement effectively creates a loop that reads
all the records from the current data file.
Eventually, the end of the file is reached, and
a new data file is opened, changing the value of @code{FILENAME}.
Once this happens, the comparison of @code{_abandon_} to @code{FILENAME}
fails, and execution continues with the first rule of the ``real'' program.
The @code{nextfile} function itself simply sets the value of @code{_abandon_}
and then executes a @code{next} statement to start the loop
going.@footnote{Some implementations of @code{awk} do not allow you to
execute @code{next} from within a function body. Some other work-around
will be necessary if you use such a version.}
@c mawk is what we're talking about.
This initial version has a subtle problem. What happens if the same data
file is listed @emph{twice} on the command line, one right after the other,
or even with just a variable assignment between the two occurrences of
the file name?
@c @findex nextfile
@c do it this way, since all the indices are merged
@cindex @code{nextfile} function
In such a case,
this code will skip right through the file, a second time, even though
it should stop when it gets to the end of the first occurrence.
Here is a second version of @code{nextfile} that remedies this problem.
@example
@group
@c file eg/lib/nextfile.awk
# nextfile --- skip remaining records in current file
# correctly handle successive occurrences of the same file
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May, 1993
# this should be read in before the "main" awk program
function nextfile() @{ _abandon_ = FILENAME; next @}
_abandon_ == FILENAME @{
if (FNR == 1)
_abandon_ = ""
else
next
@}
@c endfile
@end group
@end example
The @code{nextfile} function has not changed. It sets @code{_abandon_}
equal to the current file name and then executes a @code{next} satement.
The @code{next} statement reads the next record and increments @code{FNR},
so @code{FNR} is guaranteed to have a value of at least two.
However, if @code{nextfile} is called for the last record in the file,
then @code{awk} will close the current data file and move on to the next
one. Upon doing so, @code{FILENAME} will be set to the name of the new file,
and @code{FNR} will be reset to one. If this next file is the same as
the previous one, @code{_abandon_} will still be equal to @code{FILENAME}.
However, @code{FNR} will be equal to one, telling us that this is a new
occurrence of the file, and not the one we were reading when the
@code{nextfile} function was executed. In that case, @code{_abandon_}
is reset to the empty string, so that further executions of this rule
will fail (until the next time that @code{nextfile} is called).
If @code{FNR} is not one, then we are still in the original data file,
and the program executes a @code{next} statement to skip through it.
An important question to ask at this point is: ``Given that the
functionality of @code{nextfile} can be provided with a library file,
why is it built into @code{gawk}?'' This is an important question. Adding
features for little reason leads to larger, slower programs that are
harder to maintain.
The answer is that building @code{nextfile} into @code{gawk} provides
significant gains in efficiency. If the @code{nextfile} function is executed
at the beginning of a large data file, @code{awk} still has to scan the entire
file, splitting it up into records, just to skip over it. The built-in
@code{nextfile} can simply close the file immediately and proceed to the
next one, saving a lot of time. This is particularly important in
@code{awk}, since @code{awk} programs are generally I/O bound (i.e.@:
they spend most of their time doing input and output, instead of performing
computations).
@node Assert Function, Round Function, Nextfile Function, Library Functions
@section Assertions
@cindex assertions
@cindex @code{assert}, C version
When writing large programs, it is often useful to be able to know
that a condition or set of conditions is true. Before proceeding with a
particular computation, you make a statement about what you believe to be
the case. Such a statement is known as an
``assertion.'' The C language provides an @code{<assert.h>} header file
and corresponding @code{assert} macro that the programmer can use to make
assertions. If an assertion fails, the @code{assert} macro arranges to
print a diagnostic message describing the condition that should have
been true but was not, and then it kills the program. In C, using
@code{assert} looks this:
@example
#include <assert.h>
int myfunc(int a, double b)
@{
assert(a <= 5 && b >= 17);
@dots{}
@}
@end example
If the assertion failed, the program would print a message similar to
this:
@example
prog.c:5: assertion failed: a <= 5 && b >= 17
@end example
@findex assert
The ANSI C language makes it possible to turn the condition into a string for use
in printing the diagnostic message. This is not possible in @code{awk}, so
this @code{assert} function also requires a string version of the condition
that is being tested.
@example
@c @group
@c file eg/lib/assert.awk
# assert --- assert that a condition is true. Otherwise exit.
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May, 1993
function assert(condition, string)
@{
if (! condition) @{
printf("%s:%d: assertion failed: %s\n",
FILENAME, FNR, string) > "/dev/stderr"
_assert_exit = 1
exit 1
@}
@}
END @{
if (_assert_exit)
exit 1
@}
@c endfile
@c @end group
@end example
The @code{assert} function tests the @code{condition} parameter. If it
is false, it prints a message to standard error, using the @code{string}
parameter to describe the failed condition. It then sets the variable
@code{_assert_exit} to one, and executes the @code{exit} statement.
The @code{exit} statement jumps to the @code{END} rule. If the @code{END}
rules finds @code{_assert_exit} to be true, then it exits immediately.
The purpose of the @code{END} rule with its test is to
keep any other @code{END} rules from running. When an assertion fails, the
program should exit immediately.
If no assertions fail, then @code{_assert_exit} will still be
false when the @code{END} rule is run normally, and the rest of the
program's @code{END} rules will execute.
For all of this to work correctly, @file{assert.awk} must be the
first source file read by @code{awk}.
You would use this function in your programs this way:
@example
function myfunc(a, b)
@{
assert(a <= 5 && b >= 17, "a <= 5 && b >= 17")
@dots{}
@}
@end example
@noindent
If the assertion failed, you would see a message like this:
@example
mydata:1357: assertion failed: a <= 5 && b >= 17
@end example
There is a problem with this version of @code{assert}, that it may not
be possible to work around. An @code{END} rule is automatically added
to the program calling @code{assert}. Normally, if a program consists
of just a @code{BEGIN} rule, the input files and/or standard input are
not read. However, now that the program has an @code{END} rule, @code{awk}
will attempt to read the input data files, or standard input
(@pxref{Using BEGIN/END, , Startup and Cleanup Actions}),
most likely causing the program to hang, waiting for input.
@node Round Function, Ordinal Functions, Assert Function, Library Functions
@section Rounding Numbers
@cindex rounding
The way @code{printf} and @code{sprintf}
(@pxref{Printf, , Using @code{printf} Statements for Fancier Printing})
do rounding will often depend
upon the system's C @code{sprintf} subroutine.
On many machines,
@code{sprintf} rounding is ``unbiased,'' which means it doesn't always
round a trailing @samp{.5} up, contrary to naive expectations. In unbiased
rounding, @samp{.5} rounds to even, rather than always up, so 1.5 rounds to
2 but 4.5 rounds to 4.
The result is that if you are using a format that does
rounding (e.g., @code{"%.0f"}) you should check what your system does.
The following function does traditional rounding;
it might be useful if your awk's @code{printf} does unbiased rounding.
@findex round
@example
@c file eg/lib/round.awk
# round --- do normal rounding
#
# Arnold Robbins, arnold@@gnu.ai.mit.edu, August, 1996
# Public Domain
function round(x, ival, aval, fraction)
@{
ival = int(x) # integer part, int() truncates
# see if fractional part
if (ival == x) # no fraction
return x
if (x < 0) @{
aval = -x # absolute value
ival = int(aval)
fraction = aval - ival
if (fraction >= .5)
return int(x) - 1 # -2.5 --> -3
else
return int(x) # -2.3 --> -2
@} else @{
fraction = x - ival
if (fraction >= .5)
return ival + 1
else
return ival
@}
@}
# test harness
@{ print $0, round($0) @}
@c endfile
@end example
@node Ordinal Functions, Join Function, Round Function, Library Functions
@section Translating Between Characters and Numbers
@cindex numeric character values
@cindex values of characters as numbers
One commercial implementation of @code{awk} supplies a built-in function,
@code{ord}, which takes a character and returns the numeric value for that
character in the machine's character set. If the string passed to
@code{ord} has more than one character, only the first one is used.
The inverse of this function is @code{chr} (from the function of the same
name in Pascal), which takes a number and returns the corresponding character.
Both functions can be written very nicely in @code{awk}; there is no real
reason to build them into the @code{awk} interpreter.
@findex ord
@findex chr
@example
@group
@c file eg/lib/ord.awk
# ord.awk --- do ord and chr
#
# Global identifiers:
# _ord_: numerical values indexed by characters
# _ord_init: function to initialize _ord_
#
# Arnold Robbins
# arnold@@gnu.ai.mit.edu
# Public Domain
# 16 January, 1992
# 20 July, 1992, revised
BEGIN @{ _ord_init() @}
@c endfile
@end group
@c @group
@c file eg/lib/ord.awk
function _ord_init( low, high, i, t)
@{
low = sprintf("%c", 7) # BEL is ascii 7
if (low == "\a") @{ # regular ascii
low = 0
high = 127
@} else if (sprintf("%c", 128 + 7) == "\a") @{
# ascii, mark parity
low = 128
high = 255
@} else @{ # ebcdic(!)
low = 0
high = 255
@}
for (i = low; i <= high; i++) @{
t = sprintf("%c", i)
_ord_[t] = i
@}
@}
@c endfile
@c @end group
@end example
@cindex character sets
@cindex character encodings
@cindex ASCII
@cindex EBCDIC
@cindex mark parity
Some explanation of the numbers used by @code{chr} is worthwhile.
The most prominent character set in use today is ASCII. Although an
eight-bit byte can hold 256 distinct values (from zero to 255), ASCII only
defines characters that use the values from zero to 127.@footnote{ASCII
has been extended in many countries to use the values from 128 to 255
for country-specific characters. If your system uses these extensions,
you can simplify @code{_ord_init} to simply loop from zero to 255.}
At least one computer manufacturer that we know of
@c Pr1me, blech
uses ASCII, but with mark parity, meaning that the leftmost bit in the byte
is always one. What this means is that on those systems, characters
have numeric values from 128 to 255.
Finally, large mainframe systems use the EBCDIC character set, which
uses all 256 values.
While there are other character sets in use on some older systems,
they are not really worth worrying about.
@example
@group
@c file eg/lib/ord.awk
function ord(str, c)
@{
# only first character is of interest
c = substr(str, 1, 1)
return _ord_[c]
@}
@c endfile
@end group
@group
@c file eg/lib/ord.awk
function chr(c)
@{
# force c to be numeric by adding 0
return sprintf("%c", c + 0)
@}
@c endfile
@end group
@c @group
@c file eg/lib/ord.awk
#### test code ####
# BEGIN \
# @{
# for (;;) @{
# printf("enter a character: ")
# if (getline var <= 0)
# break
# printf("ord(%s) = %d\n", var, ord(var))
# @}
# @}
@c endfile
@c @end group
@end example
An obvious improvement to these functions would be to move the code for the
@code{@w{_ord_init}} function into the body of the @code{BEGIN} rule. It was
written this way initially for ease of development.
There is a ``test program'' in a @code{BEGIN} rule, for testing the
function. It is commented out for production use.
@node Join Function, Mktime Function, Ordinal Functions, Library Functions
@section Merging an Array Into a String
@cindex merging strings
When doing string processing, it is often useful to be able to join
all the strings in an array into one long string. The following function,
@code{join}, accomplishes this task. It is used later in several of
the application programs
(@pxref{Sample Programs, ,Practical @code{awk} Programs}).
Good function design is important; this function needs to be general, but it
should also have a reasonable default behavior. It is called with an array
and the beginning and ending indices of the elements in the array to be
merged. This assumes that the array indices are numeric---a reasonable
assumption since the array was likely created with @code{split}
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
@findex join
@example
@group
@c file eg/lib/join.awk
# join.awk --- join an array into a string
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
function join(array, start, end, sep, result, i)
@{
if (sep == "")
sep = " "
else if (sep == SUBSEP) # magic value
sep = ""
result = array[start]
for (i = start + 1; i <= end; i++)
result = result sep array[i]
return result
@}
@c endfile
@end group
@end example
An optional additional argument is the separator to use when joining the
strings back together. If the caller supplies a non-empty value,
@code{join} uses it. If it is not supplied, it will have a null
value. In this case, @code{join} uses a single blank as a default
separator for the strings. If the value is equal to @code{SUBSEP},
then @code{join} joins the strings with no separator between them.
@code{SUBSEP} serves as a ``magic'' value to indicate that there should
be no separation between the component strings.
It would be nice if @code{awk} had an assignment operator for concatenation.
The lack of an explicit operator for concatenation makes string operations
more difficult than they really need to be.
@node Mktime Function, Gettimeofday Function, Join Function, Library Functions
@section Turning Dates Into Timestamps
The @code{systime} function built in to @code{gawk}
returns the current time of day as
a timestamp in ``seconds since the Epoch.'' This timestamp
can be converted into a printable date of almost infinitely variable
format using the built-in @code{strftime} function.
(For more information on @code{systime} and @code{strftime},
@pxref{Time Functions, ,Functions for Dealing with Time Stamps}.)
@cindex converting dates to timestamps
@cindex dates, converting to timestamps
@cindex timestamps, converting from dates
An interesting but difficult problem is to convert a readable representation
of a date back into a timestamp. The ANSI C library provides a @code{mktime}
function that does the basic job, converting a canonical representation of a
date into a timestamp.
It would appear at first glance that @code{gawk} would have to supply a
@code{mktime} built-in function that was simply a ``hook'' to the C language
version. In fact though, @code{mktime} can be implemented entirely in
@code{awk}.
Here is a version of @code{mktime} for @code{awk}. It takes a simple
representation of the date and time, and converts it into a timestamp.
The code is presented here intermixed with explanatory prose. In
@ref{Extract Program, ,Extracting Programs from Texinfo Source Files},
you will see how the Texinfo source file for this @value{DOCUMENT}
can be processed to extract the code into a single source file.
The program begins with a descriptive comment and a @code{BEGIN} rule
that initializes a table @code{_tm_months}. This table is a two-dimensional
array that has the lengths of the months. The first index is zero for
regular years, and one for leap years. The values are the same for all the
months in both kinds of years, except for February; thus the use of multiple
assignment.
@example
@c @group
@c file eg/lib/mktime.awk
# mktime.awk --- convert a canonical date representation
# into a timestamp
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
BEGIN \
@{
# Initialize table of month lengths
_tm_months[0,1] = _tm_months[1,1] = 31
_tm_months[0,2] = 28; _tm_months[1,2] = 29
_tm_months[0,3] = _tm_months[1,3] = 31
_tm_months[0,4] = _tm_months[1,4] = 30
_tm_months[0,5] = _tm_months[1,5] = 31
_tm_months[0,6] = _tm_months[1,6] = 30
_tm_months[0,7] = _tm_months[1,7] = 31
_tm_months[0,8] = _tm_months[1,8] = 31
_tm_months[0,9] = _tm_months[1,9] = 30
_tm_months[0,10] = _tm_months[1,10] = 31
_tm_months[0,11] = _tm_months[1,11] = 30
_tm_months[0,12] = _tm_months[1,12] = 31
@}
@c endfile
@c @end group
@end example
The benefit of merging multiple @code{BEGIN} rules
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns})
is particularly clear when writing library files. Functions in library
files can cleanly initialize their own private data and also provide clean-up
actions in private @code{END} rules.
The next function is a simple one that computes whether a given year is or
is not a leap year. If a year is evenly divisible by four, but not evenly
divisible by 100, or if it is evenly divisible by 400, then it is a leap
year. Thus, 1904 was a leap year, 1900 was not, but 2000 will be.
@c Change this after the year 2000 to ``2000 was'' (:-)
@findex _tm_isleap
@example
@group
@c file eg/lib/mktime.awk
# decide if a year is a leap year
function _tm_isleap(year, ret)
@{
ret = (year % 4 == 0 && year % 100 != 0) ||
(year % 400 == 0)
return ret
@}
@c endfile
@end group
@end example
This function is only used a few times in this file, and its computation
could have been written @dfn{in-line} (at the point where it's used).
Making it a separate function made the original development easier, and also
avoids the possibility of typing errors when duplicating the code in
multiple places.
The next function is more interesting. It does most of the work of
generating a timestamp, which is converting a date and time into some number
of seconds since the Epoch. The caller passes an array (rather
imaginatively named @code{a}) containing six
values: the year including century, the month as a number between one and 12,
the day of the month, the hour as a number between zero and 23, the minute in
the hour, and the seconds within the minute.
The function uses several local variables to precompute the number of
seconds in an hour, seconds in a day, and seconds in a year. Often,
similar C code simply writes out the expression in-line, expecting the
compiler to do @dfn{constant folding}. E.g., most C compilers would
turn @samp{60 * 60} into @samp{3600} at compile time, instead of recomputing
it every time at run time. Precomputing these values makes the
function more efficient.
@findex _tm_addup
@example
@c @group
@c file eg/lib/mktime.awk
# convert a date into seconds
function _tm_addup(a, total, yearsecs, daysecs,
hoursecs, i, j)
@{
hoursecs = 60 * 60
daysecs = 24 * hoursecs
yearsecs = 365 * daysecs
total = (a[1] - 1970) * yearsecs
@group
# extra day for leap years
for (i = 1970; i < a[1]; i++)
if (_tm_isleap(i))
total += daysecs
@end group
@group
j = _tm_isleap(a[1])
for (i = 1; i < a[2]; i++)
total += _tm_months[j, i] * daysecs
@end group
total += (a[3] - 1) * daysecs
total += a[4] * hoursecs
total += a[5] * 60
total += a[6]
return total
@}
@c endfile
@c @end group
@end example
The function starts with a first approximation of all the seconds between
Midnight, January 1, 1970,@footnote{This is the Epoch on POSIX systems.
It may be different on other systems.} and the beginning of the current
year. It then goes through all those years, and for every leap year,
adds an additional day's worth of seconds.
The variable @code{j} holds either one or zero, if the current year is or is not
a leap year.
For every month in the current year prior to the current month, it adds
the number of seconds in the month, using the appropriate entry in the
@code{_tm_months} array.
Finally, it adds in the seconds for the number of days prior to the current
day, and the number of hours, minutes, and seconds in the current day.
The result is a count of seconds since January 1, 1970. This value is not
yet what is needed though. The reason why is described shortly.
The main @code{mktime} function takes a single character string argument.
This string is a representation of a date and time in a ``canonical''
(fixed) form. This string should be
@code{"@var{year} @var{month} @var{day} @var{hour} @var{minute} @var{second}"}.
@findex mktime
@example
@c @group
@c file eg/lib/mktime.awk
# mktime --- convert a date into seconds,
# compensate for time zone
function mktime(str, res1, res2, a, b, i, j, t, diff)
@{
i = split(str, a, " ") # don't rely on FS
if (i != 6)
return -1
# force numeric
for (j in a)
a[j] += 0
@group
# validate
if (a[1] < 1970 ||
a[2] < 1 || a[2] > 12 ||
a[3] < 1 || a[3] > 31 ||
a[4] < 0 || a[4] > 23 ||
a[5] < 0 || a[5] > 59 ||
a[6] < 0 || a[6] > 60 )
return -1
@end group
res1 = _tm_addup(a)
t = strftime("%Y %m %d %H %M %S", res1)
if (_tm_debug)
printf("(%s) -> (%s)\n", str, t) > "/dev/stderr"
split(t, b, " ")
res2 = _tm_addup(b)
diff = res1 - res2
if (_tm_debug)
printf("diff = %d seconds\n", diff) > "/dev/stderr"
res1 += diff
return res1
@}
@c endfile
@c @end group
@end example
The function first splits the string into an array, using spaces and tabs as
separators. If there are not six elements in the array, it returns an
error, signaled as the value @minus{}1.
Next, it forces each element of the array to be numeric, by adding zero to it.
The following @samp{if} statement then makes sure that each element is
within an allowable range. (This checking could be extended further, e.g.,
to make sure that the day of the month is within the correct range for the
particular month supplied.) All of this is essentially preliminary set-up
and error checking.
Recall that @code{_tm_addup} generated a value in seconds since Midnight,
January 1, 1970. This value is not directly usable as the result we want,
@emph{since the calculation does not account for the local timezone}. In other
words, the value represents the count in seconds since the Epoch, but only
for UTC (Universal Coordinated Time). If the local timezone is east or west
of UTC, then some number of hours should be either added to, or subtracted from
the resulting timestamp.
For example, 6:23 p.m. in Atlanta, Georgia (USA), is normally five hours west
of (behind) UTC. It is only four hours behind UTC if daylight savings
time is in effect.
If you are calling @code{mktime} in Atlanta, with the argument
@code{@w{"1993 5 23 18 23 12"}}, the result from @code{_tm_addup} will be
for 6:23 p.m. UTC, which is only 2:23 p.m. in Atlanta. It is necessary to
add another four hours worth of seconds to the result.
How can @code{mktime} determine how far away it is from UTC? This is
surprisingly easy. The returned timestamp represents the time passed to
@code{mktime} @emph{as UTC}. This timestamp can be fed back to
@code{strftime}, which will format it as a @emph{local} time; i.e.@: as
if it already had the UTC difference added in to it. This is done by
giving @code{@w{"%Y %m %d %H %M %S"}} to @code{strftime} as the format
argument. It returns the computed timestamp in the original string
format. The result represents a time that accounts for the UTC
difference. When the new time is converted back to a timestamp, the
difference between the two timestamps is the difference (in seconds)
between the local timezone and UTC. This difference is then added back
to the original result. An example demonstrating this is presented below.
Finally, there is a ``main'' program for testing the function.
@example
@c @group
@c file eg/lib/mktime.awk
BEGIN @{
if (_tm_test) @{
printf "Enter date as yyyy mm dd hh mm ss: "
getline _tm_test_date
t = mktime(_tm_test_date)
r = strftime("%Y %m %d %H %M %S", t)
printf "Got back (%s)\n", r
@}
@}
@c endfile
@c @end group
@end example
The entire program uses two variables that can be set on the command
line to control debugging output and to enable the test in the final
@code{BEGIN} rule. Here is the result of a test run. (Note that debugging
output is to standard error, and test output is to standard output.)
@example
@c @group
$ gawk -f mktime.awk -v _tm_test=1 -v _tm_debug=1
@print{} Enter date as yyyy mm dd hh mm ss: 1993 5 23 15 35 10
@error{} (1993 5 23 15 35 10) -> (1993 05 23 11 35 10)
@error{} diff = 14400 seconds
@print{} Got back (1993 05 23 15 35 10)
@c @end group
@end example
The time entered was 3:35 p.m. (15:35 on a 24-hour clock), on May 23, 1993.
The first line
of debugging output shows the resulting time as UTC---four hours ahead of
the local time zone. The second line shows that the difference is 14400
seconds, which is four hours. (The difference is only four hours, since
daylight savings time is in effect during May.)
The final line of test output shows that the timezone compensation
algorithm works; the returned time is the same as the entered time.
This program does not solve the general problem of turning an arbitrary date
representation into a timestamp. That problem is very involved. However,
the @code{mktime} function provides a foundation upon which to build. Other
software can convert month names into numeric months, and AM/PM times into
24-hour clocks, to generate the ``canonical'' format that @code{mktime}
requires.
@node Gettimeofday Function, Filetrans Function, Mktime Function, Library Functions
@section Managing the Time of Day
@cindex formatted timestamps
@cindex timestamps, formatted
The @code{systime} and @code{strftime} functions described in
@ref{Time Functions, ,Functions for Dealing with Time Stamps},
provide the minimum functionality necessary for dealing with the time of day
in human readable form. While @code{strftime} is extensive, the control
formats are not necessarily easy to remember or intuitively obvious when
reading a program.
The following function, @code{gettimeofday}, populates a user-supplied array
with pre-formatted time information. It returns a string with the current
time formatted in the same way as the @code{date} utility.
@findex gettimeofday
@example
@c @group
@c file eg/lib/gettime.awk
# gettimeofday --- get the time of day in a usable format
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain, May 1993
#
# Returns a string in the format of output of date(1)
# Populates the array argument time with individual values:
# time["second"] -- seconds (0 - 59)
# time["minute"] -- minutes (0 - 59)
# time["hour"] -- hours (0 - 23)
# time["althour"] -- hours (0 - 12)
# time["monthday"] -- day of month (1 - 31)
# time["month"] -- month of year (1 - 12)
# time["monthname"] -- name of the month
# time["shortmonth"] -- short name of the month
# time["year"] -- year within century (0 - 99)
# time["fullyear"] -- year with century (19xx or 20xx)
# time["weekday"] -- day of week (Sunday = 0)
# time["altweekday"] -- day of week (Monday = 0)
# time["weeknum"] -- week number, Sunday first day
# time["altweeknum"] -- week number, Monday first day
# time["dayname"] -- name of weekday
# time["shortdayname"] -- short name of weekday
# time["yearday"] -- day of year (0 - 365)
# time["timezone"] -- abbreviation of timezone name
# time["ampm"] -- AM or PM designation
@group
function gettimeofday(time, ret, now, i)
@{
# get time once, avoids unnecessary system calls
now = systime()
# return date(1)-style output
ret = strftime("%a %b %d %H:%M:%S %Z %Y", now)
# clear out target array
for (i in time)
delete time[i]
@end group
@group
# fill in values, force numeric values to be
# numeric by adding 0
time["second"] = strftime("%S", now) + 0
time["minute"] = strftime("%M", now) + 0
time["hour"] = strftime("%H", now) + 0
time["althour"] = strftime("%I", now) + 0
time["monthday"] = strftime("%d", now) + 0
time["month"] = strftime("%m", now) + 0
time["monthname"] = strftime("%B", now)
time["shortmonth"] = strftime("%b", now)
time["year"] = strftime("%y", now) + 0
time["fullyear"] = strftime("%Y", now) + 0
time["weekday"] = strftime("%w", now) + 0
time["altweekday"] = strftime("%u", now) + 0
time["dayname"] = strftime("%A", now)
time["shortdayname"] = strftime("%a", now)
time["yearday"] = strftime("%j", now) + 0
time["timezone"] = strftime("%Z", now)
time["ampm"] = strftime("%p", now)
time["weeknum"] = strftime("%U", now) + 0
time["altweeknum"] = strftime("%W", now) + 0
return ret
@}
@end group
@c endfile
@end example
The string indices are easier to use and read than the various formats
required by @code{strftime}. The @code{alarm} program presented in
@ref{Alarm Program, ,An Alarm Clock Program},
uses this function.
@c exercise!!!
The @code{gettimeofday} function is presented above as it was written. A
more general design for this function would have allowed the user to supply
an optional timestamp value that would have been used instead of the current
time.
@node Filetrans Function, Getopt Function, Gettimeofday Function, Library Functions
@section Noting Data File Boundaries
@cindex per file initialization and clean-up
The @code{BEGIN} and @code{END} rules are each executed exactly once, at
the beginning and end respectively of your @code{awk} program
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}).
We (the @code{gawk} authors) once had a user who mistakenly thought that the
@code{BEGIN} rule was executed at the beginning of each data file and the
@code{END} rule was executed at the end of each data file. When informed
that this was not the case, the user requested that we add new special
patterns to @code{gawk}, named @code{BEGIN_FILE} and @code{END_FILE}, that
would have the desired behavior. He even supplied us the code to do so.
However, after a little thought, I came up with the following library program.
It arranges to call two user-supplied functions, @code{beginfile} and
@code{endfile}, at the beginning and end of each data file.
Besides solving the problem in only nine(!) lines of code, it does so
@emph{portably}; this will work with any implementation of @code{awk}.
@example
@c @group
# transfile.awk
#
# Give the user a hook for filename transitions
#
# The user must supply functions beginfile() and endfile()
# that each take the name of the file being started or
# finished, respectively.
#
# Arnold Robbins, arnold@@gnu.ai.mit.edu, January 1992
# Public Domain
FILENAME != _oldfilename \
@{
if (_oldfilename != "")
endfile(_oldfilename)
_oldfilename = FILENAME
beginfile(FILENAME)
@}
END @{ endfile(FILENAME) @}
@c @end group
@end example
This file must be loaded before the user's ``main'' program, so that the
rule it supplies will be executed first.
This rule relies on @code{awk}'s @code{FILENAME} variable that
automatically changes for each new data file. The current file name is
saved in a private variable, @code{_oldfilename}. If @code{FILENAME} does
not equal @code{_oldfilename}, then a new data file is being processed, and
it is necessary to call @code{endfile} for the old file. Since
@code{endfile} should only be called if a file has been processed, the
program first checks to make sure that @code{_oldfilename} is not the null
string. The program then assigns the current file name to
@code{_oldfilename}, and calls @code{beginfile} for the file.
Since, like all @code{awk} variables, @code{_oldfilename} will be
initialized to the null string, this rule executes correctly even for the
first data file.
The program also supplies an @code{END} rule, to do the final processing for
the last file. Since this @code{END} rule comes before any @code{END} rules
supplied in the ``main'' program, @code{endfile} will be called first. Once
again the value of multiple @code{BEGIN} and @code{END} rules should be clear.
@findex beginfile
@findex endfile
This version has same problem as the first version of @code{nextfile}
(@pxref{Nextfile Function, ,Implementing @code{nextfile} as a Function}).
If the same data file occurs twice in a row on command line, then
@code{endfile} and @code{beginfile} will not be executed at the end of the
first pass and at the beginning of the second pass.
This version solves the problem.
@example
@c @group
@c file eg/lib/ftrans.awk
# ftrans.awk --- handle data file transitions
#
# user supplies beginfile() and endfile() functions
#
# Arnold Robbins, arnold@@gnu.ai.mit.edu. November 1992
# Public Domain
FNR == 1 @{
if (_filename_ != "")
endfile(_filename_)
_filename_ = FILENAME
beginfile(FILENAME)
@}
END @{ endfile(_filename_) @}
@c endfile
@c @end group
@end example
In @ref{Wc Program, ,Counting Things},
you will see how this library function can be used, and
how it simplifies writing the main program.
@node Getopt Function, Passwd Functions, Filetrans Function, Library Functions
@section Processing Command Line Options
@cindex @code{getopt}, C version
@cindex processing arguments
@cindex argument processing
Most utilities on POSIX compatible systems take options or ``switches'' on
the command line that can be used to change the way a program behaves.
@code{awk} is an example of such a program
(@pxref{Options, ,Command Line Options}).
Often, options take @dfn{arguments}, data that the program needs to
correctly obey the command line option. For example, @code{awk}'s
@samp{-F} option requires a string to use as the field separator.
The first occurrence on the command line of either @samp{--} or a
string that does not begin with @samp{-} ends the options.
Most Unix systems provide a C function named @code{getopt} for processing
command line arguments. The programmer provides a string describing the one
letter options. If an option requires an argument, it is followed in the
string with a colon. @code{getopt} is also passed the
count and values of the command line arguments, and is called in a loop.
@code{getopt} processes the command line arguments for option letters.
Each time around the loop, it returns a single character representing the
next option letter that it found, or @samp{?} if it found an invalid option.
When it returns @minus{}1, there are no options left on the command line.
When using @code{getopt}, options that do not take arguments can be
grouped together. Furthermore, options that take arguments require that the
argument be present. The argument can immediately follow the option letter,
or it can be a separate command line argument.
Given a hypothetical program that takes
three command line options, @samp{-a}, @samp{-b}, and @samp{-c}, and
@samp{-b} requires an argument, all of the following are valid ways of
invoking the program:
@example
@c @group
prog -a -b foo -c data1 data2 data3
prog -ac -bfoo -- data1 data2 data3
prog -acbfoo data1 data2 data3
@c @end group
@end example
Notice that when the argument is grouped with its option, the rest of
the command line argument is considered to be the option's argument.
In the above example, @samp{-acbfoo} indicates that all of the
@samp{-a}, @samp{-b}, and @samp{-c} options were supplied,
and that @samp{foo} is the argument to the @samp{-b} option.
@code{getopt} provides four external variables that the programmer can use.
@table @code
@item optind
The index in the argument value array (@code{argv}) where the first
non-option command line argument can be found.
@item optarg
The string value of the argument to an option.
@item opterr
Usually @code{getopt} prints an error message when it finds an invalid
option. Setting @code{opterr} to zero disables this feature. (An
application might wish to print its own error message.)
@item optopt
The letter representing the command line option.
While not usually documented, most versions supply this variable.
@end table
The following C fragment shows how @code{getopt} might process command line
arguments for @code{awk}.
@example
@group
int
main(int argc, char *argv[])
@{
@dots{}
/* print our own message */
opterr = 0;
@end group
@group
while ((c = getopt(argc, argv, "v:f:F:W:")) != -1) @{
switch (c) @{
case 'f': /* file */
@dots{}
break;
case 'F': /* field separator */
@dots{}
break;
case 'v': /* variable assignment */
@dots{}
break;
case 'W': /* extension */
@dots{}
break;
case '?':
default:
usage();
break;
@}
@}
@dots{}
@}
@end group
@end example
As a side point, @code{gawk} actually uses the GNU @code{getopt_long}
function to process both normal and GNU-style long options
(@pxref{Options, ,Command Line Options}).
The abstraction provided by @code{getopt} is very useful, and would be quite
handy in @code{awk} programs as well. Here is an @code{awk} version of
@code{getopt}. This function highlights one of the greatest weaknesses in
@code{awk}, which is that it is very poor at manipulating single characters.
Repeated calls to @code{substr} are necessary for accessing individual
characters (@pxref{String Functions, ,Built-in Functions for String Manipulation}).
The discussion walks through the code a bit at a time.
@example
@c @group
@c file eg/lib/getopt.awk
# getopt --- do C library getopt(3) function in awk
#
# arnold@@gnu.ai.mit.edu
# Public domain
#
# Initial version: March, 1991
# Revised: May, 1993
@group
# External variables:
# Optind -- index of ARGV for first non-option argument
# Optarg -- string value of argument to current option
# Opterr -- if non-zero, print our own diagnostic
# Optopt -- current option letter
@end group
# Returns
# -1 at end of options
# ? for unrecognized option
# <c> a character representing the current option
# Private Data
# _opti index in multi-flag option, e.g., -abc
@c endfile
@c @end group
@end example
The function starts out with some documentation: who wrote the code,
and when it was revised, followed by a list of the global variables it uses,
what the return values are and what they mean, and any global variables that
are ``private'' to this library function. Such documentation is essential
for any program, and particularly for library functions.
@findex getopt
@example
@c @group
@c file eg/lib/getopt.awk
function getopt(argc, argv, options, optl, thisopt, i)
@{
optl = length(options)
if (optl == 0) # no options given
return -1
if (argv[Optind] == "--") @{ # all done
Optind++
_opti = 0
return -1
@} else if (argv[Optind] !~ /^-[^: \t\n\f\r\v\b]/) @{
_opti = 0
return -1
@}
@c endfile
@c @end group
@end example
The function first checks that it was indeed called with a string of options
(the @code{options} parameter). If @code{options} has a zero length,
@code{getopt} immediately returns @minus{}1.
The next thing to check for is the end of the options. A @samp{--} ends the
command line options, as does any command line argument that does not begin
with a @samp{-}. @code{Optind} is used to step through the array of command
line arguments; it retains its value across calls to @code{getopt}, since it
is a global variable.
The regexp used, @code{@w{/^-[^: \t\n\f\r\v\b]/}}, is
perhaps a bit of overkill; it checks for a @samp{-} followed by anything
that is not whitespace and not a colon.
If the current command line argument does not match this pattern,
it is not an option, and it ends option processing.
@example
@group
@c file eg/lib/getopt.awk
if (_opti == 0)
_opti = 2
thisopt = substr(argv[Optind], _opti, 1)
Optopt = thisopt
i = index(options, thisopt)
if (i == 0) @{
if (Opterr)
printf("%c -- invalid option\n",
thisopt) > "/dev/stderr"
if (_opti >= length(argv[Optind])) @{
Optind++
_opti = 0
@} else
_opti++
return "?"
@}
@c endfile
@end group
@end example
The @code{_opti} variable tracks the position in the current command line
argument (@code{argv[Optind]}). In the case that multiple options were
grouped together with one @samp{-} (e.g., @samp{-abx}), it is necessary
to return them to the user one at a time.
If @code{_opti} is equal to zero, it is set to two, the index in the string
of the next character to look at (we skip the @samp{-}, which is at position
one). The variable @code{thisopt} holds the character, obtained with
@code{substr}. It is saved in @code{Optopt} for the main program to use.
If @code{thisopt} is not in the @code{options} string, then it is an
invalid option. If @code{Opterr} is non-zero, @code{getopt} prints an error
message on the standard error that is similar to the message from the C
version of @code{getopt}.
Since the option is invalid, it is necessary to skip it and move on to the
next option character. If @code{_opti} is greater than or equal to the
length of the current command line argument, then it is necessary to move on
to the next one, so @code{Optind} is incremented and @code{_opti} is reset
to zero. Otherwise, @code{Optind} is left alone and @code{_opti} is merely
incremented.
In any case, since the option was invalid, @code{getopt} returns @samp{?}.
The main program can examine @code{Optopt} if it needs to know what the
invalid option letter actually was.
@example
@group
@c file eg/lib/getopt.awk
if (substr(options, i + 1, 1) == ":") @{
# get option argument
if (length(substr(argv[Optind], _opti + 1)) > 0)
Optarg = substr(argv[Optind], _opti + 1)
else
Optarg = argv[++Optind]
_opti = 0
@} else
Optarg = ""
@c endfile
@end group
@end example
If the option requires an argument, the option letter is followed by a colon
in the @code{options} string. If there are remaining characters in the
current command line argument (@code{argv[Optind]}), then the rest of that
string is assigned to @code{Optarg}. Otherwise, the next command line
argument is used (@samp{-xFOO} vs. @samp{@w{-x FOO}}). In either case,
@code{_opti} is reset to zero, since there are no more characters left to
examine in the current command line argument.
@example
@c @group
@c file eg/lib/getopt.awk
if (_opti == 0 || _opti >= length(argv[Optind])) @{
Optind++
_opti = 0
@} else
_opti++
return thisopt
@}
@c endfile
@c @end group
@end example
Finally, if @code{_opti} is either zero or greater than the length of the
current command line argument, it means this element in @code{argv} is
through being processed, so @code{Optind} is incremented to point to the
next element in @code{argv}. If neither condition is true, then only
@code{_opti} is incremented, so that the next option letter can be processed
on the next call to @code{getopt}.
@example
@c @group
@c file eg/lib/getopt.awk
BEGIN @{
Opterr = 1 # default is to diagnose
Optind = 1 # skip ARGV[0]
# test program
if (_getopt_test) @{
while ((_go_c = getopt(ARGC, ARGV, "ab:cd")) != -1)
printf("c = <%c>, optarg = <%s>\n",
_go_c, Optarg)
printf("non-option arguments:\n")
for (; Optind < ARGC; Optind++)
printf("\tARGV[%d] = <%s>\n",
Optind, ARGV[Optind])
@}
@}
@c endfile
@c @end group
@end example
The @code{BEGIN} rule initializes both @code{Opterr} and @code{Optind} to one.
@code{Opterr} is set to one, since the default behavior is for @code{getopt}
to print a diagnostic message upon seeing an invalid option. @code{Optind}
is set to one, since there's no reason to look at the program name, which is
in @code{ARGV[0]}.
The rest of the @code{BEGIN} rule is a simple test program. Here is the
result of two sample runs of the test program.
@example
@group
$ awk -f getopt.awk -v _getopt_test=1 -- -a -cbARG bax -x
@print{} c = <a>, optarg = <>
@print{} c = <c>, optarg = <>
@print{} c = <b>, optarg = <ARG>
@print{} non-option arguments:
@print{} ARGV[3] = <bax>
@print{} ARGV[4] = <-x>
@end group
@group
$ awk -f getopt.awk -v _getopt_test=1 -- -a -x -- xyz abc
@print{} c = <a>, optarg = <>
@error{} x -- invalid option
@print{} c = <?>, optarg = <>
@print{} non-option arguments:
@print{} ARGV[4] = <xyz>
@print{} ARGV[5] = <abc>
@end group
@end example
The first @samp{--} terminates the arguments to @code{awk}, so that it does
not try to interpret the @samp{-a} etc. as its own options.
Several of the sample programs presented in
@ref{Sample Programs, ,Practical @code{awk} Programs},
use @code{getopt} to process their arguments.
@node Passwd Functions, Group Functions, Getopt Function, Library Functions
@section Reading the User Database
@cindex @file{/dev/user}
The @file{/dev/user} special file
(@pxref{Special Files, ,Special File Names in @code{gawk}})
provides access to the current user's real and effective user and group id
numbers, and if available, the user's supplementary group set.
However, since these are numbers, they do not provide very useful
information to the average user. There needs to be some way to find the
user information associated with the user and group numbers. This
section presents a suite of functions for retrieving information from the
user database. @xref{Group Functions, ,Reading the Group Database},
for a similar suite that retrieves information from the group database.
@cindex @code{getpwent}, C version
@cindex user information
@cindex login information
@cindex account information
@cindex password file
The POSIX standard does not define the file where user information is
kept. Instead, it provides the @code{<pwd.h>} header file
and several C language subroutines for obtaining user information.
The primary function is @code{getpwent}, for ``get password entry.''
The ``password'' comes from the original user database file,
@file{/etc/passwd}, which kept user information, along with the
encrypted passwords (hence the name).
While an @code{awk} program could simply read @file{/etc/passwd} directly
(the format is well known), because of the way password
files are handled on networked systems,
this file may not contain complete information about the system's set of users.
@cindex @code{pwcat} program
To be sure of being
able to produce a readable, complete version of the user database, it is
necessary to write a small C program that calls @code{getpwent}.
@code{getpwent} is defined to return a pointer to a @code{struct passwd}.
Each time it is called, it returns the next entry in the database.
When there are no more entries, it returns @code{NULL}, the null pointer.
When this happens, the C program should call @code{endpwent} to close the
database.
Here is @code{pwcat}, a C program that ``cats'' the password database.
@findex pwcat.c
@example
@c @group
@c file eg/lib/pwcat.c
/*
* pwcat.c
*
* Generate a printable version of the password database
*
* Arnold Robbins
* arnold@@gnu.ai.mit.edu
* May 1993
* Public Domain
*/
#include <stdio.h>
#include <pwd.h>
int
main(argc, argv)
int argc;
char **argv;
@{
struct passwd *p;
while ((p = getpwent()) != NULL)
printf("%s:%s:%d:%d:%s:%s:%s\n",
p->pw_name, p->pw_passwd, p->pw_uid,
p->pw_gid, p->pw_gecos, p->pw_dir, p->pw_shell);
endpwent();
exit(0);
@}
@c endfile
@c @end group
@end example
If you don't understand C, don't worry about it.
The output from @code{pwcat} is the user database, in the traditional
@file{/etc/passwd} format of colon-separated fields. The fields are:
@table @asis
@item Login name
The user's login name.
@item Encrypted password
The user's encrypted password. This may not be available on some systems.
@item User-ID
The user's numeric user-id number.
@item Group-ID
The user's numeric group-id number.
@item Full name
The user's full name, and perhaps other information associated with the
user.
@item Home directory
The user's login, or ``home'' directory (familiar to shell programmers as
@code{$HOME}).
@item Login shell
The program that will be run when the user logs in. This is usually a
shell, such as Bash (the Gnu Bourne-Again shell).
@end table
Here are a few lines representative of @code{pwcat}'s output.
@example
@c @group
$ pwcat
@print{} root:3Ov02d5VaUPB6:0:1:Operator:/:/bin/sh
@print{} nobody:*:65534:65534::/:
@print{} daemon:*:1:1::/:
@print{} sys:*:2:2::/:/bin/csh
@print{} bin:*:3:3::/bin:
@print{} arnold:xyzzy:2076:10:Arnold Robbins:/home/arnold:/bin/sh
@print{} miriam:yxaay:112:10:Miriam Robbins:/home/miriam:/bin/sh
@print{} andy:abcca2:113:10:Andy Jacobs:/home/andy:/bin/sh
@dots{}
@c @end group
@end example
With that introduction, here is a group of functions for getting user
information. There are several functions here, corresponding to the C
functions of the same name.
@findex _pw_init
@example
@c file eg/lib/passwdawk.in
@group
# passwd.awk --- access password file information
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
BEGIN @{
# tailor this to suit your system
_pw_awklib = "/usr/local/libexec/awk/"
@}
@end group
@group
function _pw_init( oldfs, oldrs, olddol0, pwcat)
@{
if (_pw_inited)
return
oldfs = FS
oldrs = RS
olddol0 = $0
FS = ":"
RS = "\n"
pwcat = _pw_awklib "pwcat"
while ((pwcat | getline) > 0) @{
_pw_byname[$1] = $0
_pw_byuid[$3] = $0
_pw_bycount[++_pw_total] = $0
@}
close(pwcat)
_pw_count = 0
_pw_inited = 1
FS = oldfs
RS = oldrs
$0 = olddol0
@}
@c endfile
@end group
@end example
The @code{BEGIN} rule sets a private variable to the directory where
@code{pwcat} is stored. Since it is used to help out an @code{awk} library
routine, we have chosen to put it in @file{/usr/local/libexec/awk}.
You might want it to be in a different directory on your system.
The function @code{_pw_init} keeps three copies of the user information
in three associative arrays. The arrays are indexed by user name
(@code{_pw_byname}), by user-id number (@code{_pw_byuid}), and by order of
occurrence (@code{_pw_bycount}).
The variable @code{_pw_inited} is used for efficiency; @code{_pw_init} only
needs to be called once.
Since this function uses @code{getline} to read information from
@code{pwcat}, it first saves the values of @code{FS}, @code{RS}, and
@code{$0}. Doing so is necessary, since these functions could be called
from anywhere within a user's program, and the user may have his or her
own values for @code{FS} and @code{RS}.
@ignore
Problem, what if FIELDWIDTHS is in use? Sigh.
@end ignore
The main part of the function uses a loop to read database lines, split
the line into fields, and then store the line into each array as necessary.
When the loop is done, @code{@w{_pw_init}} cleans up by closing the pipeline,
setting @code{@w{_pw_inited}} to one, and restoring @code{FS}, @code{RS}, and
@code{$0}. The use of @code{@w{_pw_count}} will be explained below.
@findex getpwnam
@example
@group
@c file eg/lib/passwdawk.in
function getpwnam(name)
@{
_pw_init()
if (name in _pw_byname)
return _pw_byname[name]
return ""
@}
@c endfile
@end group
@end example
The @code{getpwnam} function takes a user name as a string argument. If that
user is in the database, it returns the appropriate line. Otherwise it
returns the null string.
@findex getpwuid
@example
@group
@c file eg/lib/passwdawk.in
function getpwuid(uid)
@{
_pw_init()
if (uid in _pw_byuid)
return _pw_byuid[uid]
return ""
@}
@c endfile
@end group
@end example
Similarly,
the @code{getpwuid} function takes a user-id number argument. If that
user number is in the database, it returns the appropriate line. Otherwise it
returns the null string.
@findex getpwent
@example
@c @group
@c file eg/lib/passwdawk.in
function getpwent()
@{
_pw_init()
if (_pw_count < _pw_total)
return _pw_bycount[++_pw_count]
return ""
@}
@c endfile
@c @end group
@end example
The @code{getpwent} function simply steps through the database, one entry at
a time. It uses @code{_pw_count} to track its current position in the
@code{_pw_bycount} array.
@findex endpwent
@example
@c @group
@c file eg/lib/passwdawk.in
function endpwent()
@{
_pw_count = 0
@}
@c endfile
@c @end group
@end example
The @code{@w{endpwent}} function resets @code{@w{_pw_count}} to zero, so that
subsequent calls to @code{getpwent} will start over again.
A conscious design decision in this suite is that each subroutine calls
@code{@w{_pw_init}} to initialize the database arrays. The overhead of running
a separate process to generate the user database, and the I/O to scan it,
will only be incurred if the user's main program actually calls one of these
functions. If this library file is loaded along with a user's program, but
none of the routines are ever called, then there is no extra run-time overhead.
(The alternative would be to move the body of @code{@w{_pw_init}} into a
@code{BEGIN} rule, which would always run @code{pwcat}. This simplifies the
code but runs an extra process that may never be needed.)
In turn, calling @code{_pw_init} is not too expensive, since the
@code{_pw_inited} variable keeps the program from reading the data more than
once. If you are worried about squeezing every last cycle out of your
@code{awk} program, the check of @code{_pw_inited} could be moved out of
@code{_pw_init} and duplicated in all the other functions. In practice,
this is not necessary, since most @code{awk} programs are I/O bound, and it
would clutter up the code.
The @code{id} program in @ref{Id Program, ,Printing Out User Information},
uses these functions.
@node Group Functions, Library Names, Passwd Functions, Library Functions
@section Reading the Group Database
@cindex @code{getgrent}, C version
@cindex group information
@cindex account information
@cindex group file
Much of the discussion presented in
@ref{Passwd Functions, ,Reading the User Database},
applies to the group database as well. Although there has traditionally
been a well known file, @file{/etc/group}, in a well known format, the POSIX
standard only provides a set of C library routines
(@code{<grp.h>} and @code{getgrent})
for accessing the information.
Even though this file may exist, it likely does not have
complete information. Therefore, as with the user database, it is necessary
to have a small C program that generates the group database as its output.
@cindex @code{grcat} program
Here is @code{grcat}, a C program that ``cats'' the group database.
@findex grcat.c
@example
@c @group
@c file eg/lib/grcat.c
/*
* grcat.c
*
* Generate a printable version of the group database
*
* Arnold Robbins, arnold@@gnu.ai.mit.edu
* May 1993
* Public Domain
*/
#include <stdio.h>
#include <grp.h>
@group
int
main(argc, argv)
int argc;
char **argv;
@{
struct group *g;
int i;
@end group
while ((g = getgrent()) != NULL) @{
printf("%s:%s:%d:", g->gr_name, g->gr_passwd,
g->gr_gid);
for (i = 0; g->gr_mem[i] != NULL; i++) @{
printf("%s", g->gr_mem[i]);
if (g->gr_mem[i+1] != NULL)
putchar(',');
@}
putchar('\n');
@}
endgrent();
exit(0);
@}
@c endfile
@c @end group
@end example
Each line in the group database represent one group. The fields are
separated with colons, and represent the following information.
@table @asis
@item Group Name
The name of the group.
@item Group Password
The encrypted group password. In practice, this field is never used. It is
usually empty, or set to @samp{*}.
@item Group ID Number
The numeric group-id number. This number should be unique within the file.
@item Group Member List
A comma-separated list of user names. These users are members of the group.
Most Unix systems allow users to be members of several groups
simultaneously. If your system does, then reading @file{/dev/user} will
return those group-id numbers in @code{$5} through @code{$NF}.
(Note that @file{/dev/user} is a @code{gawk} extension;
@pxref{Special Files, ,Special File Names in @code{gawk}}.)
@end table
Here is what running @code{grcat} might produce:
@example
@group
$ grcat
@print{} wheel:*:0:arnold
@print{} nogroup:*:65534:
@print{} daemon:*:1:
@print{} kmem:*:2:
@print{} staff:*:10:arnold,miriam,andy
@print{} other:*:20:
@dots{}
@end group
@end example
Here are the functions for obtaining information from the group database.
There are several, modeled after the C library functions of the same names.
@findex _gr_init
@example
@group
@c file eg/lib/groupawk.in
# group.awk --- functions for dealing with the group file
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
BEGIN \
@{
# Change to suit your system
_gr_awklib = "/usr/local/libexec/awk/"
@}
@c endfile
@end group
@group
@c file eg/lib/groupawk.in
function _gr_init( oldfs, oldrs, olddol0, grcat, n, a, i)
@{
if (_gr_inited)
return
@end group
@group
oldfs = FS
oldrs = RS
olddol0 = $0
FS = ":"
RS = "\n"
@end group
@group
grcat = _gr_awklib "grcat"
while ((grcat | getline) > 0) @{
if ($1 in _gr_byname)
_gr_byname[$1] = _gr_byname[$1] "," $4
else
_gr_byname[$1] = $0
if ($3 in _gr_bygid)
_gr_bygid[$3] = _gr_bygid[$3] "," $4
else
_gr_bygid[$3] = $0
n = split($4, a, "[ \t]*,[ \t]*")
@end group
@group
for (i = 1; i <= n; i++)
if (a[i] in _gr_groupsbyuser)
_gr_groupsbyuser[a[i]] = \
_gr_groupsbyuser[a[i]] " " $1
else
_gr_groupsbyuser[a[i]] = $1
@end group
@group
_gr_bycount[++_gr_count] = $0
@}
@end group
@group
close(grcat)
_gr_count = 0
_gr_inited++
FS = oldfs
RS = oldrs
$0 = olddol0
@}
@c endfile
@end group
@end example
The @code{BEGIN} rule sets a private variable to the directory where
@code{grcat} is stored. Since it is used to help out an @code{awk} library
routine, we have chosen to put it in @file{/usr/local/libexec/awk}. You might
want it to be in a different directory on your system.
These routines follow the same general outline as the user database routines
(@pxref{Passwd Functions, ,Reading the User Database}).
The @code{@w{_gr_inited}} variable is used to
ensure that the database is scanned no more than once.
The @code{@w{_gr_init}} function first saves @code{FS}, @code{RS}, and
@code{$0}, and then sets @code{FS} and @code{RS} to the correct values for
scanning the group information.
The group information is stored is several associative arrays.
The arrays are indexed by group name (@code{@w{_gr_byname}}), by group-id number
(@code{@w{_gr_bygid}}), and by position in the database (@code{@w{_gr_bycount}}).
There is an additional array indexed by user name (@code{@w{_gr_groupsbyuser}}),
that is a space separated list of groups that each user belongs to.
Unlike the user database, it is possible to have multiple records in the
database for the same group. This is common when a group has a large number
of members. Such a pair of entries might look like:
@example
tvpeople:*:101:johny,jay,arsenio
tvpeople:*:101:david,conan,tom,joan
@end example
For this reason, @code{_gr_init} looks to see if a group name or
group-id number has already been seen. If it has, then the user names are
simply concatenated onto the previous list of users. (There is actually a
subtle problem with the code presented above. Suppose that
the first time there were no names. This code adds the names with
a leading comma. It also doesn't check that there is a @code{$4}.)
Finally, @code{_gr_init} closes the pipeline to @code{grcat}, restores
@code{FS}, @code{RS}, and @code{$0}, initializes @code{_gr_count} to zero
(it is used later), and makes @code{_gr_inited} non-zero.
@findex getgrnam
@example
@c @group
@c file eg/lib/groupawk.in
function getgrnam(group)
@{
_gr_init()
if (group in _gr_byname)
return _gr_byname[group]
return ""
@}
@c endfile
@c @end group
@end example
The @code{getgrnam} function takes a group name as its argument, and if that
group exists, it is returned. Otherwise, @code{getgrnam} returns the null
string.
@findex getgrgid
@example
@c @group
@c file eg/lib/groupawk.in
function getgrgid(gid)
@{
_gr_init()
if (gid in _gr_bygid)
return _gr_bygid[gid]
return ""
@}
@c endfile
@c @end group
@end example
The @code{getgrgid} function is similar, it takes a numeric group-id, and
looks up the information associated with that group-id.
@findex getgruser
@example
@group
@c file eg/lib/groupawk.in
function getgruser(user)
@{
_gr_init()
if (user in _gr_groupsbyuser)
return _gr_groupsbyuser[user]
return ""
@}
@c endfile
@end group
@end example
The @code{getgruser} function does not have a C counterpart. It takes a
user name, and returns the list of groups that have the user as a member.
@findex getgrent
@example
@c @group
@c file eg/lib/groupawk.in
function getgrent()
@{
_gr_init()
if (++gr_count in _gr_bycount)
return _gr_bycount[_gr_count]
return ""
@}
@c endfile
@c @end group
@end example
The @code{getgrent} function steps through the database one entry at a time.
It uses @code{_gr_count} to track its position in the list.
@findex endgrent
@example
@group
@c file eg/lib/groupawk.in
function endgrent()
@{
_gr_count = 0
@}
@c endfile
@end group
@end example
@code{endgrent} resets @code{_gr_count} to zero so that @code{getgrent} can
start over again.
As with the user database routines, each function calls @code{_gr_init} to
initialize the arrays. Doing so only incurs the extra overhead of running
@code{grcat} if these functions are used (as opposed to moving the body of
@code{_gr_init} into a @code{BEGIN} rule).
Most of the work is in scanning the database and building the various
associative arrays. The functions that the user calls are themselves very
simple, relying on @code{awk}'s associative arrays to do work.
The @code{id} program in @ref{Id Program, ,Printing Out User Information},
uses these functions.
@node Library Names, , Group Functions, Library Functions
@section Naming Library Function Global Variables
@cindex namespace issues in @code{awk}
@cindex documenting @code{awk} programs
@cindex programs, documenting
Due to the way the @code{awk} language evolved, variables are either
@dfn{global} (usable by the entire program), or @dfn{local} (usable just by
a specific function). There is no intermediate state analogous to
@code{static} variables in C.
Library functions often need to have global variables that they can use to
preserve state information between calls to the function. For example,
@code{getopt}'s variable @code{_opti}
(@pxref{Getopt Function, ,Processing Command Line Options}),
and the @code{_tm_months} array used by @code{mktime}
(@pxref{Mktime Function, ,Turning Dates Into Timestamps}).
Such variables are called @dfn{private}, since the only functions that need to
use them are the ones in the library.
When writing a library function, you should try to choose names for your
private variables so that they will not conflict with any variables used by
either another library function or a user's main program. For example, a
name like @samp{i} or @samp{j} is not a good choice, since user programs
often use variable names like these for their own purposes.
The example programs shown in this chapter all start the names of their
private variables with an underscore (@samp{_}). Users generally don't use
leading underscores in their variable names, so this convention immediately
decreases the chances that the variable name will be accidentally shared
with the user's program.
In addition, several of the library functions use a prefix that helps
indicate what function or set of functions uses the variables. For example,
@code{_tm_months} in @code{mktime}
(@pxref{Mktime Function, ,Turning Dates Into Timestamps}), and
@code{_pw_byname} in the user data base routines
(@pxref{Passwd Functions, ,Reading the User Database}).
This convention is recommended, since it even further decreases the chance
of inadvertent conflict among variable names.
Note that this convention can be used equally well both for variable names
and for private function names too.
While I could have re-written all the library routines to use this
convention, I did not do so, in order to show how my own @code{awk}
programming style has evolved, and to provide some basis for this
discussion.
As a final note on variable naming, if a function makes global variables
available for use by a main program, it is a good convention to start that
variable's name with a capital letter.
For example, @code{getopt}'s @code{Opterr} and @code{Optind} variables
(@pxref{Getopt Function, ,Processing Command Line Options}).
The leading capital letter indicates that it is global, while the fact that
the variable name is not all capital letters indicates that the variable is
not one of @code{awk}'s built-in variables, like @code{FS}.
It is also important that @emph{all} variables in library functions
that do not need to save state are in fact declared local. If this is
not done, the variable could accidentally be used in the user's program,
leading to bugs that are very difficult to track down.
@example
function lib_func(x, y, l1, l2)
@{
@dots{}
@var{use variable} some_var # some_var could be local
@dots{} # but is not by oversight
@}
@end example
@cindex Tcl
A different convention, common in the Tcl community, is to use a single
associative array to hold the values needed by the library function(s), or
``package.'' This significantly decreases the number of actual global names
in use. For example, the functions described in
@ref{Passwd Functions, , Reading the User Database},
might have used @code{@w{PW_data["inited"]}}, @code{@w{PW_data["total"]}},
@code{@w{PW_data["count"]}} and @code{@w{PW_data["awklib"]}}, instead of
@code{@w{_pw_inited}}, @code{@w{_pw_awklib}}, @code{@w{_pw_total}},
and @code{@w{_pw_count}}.
The conventions presented in this section are exactly that, conventions. You
are not required to write your programs this way, we merely recommend that
you do so.
@node Sample Programs, Language History, Library Functions, Top
@chapter Practical @code{awk} Programs
This chapter presents a potpourri of @code{awk} programs for your reading
enjoyment.
@iftex
There are two sections. The first presents @code{awk}
versions of several common POSIX utilities.
The second is a grab-bag of interesting programs.
@end iftex
Many of these programs use the library functions presented in
@ref{Library Functions, ,A Library of @code{awk} Functions}.
@menu
* Clones:: Clones of common utilities.
* Miscellaneous Programs:: Some interesting @code{awk} programs.
@end menu
@node Clones, Miscellaneous Programs, Sample Programs, Sample Programs
@section Re-inventing Wheels for Fun and Profit
This section presents a number of POSIX utilities that are implemented in
@code{awk}. Re-inventing these programs in @code{awk} is often enjoyable,
since the algorithms can be very clearly expressed, and usually the code is
very concise and simple. This is true because @code{awk} does so much for you.
It should be noted that these programs are not necessarily intended to
replace the installed versions on your system. Instead, their
purpose is to illustrate @code{awk} language programming for ``real world''
tasks.
The programs are presented in alphabetical order.
@menu
* Cut Program:: The @code{cut} utility.
* Egrep Program:: The @code{egrep} utility.
* Id Program:: The @code{id} utility.
* Split Program:: The @code{split} utility.
* Tee Program:: The @code{tee} utility.
* Uniq Program:: The @code{uniq} utility.
* Wc Program:: The @code{wc} utility.
@end menu
@node Cut Program, Egrep Program, Clones, Clones
@subsection Cutting Out Fields and Columns
@cindex @code{cut} utility
The @code{cut} utility selects, or ``cuts,'' either characters or fields
from its standard
input and sends them to its standard output. @code{cut} can cut out either
a list of characters, or a list of fields. By default, fields are separated
by tabs, but you may supply a command line option to change the field
@dfn{delimiter}, i.e.@: the field separator character. @code{cut}'s definition
of fields is less general than @code{awk}'s.
A common use of @code{cut} might be to pull out just the login name of
logged-on users from the output of @code{who}. For example, the following
pipeline generates a sorted, unique list of the logged on users:
@example
who | cut -c1-8 | sort | uniq
@end example
The options for @code{cut} are:
@table @code
@item -c @var{list}
Use @var{list} as the list of characters to cut out. Items within the list
may be separated by commas, and ranges of characters can be separated with
dashes. The list @samp{1-8,15,22-35} specifies characters one through
eight, 15, and 22 through 35.
@item -f @var{list}
Use @var{list} as the list of fields to cut out.
@item -d @var{delim}
Use @var{delim} as the field separator character instead of the tab
character.
@item -s
Suppress printing of lines that do not contain the field delimiter.
@end table
The @code{awk} implementation of @code{cut} uses the @code{getopt} library
function (@pxref{Getopt Function, ,Processing Command Line Options}),
and the @code{join} library function
(@pxref{Join Function, ,Merging an Array Into a String}).
The program begins with a comment describing the options and a @code{usage}
function which prints out a usage message and exits. @code{usage} is called
if invalid arguments are supplied.
@findex cut.awk
@example
@c @group
@c file eg/prog/cut.awk
# cut.awk --- implement cut in awk
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
# Options:
# -f list Cut fields
# -d c Field delimiter character
# -c list Cut characters
#
# -s Suppress lines without the delimiter character
function usage( e1, e2)
@{
e1 = "usage: cut [-f list] [-d c] [-s] [files...]"
e2 = "usage: cut [-c list] [files...]"
print e1 > "/dev/stderr"
print e2 > "/dev/stderr"
exit 1
@}
@c endfile
@c @end group
@end example
@noindent
The variables @code{e1} and @code{e2} are used so that the function
fits nicely on the
@iftex
page.
@end iftex
@ifinfo
screen.
@end ifinfo
Next comes a @code{BEGIN} rule that parses the command line options.
It sets @code{FS} to a single tab character, since that is @code{cut}'s
default field separator. The output field separator is also set to be the
same as the input field separator. Then @code{getopt} is used to step
through the command line options. One or the other of the variables
@code{by_fields} or @code{by_chars} is set to true, to indicate that
processing should be done by fields or by characters respectively.
When cutting by characters, the output field separator is set to the null
string.
@example
@c @group
@c file eg/prog/cut.awk
BEGIN \
@{
FS = "\t" # default
OFS = FS
while ((c = getopt(ARGC, ARGV, "sf:c:d:")) != -1) @{
if (c == "f") @{
by_fields = 1
fieldlist = Optarg
@group
@} else if (c == "c") @{
by_chars = 1
fieldlist = Optarg
OFS = ""
@} else if (c == "d") @{
if (length(Optarg) > 1) @{
printf("Using first character of %s" \
" for delimiter\n", Optarg) > "/dev/stderr"
Optarg = substr(Optarg, 1, 1)
@}
FS = Optarg
OFS = FS
if (FS == " ") # defeat awk semantics
FS = "[ ]"
@} else if (c == "s")
suppress++
else
usage()
@}
@end group
for (i = 1; i < Optind; i++)
ARGV[i] = ""
@c endfile
@c @end group
@end example
Special care is taken when the field delimiter is a space. Using
@code{@w{" "}} (a single space) for the value of @code{FS} is
incorrect---@code{awk} would
separate fields with runs of spaces, tabs and/or newlines, and we want them to be
separated with individual spaces. Also, note that after @code{getopt} is
through, we have to clear out all the elements of @code{ARGV} from one to
@code{Optind}, so that @code{awk} will not try to process the command line
options as file names.
After dealing with the command line options, the program verifies that the
options make sense. Only one or the other of @samp{-c} and @samp{-f} should
be used, and both require a field list. Then either @code{set_fieldlist} or
@code{set_charlist} is called to pull apart the list of fields or
characters.
@example
@c @group
@c file eg/prog/cut.awk
if (by_fields && by_chars)
usage()
if (by_fields == 0 && by_chars == 0)
by_fields = 1 # default
if (fieldlist == "") @{
print "cut: needs list for -c or -f" > "/dev/stderr"
exit 1
@}
@group
if (by_fields)
set_fieldlist()
else
set_charlist()
@}
@c endfile
@end group
@end example
Here is @code{set_fieldlist}. It first splits the field list apart
at the commas, into an array. Then, for each element of the array, it
looks to see if it is actually a range, and if so splits it apart. The range
is verified to make sure the first number is smaller than the second.
Each number in the list is added to the @code{flist} array, which simply
lists the fields that will be printed.
Normal field splitting is used.
The program lets @code{awk}
handle the job of doing the field splitting.
@example
@c @group
@c file eg/prog/cut.awk
function set_fieldlist( n, m, i, j, k, f, g)
@{
n = split(fieldlist, f, ",")
j = 1 # index in flist
for (i = 1; i <= n; i++) @{
if (index(f[i], "-") != 0) @{ # a range
m = split(f[i], g, "-")
if (m != 2 || g[1] >= g[2]) @{
printf("bad field list: %s\n",
f[i]) > "/dev/stderr"
exit 1
@}
for (k = g[1]; k <= g[2]; k++)
flist[j++] = k
@} else
flist[j++] = f[i]
@}
nfields = j - 1
@}
@c endfile
@c @end group
@end example
The @code{set_charlist} function is more complicated than @code{set_fieldlist}.
The idea here is to use @code{gawk}'s @code{FIELDWIDTHS} variable
(@pxref{Constant Size, ,Reading Fixed-width Data}),
which describes constant width input. When using a character list, that is
exactly what we have.
Setting up @code{FIELDWIDTHS} is more complicated than simply listing the
fields that need to be printed. We have to keep track of the fields to be
printed, and also the intervening characters that have to be skipped.
For example, suppose you wanted characters one through eight, 15, and
22 through 35. You would use @samp{-c 1-8,15,22-35}. The necessary value
for @code{FIELDWIDTHS} would be @code{@w{"8 6 1 6 14"}}. This gives us five
fields, and what should be printed are @code{$1}, @code{$3}, and @code{$5}.
The intermediate fields are ``filler,'' stuff in between the desired data.
@code{flist} lists the fields to be printed, and @code{t} tracks the
complete field list, including filler fields.
@example
@c @group
@c file eg/prog/cut.awk
function set_charlist( field, i, j, f, g, t,
filler, last, len)
@{
field = 1 # count total fields
n = split(fieldlist, f, ",")
j = 1 # index in flist
for (i = 1; i <= n; i++) @{
if (index(f[i], "-") != 0) @{ # range
m = split(f[i], g, "-")
if (m != 2 || g[1] >= g[2]) @{
printf("bad character list: %s\n",
f[i]) > "/dev/stderr"
exit 1
@}
len = g[2] - g[1] + 1
if (g[1] > 1) # compute length of filler
filler = g[1] - last - 1
else
filler = 0
if (filler)
t[field++] = filler
t[field++] = len # length of field
last = g[2]
flist[j++] = field - 1
@} else @{
if (f[i] > 1)
filler = f[i] - last - 1
else
filler = 0
if (filler)
t[field++] = filler
t[field++] = 1
last = f[i]
flist[j++] = field - 1
@}
@}
@group
FIELDWIDTHS = join(t, 1, field - 1)
nfields = j - 1
@}
@end group
@c endfile
@end example
Here is the rule that actually processes the data. If the @samp{-s} option
was given, then @code{suppress} will be true. The first @code{if} statement
makes sure that the input record does have the field separator. If
@code{cut} is processing fields, @code{suppress} is true, and the field
separator character is not in the record, then the record is skipped.
If the record is valid, then at this point, @code{gawk} has split the data
into fields, either using the character in @code{FS} or using fixed-length
fields and @code{FIELDWIDTHS}. The loop goes through the list of fields
that should be printed. If the corresponding field has data in it, it is
printed. If the next field also has data, then the separator character is
written out in between the fields.
@c 2e: Could use `index($0, FS) != 0' instead of `$0 !~ FS', below
@example
@c @group
@c file eg/prog/cut.awk
@{
if (by_fields && suppress && $0 !~ FS)
next
for (i = 1; i <= nfields; i++) @{
if ($flist[i] != "") @{
printf "%s", $flist[i]
if (i < nfields && $flist[i+1] != "")
printf "%s", OFS
@}
@}
print ""
@}
@c endfile
@c @end group
@end example
This version of @code{cut} relies on @code{gawk}'s @code{FIELDWIDTHS}
variable to do the character-based cutting. While it would be possible in
other @code{awk} implementations to use @code{substr}
(@pxref{String Functions, ,Built-in Functions for String Manipulation}),
it would also be extremely painful to do so.
The @code{FIELDWIDTHS} variable supplies an elegant solution to the problem
of picking the input line apart by characters.
@node Egrep Program, Id Program, Cut Program, Clones
@subsection Searching for Regular Expressions in Files
@cindex @code{egrep} utility
The @code{egrep} utility searches files for patterns. It uses regular
expressions that are almost identical to those available in @code{awk}
(@pxref{Regexp Constants, ,Regular Expression Constants}). It is used this way:
@example
egrep @r{[} @var{options} @r{]} '@var{pattern}' @var{files} @dots{}
@end example
The @var{pattern} is a regexp.
In typical usage, the regexp is quoted to prevent the shell from expanding
any of the special characters as file name wildcards.
Normally, @code{egrep} prints the
lines that matched. If multiple file names are provided on the command
line, each output line is preceded by the name of the file and a colon.
@c NEEDED
@page
The options are:
@table @code
@item -c
Print out a count of the lines that matched the pattern, instead of the
lines themselves.
@item -s
Be silent. No output is produced, and the exit value indicates whether
or not the pattern was matched.
@item -v
Invert the sense of the test. @code{egrep} prints the lines that do
@emph{not} match the pattern, and exits successfully if the pattern was not
matched.
@item -i
Ignore case distinctions in both the pattern and the input data.
@item -l
Only print the names of the files that matched, not the lines that matched.
@item -e @var{pattern}
Use @var{pattern} as the regexp to match. The purpose of the @samp{-e}
option is to allow patterns that start with a @samp{-}.
@end table
This version uses the @code{getopt} library function
(@pxref{Getopt Function, ,Processing Command Line Options}),
and the file transition library program
(@pxref{Filetrans Function, ,Noting Data File Boundaries}).
The program begins with a descriptive comment, and then a @code{BEGIN} rule
that processes the command line arguments with @code{getopt}. The @samp{-i}
(ignore case) option is particularly easy with @code{gawk}; we just use the
@code{IGNORECASE} built in variable
(@pxref{Built-in Variables}).
@findex egrep.awk
@example
@c @group
@c file eg/prog/egrep.awk
# egrep.awk --- simulate egrep in awk
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
# Options:
# -c count of lines
# -s silent - use exit value
# -v invert test, success if no match
# -i ignore case
# -l print filenames only
# -e argument is pattern
BEGIN @{
while ((c = getopt(ARGC, ARGV, "ce:svil")) != -1) @{
if (c == "c")
count_only++
else if (c == "s")
no_print++
else if (c == "v")
invert++
else if (c == "i")
IGNORECASE = 1
else if (c == "l")
filenames_only++
else if (c == "e")
pattern = Optarg
else
usage()
@}
@c endfile
@c @end group
@end example
Next comes the code that handles the @code{egrep} specific behavior. If no
pattern was supplied with @samp{-e}, the first non-option on the command
line is used. The @code{awk} command line arguments up to @code{ARGV[Optind]}
are cleared, so that @code{awk} won't try to process them as files. If no
files were specified, the standard input is used, and if multiple files were
specified, we make sure to note this so that the file names can precede the
matched lines in the output.
The last two lines are commented out, since they are not needed in
@code{gawk}. They should be uncommented if you have to use another version
of @code{awk}.
@example
@c @group
@c file eg/prog/egrep.awk
if (pattern == "")
pattern = ARGV[Optind++]
for (i = 1; i < Optind; i++)
ARGV[i] = ""
if (Optind >= ARGC) @{
ARGV[1] = "-"
ARGC = 2
@} else if (ARGC - Optind > 1)
do_filenames++
# if (IGNORECASE)
# pattern = tolower(pattern)
@}
@c endfile
@c @end group
@end example
The next set of lines should be uncommented if you are not using
@code{gawk}. This rule translates all the characters in the input line
into lower-case if the @samp{-i} option was specified. The rule is
commented out since it is not necessary with @code{gawk}.
@c bug: if a match happens, we output the translated line, not the original
@example
@c @group
@c file eg/prog/egrep.awk
#@{
# if (IGNORECASE)
# $0 = tolower($0)
#@}
@c endfile
@c @end group
@end example
The @code{beginfile} function is called by the rule in @file{ftrans.awk}
when each new file is processed. In this case, it is very simple; all it
does is initialize a variable @code{fcount} to zero. @code{fcount} tracks
how many lines in the current file matched the pattern.
@example
@group
@c file eg/prog/egrep.awk
function beginfile(junk)
@{
fcount = 0
@}
@c endfile
@end group
@end example
The @code{endfile} function is called after each file has been processed.
It is used only when the user wants a count of the number of lines that
matched. @code{no_print} will be true only if the exit status is desired.
@code{count_only} will be true if line counts are desired. @code{egrep}
will therefore only print line counts if printing and counting are enabled.
The output format must be adjusted depending upon the number of files to be
processed. Finally, @code{fcount} is added to @code{total}, so that we
know how many lines altogether matched the pattern.
@example
@c @group
@c file eg/prog/egrep.awk
function endfile(file)
@{
if (! no_print && count_only)
if (do_filenames)
print file ":" fcount
else
print fcount
total += fcount
@}
@c endfile
@c @end group
@end example
This rule does most of the work of matching lines. The variable
@code{matches} will be true if the line matched the pattern. If the user
wants lines that did not match, the sense of the @code{matches} is inverted
using the @samp{!} operator. @code{fcount} is incremented with the value of
@code{matches}, which will be either one or zero, depending upon a
successful or unsuccessful match. If the line did not match, the
@code{next} statement just moves on to the next record.
There are several optimizations for performance in the following few lines
of code. If the user only wants exit status (@code{no_print} is true), and
we don't have to count lines, then it is enough to know that one line in
this file matched, and we can skip on to the next file with @code{nextfile}.
Along similar lines, if we are only printing file names, and we
don't need to count lines, we can print the file name, and then skip to the
next file with @code{nextfile}.
Finally, each line is printed, with a leading filename and colon if
necessary.
@ignore
2e: note, probably better to recode the last few lines as
if (! count_only) @{
if (no_print)
nextfile
if (filenames_only) @{
print FILENAME
nextfile
@}
if (do_filenames)
print FILENAME ":" $0
else
print
@}
@end ignore
@example
@c @group
@c file eg/prog/egrep.awk
@{
matches = ($0 ~ pattern)
if (invert)
matches = ! matches
fcount += matches # 1 or 0
@group
if (! matches)
next
@end group
if (no_print && ! count_only)
nextfile
if (filenames_only && ! count_only) @{
print FILENAME
nextfile
@}
if (do_filenames && ! count_only)
print FILENAME ":" $0
else if (! count_only)
print
@}
@c endfile
@c @end group
@end example
@c @strong{Exercise}: rearrange the code inside @samp{if (! count_only)}.
The @code{END} rule takes care of producing the correct exit status. If
there were no matches, the exit status is one, otherwise it is zero.
@example
@c @group
@c file eg/prog/egrep.awk
END \
@{
if (total == 0)
exit 1
exit 0
@}
@c endfile
@c @end group
@end example
The @code{usage} function prints a usage message in case of invalid options
and then exits.
@example
@c @group
@c file eg/prog/egrep.awk
function usage( e)
@{
e = "Usage: egrep [-csvil] [-e pat] [files ...]"
print e > "/dev/stderr"
exit 1
@}
@c endfile
@c @end group
@end example
The variable @code{e} is used so that the function fits nicely
on the printed page.
@cindex backslash continuation
Just a note on programming style. You may have noticed that the @code{END}
rule uses backslash continuation, with the open brace on a line by
itself. This is so that it more closely resembles the way functions
are written. Many of the examples
@iftex
in this chapter
@end iftex
use this style. You can decide for yourself if you like writing
your @code{BEGIN} and @code{END} rules this way,
or not.
@node Id Program, Split Program, Egrep Program, Clones
@subsection Printing Out User Information
@cindex @code{id} utility
The @code{id} utility lists a user's real and effective user-id numbers,
real and effective group-id numbers, and the user's group set, if any.
@code{id} will only print the effective user-id and group-id if they are
different from the real ones. If possible, @code{id} will also supply the
corresponding user and group names. The output might look like this:
@example
$ id
@print{} uid=2076(arnold) gid=10(staff) groups=10(staff),4(tty)
@end example
This information is exactly what is provided by @code{gawk}'s
@file{/dev/user} special file (@pxref{Special Files, ,Special File Names in @code{gawk}}).
However, the @code{id} utility provides a more palatable output than just a
string of numbers.
Here is a simple version of @code{id} written in @code{awk}.
It uses the user database library functions
(@pxref{Passwd Functions, ,Reading the User Database}),
and the group database library functions
(@pxref{Group Functions, ,Reading the Group Database}).
The program is fairly straightforward. All the work is done in the
@code{BEGIN} rule. The user and group id numbers are obtained from
@file{/dev/user}. If there is no support for @file{/dev/user}, the program
gives up.
The code is repetitive. The entry in the user database for the real user-id
number is split into parts at the @samp{:}. The name is the first field.
Similar code is used for the effective user-id number, and the group
numbers.
@findex id.awk
@example
@c @group
@c file eg/prog/id.awk
# id.awk --- implement id in awk
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
# output is:
# uid=12(foo) euid=34(bar) gid=3(baz) \
# egid=5(blat) groups=9(nine),2(two),1(one)
BEGIN \
@{
if ((getline < "/dev/user") < 0) @{
err = "id: no /dev/user support - cannot run"
print err > "/dev/stderr"
exit 1
@}
close("/dev/user")
uid = $1
euid = $2
gid = $3
egid = $4
printf("uid=%d", uid)
pw = getpwuid(uid)
@group
if (pw != "") @{
split(pw, a, ":")
printf("(%s)", a[1])
@}
@end group
if (euid != uid) @{
printf(" euid=%d", euid)
pw = getpwuid(euid)
if (pw != "") @{
split(pw, a, ":")
printf("(%s)", a[1])
@}
@}
printf(" gid=%d", gid)
pw = getgrgid(gid)
if (pw != "") @{
split(pw, a, ":")
printf("(%s)", a[1])
@}
if (egid != gid) @{
printf(" egid=%d", egid)
pw = getgrgid(egid)
if (pw != "") @{
split(pw, a, ":")
printf("(%s)", a[1])
@}
@}
if (NF > 4) @{
printf(" groups=");
for (i = 5; i <= NF; i++) @{
printf("%d", $i)
pw = getgrgid($i)
if (pw != "") @{
split(pw, a, ":")
printf("(%s)", a[1])
@}
@group
if (i < NF)
printf(",")
@end group
@}
@}
print ""
@}
@c endfile
@c @end group
@end example
@c exercise!!!
@ignore
The POSIX version of @code{id} takes arguments that control which
information is printed. Modify this version to accept the same
arguments and perform in the same way.
@end ignore
@node Split Program, Tee Program, Id Program, Clones
@subsection Splitting a Large File Into Pieces
@cindex @code{split} utility
The @code{split} program splits large text files into smaller pieces. By default,
the output files are named @file{xaa}, @file{xab}, and so on. Each file has
1000 lines in it, with the likely exception of the last file. To change the
number of lines in each file, you supply a number on the command line
preceded with a minus, e.g., @samp{-500} for files with 500 lines in them
instead of 1000. To change the name of the output files to something like
@file{myfileaa}, @file{myfileab}, and so on, you supply an additional
argument that specifies the filename.
Here is a version of @code{split} in @code{awk}. It uses the @code{ord} and
@code{chr} functions presented in
@ref{Ordinal Functions, ,Translating Between Characters and Numbers}.
The program first sets its defaults, and then tests to make sure there are
not too many arguments. It then looks at each argument in turn. The
first argument could be a minus followed by a number. If it is, this happens
to look like a negative number, so it is made positive, and that is the
count of lines. The data file name is skipped over, and the final argument
is used as the prefix for the output file names.
@findex split.awk
@example
@c @group
@c file eg/prog/split.awk
# split.awk --- do split in awk
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
# usage: split [-num] [file] [outname]
BEGIN @{
outfile = "x" # default
count = 1000
if (ARGC > 4)
usage()
i = 1
if (ARGV[i] ~ /^-[0-9]+$/) @{
count = -ARGV[i]
ARGV[i] = ""
i++
@}
# test argv in case reading from stdin instead of file
if (i in ARGV)
i++ # skip data file name
if (i in ARGV) @{
outfile = ARGV[i]
ARGV[i] = ""
@}
s1 = s2 = "a"
out = (outfile s1 s2)
@}
@c endfile
@c @end group
@end example
The next rule does most of the work. @code{tcount} (temporary count) tracks
how many lines have been printed to the output file so far. If it is greater
than @code{count}, it is time to close the current file and start a new one.
@code{s1} and @code{s2} track the current suffixes for the file name. If
they are both @samp{z}, the file is just too big. Otherwise, @code{s1}
moves to the next letter in the alphabet and @code{s2} starts over again at
@samp{a}.
@example
@c @group
@c file eg/prog/split.awk
@{
if (++tcount > count) @{
close(out)
if (s2 == "z") @{
if (s1 == "z") @{
printf("split: %s is too large to split\n", \
FILENAME) > "/dev/stderr"
exit 1
@}
s1 = chr(ord(s1) + 1)
s2 = "a"
@} else
s2 = chr(ord(s2) + 1)
out = (outfile s1 s2)
tcount = 1
@}
print > out
@}
@c endfile
@c @end group
@end example
The @code{usage} function simply prints an error message and exits.
@example
@c @group
@c file eg/prog/split.awk
function usage( e)
@{
e = "usage: split [-num] [file] [outname]"
print e > "/dev/stderr"
exit 1
@}
@c endfile
@c @end group
@end example
@noindent
The variable @code{e} is used so that the function
fits nicely on the
@iftex
page.
@end iftex
@ifinfo
screen.
@end ifinfo
This program is a bit sloppy; it relies on @code{awk} to close the last file
for it automatically, instead of doing it in an @code{END} rule.
@node Tee Program, Uniq Program, Split Program, Clones
@subsection Duplicating Output Into Multiple Files
@cindex @code{tee} utility
The @code{tee} program is known as a ``pipe fitting.'' @code{tee} copies
its standard input to its standard output, and also duplicates it to the
files named on the command line. Its usage is:
@example
tee @r{[}-a@r{]} file @dots{}
@end example
The @samp{-a} option tells @code{tee} to append to the named files, instead of
truncating them and starting over.
The @code{BEGIN} rule first makes a copy of all the command line arguments,
into an array named @code{copy}.
@code{ARGV[0]} is not copied, since it is not needed.
@code{tee} cannot use @code{ARGV} directly, since @code{awk} will attempt to
process each file named in @code{ARGV} as input data.
If the first argument is @samp{-a}, then the flag variable
@code{append} is set to true, and both @code{ARGV[1]} and
@code{copy[1]} are deleted. If @code{ARGC} is less than two, then no file
names were supplied, and @code{tee} prints a usage message and exits.
Finally, @code{awk} is forced to read the standard input by setting
@code{ARGV[1]} to @code{"-"}, and @code{ARGC} to two.
@c 2e: the `ARGC--' in the `if (ARGV[1] == "-a")' isn't needed.
@findex tee.awk
@example
@group
@c file eg/prog/tee.awk
# tee.awk --- tee in awk
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
# Revised December 1995
@end group
@group
BEGIN \
@{
for (i = 1; i < ARGC; i++)
copy[i] = ARGV[i]
@end group
@group
if (ARGV[1] == "-a") @{
append = 1
delete ARGV[1]
delete copy[1]
ARGC--
@}
@end group
@group
if (ARGC < 2) @{
print "usage: tee [-a] file ..." > "/dev/stderr"
exit 1
@}
@end group
@group
ARGV[1] = "-"
ARGC = 2
@}
@c endfile
@end group
@end example
The single rule does all the work. Since there is no pattern, it is
executed for each line of input. The body of the rule simply prints the
line into each file on the command line, and then to the standard output.
@example
@group
@c file eg/prog/tee.awk
@{
# moving the if outside the loop makes it run faster
if (append)
for (i in copy)
print >> copy[i]
else
for (i in copy)
print > copy[i]
print
@}
@c endfile
@end group
@end example
It would have been possible to code the loop this way:
@example
for (i in copy)
if (append)
print >> copy[i]
else
print > copy[i]
@end example
@noindent
This is more concise, but it is also less efficient. The @samp{if} is
tested for each record and for each output file. By duplicating the loop
body, the @samp{if} is only tested once for each input record. If there are
@var{N} input records and @var{M} input files, the first method only
executes @var{N} @samp{if} statements, while the second would execute
@var{N}@code{*}@var{M} @samp{if} statements.
Finally, the @code{END} rule cleans up, by closing all the output files.
@example
@c @group
@c file eg/prog/tee.awk
END \
@{
for (i in copy)
close(copy[i])
@}
@c endfile
@c @end group
@end example
@node Uniq Program, Wc Program, Tee Program, Clones
@subsection Printing Non-duplicated Lines of Text
@cindex @code{uniq} utility
The @code{uniq} utility reads sorted lines of data on its standard input,
and (by default) removes duplicate lines. In other words, only unique lines
are printed, hence the name. @code{uniq} has a number of options. The usage is:
@example
uniq @r{[}-udc @r{[}-@var{n}@r{]]} @r{[}+@var{n}@r{]} @r{[} @var{input file} @r{[} @var{output file} @r{]]}
@end example
The option meanings are:
@table @code
@item -d
Only print repeated lines.
@item -u
Only print non-repeated lines.
@item -c
Count lines. This option overrides @samp{-d} and @samp{-u}. Both repeated
and non-repeated lines are counted.
@item -@var{n}
Skip @var{n} fields before comparing lines. The definition of fields
is similar to @code{awk}'s default: non-whitespace characters separated
by runs of spaces and/or tabs.
@item +@var{n}
Skip @var{n} characters before comparing lines. Any fields specified with
@samp{-@var{n}} are skipped first.
@item @var{input file}
Data is read from the input file named on the command line, instead of from
the standard input.
@item @var{output file}
The generated output is sent to the named output file, instead of to the
standard output.
@end table
Normally @code{uniq} behaves as if both the @samp{-d} and @samp{-u} options
had been provided.
Here is an @code{awk} implementation of @code{uniq}. It uses the
@code{getopt} library function
(@pxref{Getopt Function, ,Processing Command Line Options}),
and the @code{join} library function
(@pxref{Join Function, ,Merging an Array Into a String}).
The program begins with a @code{usage} function and then a brief outline of
the options and their meanings in a comment.
The @code{BEGIN} rule deals with the command line arguments and options. It
uses a trick to get @code{getopt} to handle options of the form @samp{-25},
treating such an option as the option letter @samp{2} with an argument of
@samp{5}. If indeed two or more digits were supplied (@code{Optarg} looks
like a number), @code{Optarg} is
concatenated with the option digit, and then result is added to zero to make
it into a number. If there is only one digit in the option, then
@code{Optarg} is not needed, and @code{Optind} must be decremented so that
@code{getopt} will process it next time. This code is admittedly a bit
tricky.
If no options were supplied, then the default is taken, to print both
repeated and non-repeated lines. The output file, if provided, is assigned
to @code{outputfile}. Earlier, @code{outputfile} was initialized to the
standard output, @file{/dev/stdout}.
@findex uniq.awk
@example
@c @group
@c file eg/prog/uniq.awk
# uniq.awk --- do uniq in awk
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
@group
function usage( e)
@{
e = "Usage: uniq [-udc [-n]] [+n] [ in [ out ]]"
print e > "/dev/stderr"
exit 1
@}
@end group
@group
# -c count lines. overrides -d and -u
# -d only repeated lines
# -u only non-repeated lines
# -n skip n fields
# +n skip n characters, skip fields first
@end group
BEGIN \
@{
count = 1
outputfile = "/dev/stdout"
opts = "udc0:1:2:3:4:5:6:7:8:9:"
while ((c = getopt(ARGC, ARGV, opts)) != -1) @{
if (c == "u")
non_repeated_only++
else if (c == "d")
repeated_only++
else if (c == "c")
do_count++
else if (index("0123456789", c) != 0) @{
# getopt requires args to options
# this messes us up for things like -5
if (Optarg ~ /^[0-9]+$/)
fcount = (c Optarg) + 0
else @{
fcount = c + 0
Optind--
@}
@} else
usage()
@}
if (ARGV[Optind] ~ /^\+[0-9]+$/) @{
charcount = substr(ARGV[Optind], 2) + 0
Optind++
@}
for (i = 1; i < Optind; i++)
ARGV[i] = ""
if (repeated_only == 0 && non_repeated_only == 0)
repeated_only = non_repeated_only = 1
@group
if (ARGC - Optind == 2) @{
outputfile = ARGV[ARGC - 1]
ARGV[ARGC - 1] = ""
@}
@}
@c endfile
@end group
@end example
The following function, @code{are_equal}, compares the current line,
@code{$0}, to the
previous line, @code{last}. It handles skipping fields and characters.
If no field count and no character count were specified, @code{are_equal}
simply returns one or zero depending upon the result of a simple string
comparison of @code{last} and @code{$0}. Otherwise, things get more
complicated.
If fields have to be skipped, each line is broken into an array using
@code{split}
(@pxref{String Functions, ,Built-in Functions for String Manipulation}),
and then the desired fields are joined back into a line using @code{join}.
The joined lines are stored in @code{clast} and @code{cline}.
If no fields are skipped, @code{clast} and @code{cline} are set to
@code{last} and @code{$0} respectively.
Finally, if characters are skipped, @code{substr} is used to strip off the
leading @code{charcount} characters in @code{clast} and @code{cline}. The
two strings are then compared, and @code{are_equal} returns the result.
@example
@c @group
@c file eg/prog/uniq.awk
function are_equal( n, m, clast, cline, alast, aline)
@{
if (fcount == 0 && charcount == 0)
return (last == $0)
if (fcount > 0) @{
n = split(last, alast)
m = split($0, aline)
clast = join(alast, fcount+1, n)
cline = join(aline, fcount+1, m)
@} else @{
clast = last
cline = $0
@}
if (charcount) @{
clast = substr(clast, charcount + 1)
cline = substr(cline, charcount + 1)
@}
return (clast == cline)
@}
@c endfile
@c @end group
@end example
The following two rules are the body of the program. The first one is
executed only for the very first line of data. It sets @code{last} equal to
@code{$0}, so that subsequent lines of text have something to be compared to.
The second rule does the work. The variable @code{equal} will be one or zero
depending upon the results of @code{are_equal}'s comparison. If @code{uniq}
is counting repeated lines, then the @code{count} variable is incremented if
the lines are equal. Otherwise the line is printed and @code{count} is
reset, since the two lines are not equal.
If @code{uniq} is not counting, @code{count} is incremented if the lines are
equal. Otherwise, if @code{uniq} is counting repeated lines, and more than
one line has been seen, or if @code{uniq} is counting non-repeated lines,
and only one line has been seen, then the line is printed, and @code{count}
is reset.
Finally, similar logic is used in the @code{END} rule to print the final
line of input data.
@example
@c @group
@c file eg/prog/uniq.awk
@group
NR == 1 @{
last = $0
next
@}
@end group
@{
equal = are_equal()
if (do_count) @{ # overrides -d and -u
if (equal)
count++
else @{
printf("%4d %s\n", count, last) > outputfile
last = $0
count = 1 # reset
@}
next
@}
if (equal)
count++
else @{
if ((repeated_only && count > 1) ||
(non_repeated_only && count == 1))
print last > outputfile
last = $0
count = 1
@}
@}
@group
END @{
if (do_count)
printf("%4d %s\n", count, last) > outputfile
else if ((repeated_only && count > 1) ||
(non_repeated_only && count == 1))
print last > outputfile
@}
@end group
@c endfile
@c @end group
@end example
@node Wc Program, , Uniq Program, Clones
@subsection Counting Things
@cindex @code{wc} utility
The @code{wc} (word count) utility counts lines, words, and characters in
one or more input files. Its usage is:
@example
wc @r{[}-lwc@r{]} @r{[} @var{files} @dots{} @r{]}
@end example
If no files are specified on the command line, @code{wc} reads its standard
input. If there are multiple files, it will also print total counts for all
the files. The options and their meanings are:
@table @code
@item -l
Only count lines.
@item -w
Only count words.
A ``word'' is a contiguous sequence of non-whitespace characters, separated
by spaces and/or tabs. Happily, this is the normal way @code{awk} separates
fields in its input data.
@item -c
Only count characters.
@end table
Implementing @code{wc} in @code{awk} is particularly elegant, since
@code{awk} does a lot of the work for us; it splits lines into words (i.e.@:
fields) and counts them, it counts lines (i.e.@: records) for us, and it can
easily tell us how long a line is.
This version uses the @code{getopt} library function
(@pxref{Getopt Function, ,Processing Command Line Options}),
and the file transition functions
(@pxref{Filetrans Function, ,Noting Data File Boundaries}).
This version has one major difference from traditional versions of @code{wc}.
Our version always prints the counts in the order lines, words,
and characters. Traditional versions note the order of the @samp{-l},
@samp{-w}, and @samp{-c} options on the command line, and print the counts
in that order.
The @code{BEGIN} rule does the argument processing.
The variable @code{print_total} will
be true if more than one file was named on the command line.
@findex wc.awk
@example
@c @group
@c file eg/prog/wc.awk
# wc.awk --- count lines, words, characters
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
# Options:
# -l only count lines
# -w only count words
# -c only count characters
#
# Default is to count lines, words, characters
BEGIN @{
# let getopt print a message about
# invalid options. we ignore them
while ((c = getopt(ARGC, ARGV, "lwc")) != -1) @{
if (c == "l")
do_lines = 1
else if (c == "w")
do_words = 1
else if (c == "c")
do_chars = 1
@}
for (i = 1; i < Optind; i++)
ARGV[i] = ""
# if no options, do all
if (! do_lines && ! do_words && ! do_chars)
do_lines = do_words = do_chars = 1
print_total = (ARGC - i > 2)
@}
@c endfile
@c @end group
@end example
The @code{beginfile} function is simple; it just resets the counts of lines,
words, and characters to zero, and saves the current file name in
@code{fname}.
The @code{endfile} function adds the current file's numbers to the running
totals of lines, words, and characters. It then prints out those numbers
for the file that was just read. It relies on @code{beginfile} to reset the
numbers for the following data file.
@example
@c @group
@c file eg/prog/wc.awk
function beginfile(file)
@{
chars = lines = words = 0
fname = FILENAME
@}
function endfile(file)
@{
tchars += chars
tlines += lines
twords += words
@group
if (do_lines)
printf "\t%d", lines
@end group
if (do_words)
printf "\t%d", words
if (do_chars)
printf "\t%d", chars
printf "\t%s\n", fname
@}
@c endfile
@c @end group
@end example
There is one rule that is executed for each line. It adds the length of the
record to @code{chars}. It has to add one, since the newline character
separating records (the value of @code{RS}) is not part of the record
itself. @code{lines} is incremented for each line read, and @code{words} is
incremented by the value of @code{NF}, the number of ``words'' on this
line.@footnote{Examine the code in
@ref{Filetrans Function, ,Noting Data File Boundaries}.
Why must @code{wc} use a separate @code{lines} variable, instead of using
the value of @code{FNR} in @code{endfile}?}
Finally, the @code{END} rule simply prints the totals for all the files.
@example
@c @group
@c file eg/prog/wc.awk
# do per line
@{
chars += length($0) + 1 # get newline
lines++
words += NF
@}
END @{
if (print_total) @{
if (do_lines)
printf "\t%d", tlines
if (do_words)
printf "\t%d", twords
if (do_chars)
printf "\t%d", tchars
print "\ttotal"
@}
@}
@c endfile
@c @end group
@end example
@node Miscellaneous Programs, , Clones, Sample Programs
@section A Grab Bag of @code{awk} Programs
This section is a large ``grab bag'' of miscellaneous programs.
We hope you find them both interesting and enjoyable.
@menu
* Dupword Program:: Finding duplicated words in a document.
* Alarm Program:: An alarm clock.
* Translate Program:: A program similar to the @code{tr} utility.
* Labels Program:: Printing mailing labels.
* Word Sorting:: A program to produce a word usage count.
* History Sorting:: Eliminating duplicate entries from a history
file.
* Extract Program:: Pulling out programs from Texinfo source
files.
* Simple Sed:: A Simple Stream Editor.
* Igawk Program:: A wrapper for @code{awk} that includes files.
@end menu
@node Dupword Program, Alarm Program, Miscellaneous Programs, Miscellaneous Programs
@subsection Finding Duplicated Words in a Document
A common error when writing large amounts of prose is to accidentally
duplicate words. Often you will see this in text as something like ``the
the program does the following @dots{}.'' When the text is on-line, often
the duplicated words occur at the end of one line and the beginning of
another, making them very difficult to spot.
@c as here!
This program, @file{dupword.awk}, scans through a file one line at a time,
and looks for adjacent occurrences of the same word. It also saves the last
word on a line (in the variable @code{prev}) for comparison with the first
word on the next line.
The first two statements make sure that the line is all lower-case, so that,
for example,
``The'' and ``the'' compare equal to each other. The second statement
removes all non-alphanumeric and non-whitespace characters from the line, so
that punctuation does not affect the comparison either. This sometimes
leads to reports of duplicated words that really are different, but this is
unusual.
@c FIXME: add check for $i != ""
@findex dupword.awk
@example
@group
@c file eg/prog/dupword.awk
# dupword --- find duplicate words in text
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# December 1991
@{
$0 = tolower($0)
gsub(/[^A-Za-z0-9 \t]/, "");
if ($1 == prev)
printf("%s:%d: duplicate %s\n",
FILENAME, FNR, $1)
for (i = 2; i <= NF; i++)
if ($i == $(i-1))
printf("%s:%d: duplicate %s\n",
FILENAME, FNR, $i)
prev = $NF
@}
@c endfile
@end group
@end example
@node Alarm Program, Translate Program, Dupword Program, Miscellaneous Programs
@subsection An Alarm Clock Program
The following program is a simple ``alarm clock'' program.
You give it a time of day, and an optional message. At the given time,
it prints the message on the standard output. In addition, you can give it
the number of times to repeat the message, and also a delay between
repetitions.
This program uses the @code{gettimeofday} function from
@ref{Gettimeofday Function, ,Managing the Time of Day}.
All the work is done in the @code{BEGIN} rule. The first part is argument
checking and setting of defaults; the delay, the count, and the message to
print. If the user supplied a message, but it does not contain the ASCII BEL
character (known as the ``alert'' character, @samp{\a}), then it is added to
the message. (On many systems, printing the ASCII BEL generates some sort
of audible alert. Thus, when the alarm goes off, the system calls attention
to itself, in case the user is not looking at their computer or terminal.)
@findex alarm.awk
@example
@c @group
@c file eg/prog/alarm.awk
# alarm --- set an alarm
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
# usage: alarm time [ "message" [ count [ delay ] ] ]
BEGIN \
@{
# Initial argument sanity checking
usage1 = "usage: alarm time ['message' [count [delay]]]"
usage2 = sprintf("\t(%s) time ::= hh:mm", ARGV[1])
if (ARGC < 2) @{
print usage > "/dev/stderr"
exit 1
@} else if (ARGC == 5) @{
delay = ARGV[4] + 0
count = ARGV[3] + 0
message = ARGV[2]
@} else if (ARGC == 4) @{
count = ARGV[3] + 0
message = ARGV[2]
@} else if (ARGC == 3) @{
message = ARGV[2]
@} else if (ARGV[1] !~ /[0-9]?[0-9]:[0-9][0-9]/) @{
print usage1 > "/dev/stderr"
print usage2 > "/dev/stderr"
exit 1
@}
# set defaults for once we reach the desired time
if (delay == 0)
delay = 180 # 3 minutes
if (count == 0)
count = 5
@group
if (message == "")
message = sprintf("\aIt is now %s!\a", ARGV[1])
else if (index(message, "\a") == 0)
message = "\a" message "\a"
@end group
@c endfile
@end example
The next section of code turns the alarm time into hours and minutes,
and converts it if necessary to a 24-hour clock. Then it turns that
time into a count of the seconds since midnight. Next it turns the current
time into a count of seconds since midnight. The difference between the two
is how long to wait before setting off the alarm.
@example
@c @group
@c file eg/prog/alarm.awk
# split up dest time
split(ARGV[1], atime, ":")
hour = atime[1] + 0 # force numeric
minute = atime[2] + 0 # force numeric
# get current broken down time
gettimeofday(now)
# if time given is 12-hour hours and it's after that
# hour, e.g., `alarm 5:30' at 9 a.m. means 5:30 p.m.,
# then add 12 to real hour
if (hour < 12 && now["hour"] > hour)
hour += 12
# set target time in seconds since midnight
target = (hour * 60 * 60) + (minute * 60)
# get current time in seconds since midnight
current = (now["hour"] * 60 * 60) + \
(now["minute"] * 60) + now["second"]
# how long to sleep for
naptime = target - current
if (naptime <= 0) @{
print "time is in the past!" > "/dev/stderr"
exit 1
@}
@c endfile
@c @end group
@end example
Finally, the program uses the @code{system} function
(@pxref{I/O Functions, ,Built-in Functions for Input/Output})
to call the @code{sleep} utility. The @code{sleep} utility simply pauses
for the given number of seconds. If the exit status is not zero,
the program assumes that @code{sleep} was interrupted, and exits. If
@code{sleep} exited with an OK status (zero), then the program prints the
message in a loop, again using @code{sleep} to delay for however many
seconds are necessary.
@example
@c @group
@c file eg/prog/alarm.awk
# zzzzzz..... go away if interrupted
if (system(sprintf("sleep %d", naptime)) != 0)
exit 1
# time to notify!
command = sprintf("sleep %d", delay)
for (i = 1; i <= count; i++) @{
print message
# if sleep command interrupted, go away
if (system(command) != 0)
break
@}
exit 0
@}
@c endfile
@c @end group
@end example
@node Translate Program, Labels Program, Alarm Program, Miscellaneous Programs
@subsection Transliterating Characters
The system @code{tr} utility transliterates characters. For example, it is
often used to map upper-case letters into lower-case, for further
processing.
@example
@var{generate data} | tr '[A-Z]' '[a-z]' | @var{process data} @dots{}
@end example
You give @code{tr} two lists of characters enclosed in square brackets.
Usually, the lists are quoted to keep the shell from attempting to do a
filename expansion.@footnote{On older, non-POSIX systems, @code{tr} often
does not require that the lists be enclosed in square brackets and quoted.
This is a feature.} When processing the input, the
first character in the first list is replaced with the first character in the
second list, the second character in the first list is replaced with the
second character in the second list, and so on.
If there are more characters in the ``from'' list than in the ``to'' list,
the last character of the ``to'' list is used for the remaining characters
in the ``from'' list.
Some time ago,
@c early or mid-1989!
a user proposed to us that we add a transliteration function to @code{gawk}.
Being opposed to ``creeping featurism,'' I wrote the following program to
prove that character transliteration could be done with a user-level
function. This program is not as complete as the system @code{tr} utility,
but it will do most of the job.
The @code{translate} program demonstrates one of the few weaknesses of
standard
@code{awk}: dealing with individual characters is very painful, requiring
repeated use of the @code{substr}, @code{index}, and @code{gsub} built-in
functions
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).@footnote{This
program was written before @code{gawk} acquired the ability to
split each character in a string into separate array elements.
How might this ability simplify the program?}
There are two functions. The first, @code{stranslate}, takes three
arguments.
@table @code
@item from
A list of characters to translate from.
@item to
A list of characters to translate to.
@item target
The string to do the translation on.
@end table
Associative arrays make the translation part fairly easy. @code{t_ar} holds
the ``to'' characters, indexed by the ``from'' characters. Then a simple
loop goes through @code{from}, one character at a time. For each character
in @code{from}, if the character appears in @code{target}, @code{gsub}
is used to change it to the corresponding @code{to} character.
The @code{translate} function simply calls @code{stranslate} using @code{$0}
as the target. The main program sets two global variables, @code{FROM} and
@code{TO}, from the command line, and then changes @code{ARGV} so that
@code{awk} will read from the standard input.
Finally, the processing rule simply calls @code{translate} for each record.
@findex translate.awk
@example
@c @group
@c file eg/prog/translate.awk
# translate --- do tr like stuff
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# August 1989
# bugs: does not handle things like: tr A-Z a-z, it has
# to be spelled out. However, if `to' is shorter than `from',
# the last character in `to' is used for the rest of `from'.
function stranslate(from, to, target, lf, lt, t_ar, i, c)
@{
lf = length(from)
lt = length(to)
for (i = 1; i <= lt; i++)
t_ar[substr(from, i, 1)] = substr(to, i, 1)
if (lt < lf)
for (; i <= lf; i++)
t_ar[substr(from, i, 1)] = substr(to, lt, 1)
for (i = 1; i <= lf; i++) @{
c = substr(from, i, 1)
if (index(target, c) > 0)
gsub(c, t_ar[c], target)
@}
return target
@}
@group
function translate(from, to)
@{
return $0 = stranslate(from, to, $0)
@}
@end group
# main program
BEGIN @{
if (ARGC < 3) @{
print "usage: translate from to" > "/dev/stderr"
exit
@}
FROM = ARGV[1]
TO = ARGV[2]
ARGC = 2
ARGV[1] = "-"
@}
@{
translate(FROM, TO)
print
@}
@c endfile
@c @end group
@end example
While it is possible to do character transliteration in a user-level
function, it is not necessarily efficient, and we started to consider adding
a built-in function. However, shortly after writing this program, we learned
that the System V Release 4 @code{awk} had added the @code{toupper} and
@code{tolower} functions. These functions handle the vast majority of the
cases where character transliteration is necessary, and so we chose to
simply add those functions to @code{gawk} as well, and then leave well
enough alone.
An obvious improvement to this program would be to set up the
@code{t_ar} array only once, in a @code{BEGIN} rule. However, this
assumes that the ``from'' and ``to'' lists
will never change throughout the lifetime of the program.
@node Labels Program, Word Sorting, Translate Program, Miscellaneous Programs
@subsection Printing Mailing Labels
Here is a ``real world''@footnote{``Real world'' is defined as
``a program actually used to get something done.''}
program. This script reads lists of names and
addresses, and generates mailing labels. Each page of labels has 20 labels
on it, two across and ten down. The addresses are guaranteed to be no more
than five lines of data. Each address is separated from the next by a blank
line.
The basic idea is to read 20 labels worth of data. Each line of each label
is stored in the @code{line} array. The single rule takes care of filling
the @code{line} array and printing the page when 20 labels have been read.
The @code{BEGIN} rule simply sets @code{RS} to the empty string, so that
@code{awk} will split records at blank lines
(@pxref{Records, ,How Input is Split into Records}).
It sets @code{MAXLINES} to 100, since @code{MAXLINE} is the maximum number
of lines on the page (20 * 5 = 100).
Most of the work is done in the @code{printpage} function.
The label lines are stored sequentially in the @code{line} array. But they
have to be printed horizontally; @code{line[1]} next to @code{line[6]},
@code{line[2]} next to @code{line[7]}, and so on. Two loops are used to
accomplish this. The outer loop, controlled by @code{i}, steps through
every 10 lines of data; this is each row of labels. The inner loop,
controlled by @code{j}, goes through the lines within the row.
As @code{j} goes from zero to four, @samp{i+j} is the @code{j}'th line in
the row, and @samp{i+j+5} is the entry next to it. The output ends up
looking something like this:
@example
line 1 line 6
line 2 line 7
line 3 line 8
line 4 line 9
line 5 line 10
@end example
As a final note, at lines 21 and 61, an extra blank line is printed, to keep
the output lined up on the labels. This is dependent on the particular
brand of labels in use when the program was written. You will also note
that there are two blank lines at the top and two blank lines at the bottom.
The @code{END} rule arranges to flush the final page of labels; there may
not have been an even multiple of 20 labels in the data.
@findex labels.awk
@example
@c @group
@c file eg/prog/labels.awk
# labels.awk
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# June 1992
# Program to print labels. Each label is 5 lines of data
# that may have blank lines. The label sheets have 2
# blank lines at the top and 2 at the bottom.
BEGIN @{ RS = "" ; MAXLINES = 100 @}
function printpage( i, j)
@{
if (Nlines <= 0)
return
printf "\n\n" # header
for (i = 1; i <= Nlines; i += 10) @{
if (i == 21 || i == 61)
print ""
for (j = 0; j < 5; j++) @{
if (i + j > MAXLINES)
break
printf " %-41s %s\n", line[i+j], line[i+j+5]
@}
print ""
@}
printf "\n\n" # footer
for (i in line)
line[i] = ""
@}
# main rule
@{
if (Count >= 20) @{
printpage()
Count = 0
Nlines = 0
@}
n = split($0, a, "\n")
for (i = 1; i <= n; i++)
line[++Nlines] = a[i]
for (; i <= 5; i++)
line[++Nlines] = ""
Count++
@}
END \
@{
printpage()
@}
@c endfile
@c @end group
@end example
@node Word Sorting, History Sorting, Labels Program, Miscellaneous Programs
@subsection Generating Word Usage Counts
The following @code{awk} program prints
the number of occurrences of each word in its input. It illustrates the
associative nature of @code{awk} arrays by using strings as subscripts. It
also demonstrates the @samp{for @var{x} in @var{array}} construction.
Finally, it shows how @code{awk} can be used in conjunction with other
utility programs to do a useful task of some complexity with a minimum of
effort. Some explanations follow the program listing.
@example
awk '
# Print list of word frequencies
@{
for (i = 1; i <= NF; i++)
freq[$i]++
@}
END @{
for (word in freq)
printf "%s\t%d\n", word, freq[word]
@}'
@end example
The first thing to notice about this program is that it has two rules. The
first rule, because it has an empty pattern, is executed on every line of
the input. It uses @code{awk}'s field-accessing mechanism
(@pxref{Fields, ,Examining Fields}) to pick out the individual words from
the line, and the built-in variable @code{NF} (@pxref{Built-in Variables})
to know how many fields are available.
For each input word, an element of the array @code{freq} is incremented to
reflect that the word has been seen an additional time.
The second rule, because it has the pattern @code{END}, is not executed
until the input has been exhausted. It prints out the contents of the
@code{freq} table that has been built up inside the first action.
This program has several problems that would prevent it from being
useful by itself on real text files:
@itemize @bullet
@item
Words are detected using the @code{awk} convention that fields are
separated by whitespace and that other characters in the input (except
newlines) don't have any special meaning to @code{awk}. This means that
punctuation characters count as part of words.
@item
The @code{awk} language considers upper- and lower-case characters to be
distinct. Therefore, @samp{bartender} and @samp{Bartender} are not treated
as the same word. This is undesirable since, in normal text, words
are capitalized if they begin sentences, and a frequency analyzer should not
be sensitive to capitalization.
@item
The output does not come out in any useful order. You're more likely to be
interested in which words occur most frequently, or having an alphabetized
table of how frequently each word occurs.
@end itemize
The way to solve these problems is to use some of the more advanced
features of the @code{awk} language. First, we use @code{tolower} to remove
case distinctions. Next, we use @code{gsub} to remove punctuation
characters. Finally, we use the system @code{sort} utility to process the
output of the @code{awk} script. Here is the new version of
the program:
@findex wordfreq.sh
@example
@c file eg/prog/wordfreq.awk
# Print list of word frequencies
@{
$0 = tolower($0) # remove case distinctions
gsub(/[^a-z0-9_ \t]/, "", $0) # remove punctuation
for (i = 1; i <= NF; i++)
freq[$i]++
@}
@c endfile
END @{
for (word in freq)
printf "%s\t%d\n", word, freq[word]
@}
@end example
Assuming we have saved this program in a file named @file{wordfreq.awk},
and that the data is in @file{file1}, the following pipeline
@example
awk -f wordfreq.awk file1 | sort +1 -nr
@end example
@noindent
produces a table of the words appearing in @file{file1} in order of
decreasing frequency.
The @code{awk} program suitably massages the data and produces a word
frequency table, which is not ordered.
The @code{awk} script's output is then sorted by the @code{sort} utility and
printed on the terminal. The options given to @code{sort} in this example
specify to sort using the second field of each input line (skipping one field),
that the sort keys should be treated as numeric quantities (otherwise
@samp{15} would come before @samp{5}), and that the sorting should be done
in descending (reverse) order.
We could have even done the @code{sort} from within the program, by
changing the @code{END} action to:
@example
@c file eg/prog/wordfreq.awk
END @{
sort = "sort +1 -nr"
for (word in freq)
printf "%s\t%d\n", word, freq[word] | sort
close(sort)
@}
@c endfile
@end example
You would have to use this way of sorting on systems that do not
have true pipes.
See the general operating system documentation for more information on how
to use the @code{sort} program.
@node History Sorting, Extract Program, Word Sorting, Miscellaneous Programs
@subsection Removing Duplicates from Unsorted Text
The @code{uniq} program
(@pxref{Uniq Program, ,Printing Non-duplicated Lines of Text}),
removes duplicate lines from @emph{sorted} data.
Suppose, however, you need to remove duplicate lines from a data file, but
that you wish to preserve the order the lines are in? A good example of
this might be a shell history file. The history file keeps a copy of all
the commands you have entered, and it is not unusual to repeat a command
several times in a row. Occasionally you might wish to compact the history
by removing duplicate entries. Yet it is desirable to maintain the order
of the original commands.
This simple program does the job. It uses two arrays. The @code{data}
array is indexed by the text of each line.
For each line, @code{data[$0]} is incremented.
If a particular line has not
been seen before, then @code{data[$0]} will be zero.
In that case, the text of the line is stored in @code{lines[count]}.
Each element of @code{lines} is a unique command, and the indices of
@code{lines} indicate the order in which those lines were encountered.
The @code{END} rule simply prints out the lines, in order.
@cindex Rakitzis, Byron
@findex histsort.awk
@example
@group
@c file eg/prog/histsort.awk
# histsort.awk --- compact a shell history file
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
# Thanks to Byron Rakitzis for the general idea
@{
if (data[$0]++ == 0)
lines[++count] = $0
@}
END @{
for (i = 1; i <= count; i++)
print lines[i]
@}
@c endfile
@end group
@end example
This program also provides a foundation for generating other useful
information. For example, using the following @code{print} satement in the
@code{END} rule would indicate how often a particular command was used.
@example
print data[lines[i]], lines[i]
@end example
This works because @code{data[$0]} was incremented each time a line was
seen.
@node Extract Program, Simple Sed, History Sorting, Miscellaneous Programs
@subsection Extracting Programs from Texinfo Source Files
@iftex
Both this chapter and the previous chapter
(@ref{Library Functions, ,A Library of @code{awk} Functions}),
present a large number of @code{awk} programs.
@end iftex
@ifinfo
The nodes
@ref{Library Functions, ,A Library of @code{awk} Functions},
and @ref{Sample Programs, ,Practical @code{awk} Programs},
are the top level nodes for a large number of @code{awk} programs.
@end ifinfo
If you wish to experiment with these programs, it is tedious to have to type
them in by hand. Here we present a program that can extract parts of a
Texinfo input file into separate files.
This @value{DOCUMENT} is written in Texinfo, the GNU project's document
formatting language. A single Texinfo source file can be used to produce both
printed and on-line documentation.
@iftex
Texinfo is fully documented in @cite{Texinfo---The GNU Documentation Format},
available from the Free Software Foundation.
@end iftex
@ifinfo
The Texinfo language is described fully, starting with
@ref{Top, , Introduction, texi, Texinfo---The GNU Documentation Format}.
@end ifinfo
For our purposes, it is enough to know three things about Texinfo input
files.
@itemize @bullet
@item
The ``at'' symbol, @samp{@@}, is special in Texinfo, much like @samp{\} in C
or @code{awk}. Literal @samp{@@} symbols are represented in Texinfo source
files as @samp{@@@@}.
@item
Comments start with either @samp{@@c} or @samp{@@comment}.
The file extraction program will work by using special comments that start
at the beginning of a line.
@item
Example text that should not be split across a page boundary is bracketed
between lines containing @samp{@@group} and @samp{@@end group} commands.
@end itemize
The following program, @file{extract.awk}, reads through a Texinfo source
file, and does two things, based on the special comments.
Upon seeing @samp{@w{@@c system @dots{}}},
it runs a command, by extracting the command text from the
control line and passing it on to the @code{system} function
(@pxref{I/O Functions, ,Built-in Functions for Input/Output}).
Upon seeing @samp{@@c file @var{filename}}, each subsequent line is sent to
the file @var{filename}, until @samp{@@c endfile} is encountered.
The rules in @file{extract.awk} will match either @samp{@@c} or
@samp{@@comment} by letting the @samp{omment} part be optional.
Lines containing @samp{@@group} and @samp{@@end group} are simply removed.
@file{extract.awk} uses the @code{join} library function
(@pxref{Join Function, ,Merging an Array Into a String}).
The example programs in the on-line Texinfo source for @cite{@value{TITLE}}
(@file{gawk.texi}) have all been bracketed inside @samp{file},
and @samp{endfile} lines. The @code{gawk} distribution uses a copy of
@file{extract.awk} to extract the sample
programs and install many of them in a standard directory, where
@code{gawk} can find them.
@file{extract.awk} begins by setting @code{IGNORECASE} to one, so that
mixed upper-case and lower-case letters in the directives won't matter.
The first rule handles calling @code{system}, checking that a command was
given (@code{NF} is at least three), and also checking that the command
exited with a zero exit status, signifying OK.
@findex extract.awk
@example
@c @group
@c file eg/prog/extract.awk
# extract.awk --- extract files and run programs
# from texinfo files
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# May 1993
BEGIN @{ IGNORECASE = 1 @}
@group
/^@@c(omment)?[ \t]+system/ \
@{
if (NF < 3) @{
e = (FILENAME ":" FNR)
e = (e ": badly formed `system' line")
print e > "/dev/stderr"
next
@}
$1 = ""
$2 = ""
stat = system($0)
if (stat != 0) @{
e = (FILENAME ":" FNR)
e = (e ": warning: system returned " stat)
print e > "/dev/stderr"
@}
@}
@end group
@c endfile
@end example
@noindent
The variable @code{e} is used so that the function
fits nicely on the
@iftex
page.
@end iftex
@ifinfo
screen.
@end ifinfo
The second rule handles moving data into files. It verifies that a file
name was given in the directive. If the file named is not the current file,
then the current file is closed. This means that an @samp{@@c endfile} was
not given for that file. (We should probably print a diagnostic in this
case, although at the moment we do not.)
The @samp{for} loop does the work. It reads lines using @code{getline}
(@pxref{Getline, ,Explicit Input with @code{getline}}).
For an unexpected end of file, it calls the @code{@w{unexpected_eof}}
function. If the line is an ``endfile'' line, then it breaks out of
the loop.
If the line is an @samp{@@group} or @samp{@@end group} line, then it
ignores it, and goes on to the next line.
Most of the work is in the following few lines. If the line has no @samp{@@}
symbols, it can be printed directly. Otherwise, each leading @samp{@@} must be
stripped off.
To remove the @samp{@@} symbols, the line is split into separate elements of
the array @code{a}, using the @code{split} function
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
Each element of @code{a} that is empty indicates two successive @samp{@@}
symbols in the original line. For each two empty elements (@samp{@@@@} in
the original file), we have to add back in a single @samp{@@} symbol.
When the processing of the array is finished, @code{join} is called with the
value of @code{SUBSEP}, to rejoin the pieces back into a single
line. That line is then printed to the output file.
@example
@c @group
@c file eg/prog/extract.awk
@group
/^@@c(omment)?[ \t]+file/ \
@{
if (NF != 3) @{
e = (FILENAME ":" FNR ": badly formed `file' line")
print e > "/dev/stderr"
next
@}
@end group
if ($3 != curfile) @{
if (curfile != "")
close(curfile)
curfile = $3
@}
for (;;) @{
if ((getline line) <= 0)
unexpected_eof()
if (line ~ /^@@c(omment)?[ \t]+endfile/)
break
else if (line ~ /^@@(end[ \t]+)?group/)
continue
if (index(line, "@@") == 0) @{
print line > curfile
continue
@}
n = split(line, a, "@@")
@group
# if a[1] == "", means leading @@,
# don't add one back in.
@end group
for (i = 2; i <= n; i++) @{
if (a[i] == "") @{ # was an @@@@
a[i] = "@@"
if (a[i+1] == "")
i++
@}
@}
print join(a, 1, n, SUBSEP) > curfile
@}
@}
@c endfile
@c @end group
@end example
An important thing to note is the use of the @samp{>} redirection.
Output done with @samp{>} only opens the file once; it stays open and
subsequent output is appended to the file
(@pxref{Redirection, , Redirecting Output of @code{print} and @code{printf}}).
This allows us to easily mix program text and explanatory prose for the same
sample source file (as has been done here!) without any hassle. The file is
only closed when a new data file name is encountered, or at the end of the
input file.
Finally, the function @code{@w{unexpected_eof}} prints an appropriate
error message and then exits.
The @code{END} rule handles the final cleanup, closing the open file.
@example
@c file eg/prog/extract.awk
@group
function unexpected_eof()
@{
printf("%s:%d: unexpected EOF or error\n", \
FILENAME, FNR) > "/dev/stderr"
exit 1
@}
@end group
END @{
if (curfile)
close(curfile)
@}
@c endfile
@end example
@node Simple Sed, Igawk Program, Extract Program, Miscellaneous Programs
@subsection A Simple Stream Editor
@cindex @code{sed} utility
The @code{sed} utility is a ``stream editor,'' a program that reads a
stream of data, makes changes to it, and passes the modified data on.
It is often used to make global changes to a large file, or to a stream
of data generated by a pipeline of commands.
While @code{sed} is a complicated program in its own right, its most common
use is to perform global substitutions in the middle of a pipeline:
@example
command1 < orig.data | sed 's/old/new/g' | command2 > result
@end example
Here, the @samp{s/old/new/g} tells @code{sed} to look for the regexp
@samp{old} on each input line, and replace it with the text @samp{new},
globally (i.e.@: all the occurrences on a line). This is similar to
@code{awk}'s @code{gsub} function
(@pxref{String Functions, , Built-in Functions for String Manipulation}).
The following program, @file{awksed.awk}, accepts at least two command line
arguments; the pattern to look for and the text to replace it with. Any
additional arguments are treated as data file names to process. If none
are provided, the standard input is used.
@cindex Brennan, Michael
@cindex @code{awksed}
@cindex simple stream editor
@cindex stream editor, simple
@example
@c @group
@c file eg/prog/awksed.awk
# awksed.awk --- do s/foo/bar/g using just print
# Thanks to Michael Brennan for the idea
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# August 1995
@group
function usage()
@{
print "usage: awksed pat repl [files...]" > "/dev/stderr"
exit 1
@}
@end group
BEGIN @{
# validate arguments
if (ARGC < 3)
usage()
RS = ARGV[1]
ORS = ARGV[2]
# don't use arguments as files
ARGV[1] = ARGV[2] = ""
@}
# look ma, no hands!
@{
if (RT == "")
printf "%s", $0
else
print
@}
@c endfile
@c @end group
@end example
The program relies on @code{gawk}'s ability to have @code{RS} be a regexp
and on the setting of @code{RT} to the actual text that terminated the
record (@pxref{Records, ,How Input is Split into Records}).
The idea is to have @code{RS} be the pattern to look for. @code{gawk}
will automatically set @code{$0} to the text between matches of the pattern.
This is text that we wish to keep, unmodified. Then, by setting @code{ORS}
to the replacement text, a simple @code{print} statement will output the
text we wish to keep, followed by the replacement text.
There is one wrinkle to this scheme, which is what to do if the last record
doesn't end with text that matches @code{RS}? Using a @code{print}
statement unconditionally prints the replacement text, which is not correct.
However, if the file did not end in text that matches @code{RS}, @code{RT}
will be set to the null string. In this case, we can print @code{$0} using
@code{printf}
(@pxref{Printf, ,Using @code{printf} Statements for Fancier Printing}).
The @code{BEGIN} rule handles the setup, checking for the right number
of arguments, and calling @code{usage} if there is a problem. Then it sets
@code{RS} and @code{ORS} from the command line arguments, and sets
@code{ARGV[1]} and @code{ARGV[2]} to the null string, so that they will
not be treated as file names
(@pxref{ARGC and ARGV, , Using @code{ARGC} and @code{ARGV}}).
The @code{usage} function prints an error message and exits.
Finally, the single rule handles the printing scheme outlined above,
using @code{print} or @code{printf} as appropriate, depending upon the
value of @code{RT}.
@ignore
Exercise, compare the performance of this version with the more
straightforward:
BEGIN {
pat = ARGV[1]
repl = ARGV[2]
ARGV[1] = ARGV[2] = ""
}
{ gsub(pat, repl); print }
Exercise: what are the advantages and disadvantages of this version vs. sed?
Advantage: egrep regexps
speed (?)
Disadvantage: no & in replacement text
Others?
@end ignore
@node Igawk Program, , Simple Sed, Miscellaneous Programs
@subsection An Easy Way to Use Library Functions
Using library functions in @code{awk} can be very beneficial. It
encourages code re-use and the writing of general functions. Programs are
smaller, and therefore clearer.
However, using library functions is only easy when writing @code{awk}
programs; it is painful when running them, requiring multiple @samp{-f}
options. If @code{gawk} is unavailable, then so too is the @code{AWKPATH}
environment variable and the ability to put @code{awk} functions into a
library directory (@pxref{Options, ,Command Line Options}).
It would be nice to be able to write programs like so:
@example
# library functions
@@include getopt.awk
@@include join.awk
@dots{}
# main program
BEGIN @{
while ((c = getopt(ARGC, ARGV, "a:b:cde")) != -1)
@dots{}
@dots{}
@}
@end example
The following program, @file{igawk.sh}, provides this service.
It simulates @code{gawk}'s searching of the @code{AWKPATH} variable,
and also allows @dfn{nested} includes; i.e.@: a file that has been included
with @samp{@@include} can contain further @samp{@@include} statements.
@code{igawk} will make an effort to only include files once, so that nested
includes don't accidentally include a library function twice.
@code{igawk} should behave externally just like @code{gawk}. This means it
should accept all of @code{gawk}'s command line arguments, including the
ability to have multiple source files specified via @samp{-f}, and the
ability to mix command line and library source files.
The program is written using the POSIX Shell (@code{sh}) command language.
The way the program works is as follows:
@enumerate
@item
Loop through the arguments, saving anything that doesn't represent
@code{awk} source code for later, when the expanded program is run.
@item
For any arguments that do represent @code{awk} text, put the arguments into
a temporary file that will be expanded. There are two cases.
@enumerate a
@item
Literal text, provided with @samp{--source} or @samp{--source=}. This
text is just echoed directly. The @code{echo} program will automatically
supply a trailing newline.
@item
File names provided with @samp{-f}. We use a neat trick, and echo
@samp{@@include @var{filename}} into the temporary file. Since the file
inclusion program will work the way @code{gawk} does, this will get the text
of the file included into the program at the correct point.
@end enumerate
@item
Run an @code{awk} program (naturally) over the temporary file to expand
@samp{@@include} statements. The expanded program is placed in a second
temporary file.
@item
Run the expanded program with @code{gawk} and any other original command line
arguments that the user supplied (such as the data file names).
@end enumerate
The initial part of the program turns on shell tracing if the first
argument was @samp{debug}. Otherwise, a shell @code{trap} statement
arranges to clean up any temporary files on program exit or upon an
interrupt.
@c 2e: For the temp file handling, go with Darrel's ig=${TMP:-/tmp}/igs.$$
@c 2e: or something as similar as possible.
The next part loops through all the command line arguments.
There are several cases of interest.
@table @code
@item --
This ends the arguments to @code{igawk}. Anything else should be passed on
to the user's @code{awk} program without being evaluated.
@item -W
This indicates that the next option is specific to @code{gawk}. To make
argument processing easier, the @samp{-W} is appended to the front of the
remaining arguments and the loop continues. (This is an @code{sh}
programming trick. Don't worry about it if you are not familiar with
@code{sh}.)
@item -v
@itemx -F
These are saved and passed on to @code{gawk}.
@item -f
@itemx --file
@itemx --file=
@itemx -Wfile=
The file name is saved to the temporary file @file{/tmp/ig.s.$$} with an
@samp{@@include} statement.
The @code{sed} utility is used to remove the leading option part of the
argument (e.g., @samp{--file=}).
@item --source
@itemx --source=
@itemx -Wsource=
The source text is echoed into @file{/tmp/ig.s.$$}.
@item --version
@itemx --version
@itemx -Wversion
@code{igawk} prints its version number, and runs @samp{gawk --version}
to get the @code{gawk} version information, and then exits.
@end table
If none of @samp{-f}, @samp{--file}, @samp{-Wfile}, @samp{--source},
or @samp{-Wsource}, were supplied, then the first non-option argument
should be the @code{awk} program. If there are no command line
arguments left, @code{igawk} prints an error message and exits.
Otherwise, the first argument is echoed into @file{/tmp/ig.s.$$}.
In any case, after the arguments have been processed,
@file{/tmp/ig.s.$$} contains the complete text of the original @code{awk}
program.
The @samp{$$} in @code{sh} represents the current process ID number.
It is often used in shell programs to generate unique temporary file
names. This allows multiple users to run @code{igawk} without worrying
that the temporary file names will clash.
@cindex @code{sed} utility
Here's the program:
@findex igawk.sh
@example
@c @group
@c file eg/prog/igawk.sh
#! /bin/sh
# igawk --- like gawk but do @@include processing
# Arnold Robbins, arnold@@gnu.ai.mit.edu, Public Domain
# July 1993
if [ "$1" = debug ]
then
set -x
shift
else
# cleanup on exit, hangup, interrupt, quit, termination
trap 'rm -f /tmp/ig.[se].$$' 0 1 2 3 15
fi
while [ $# -ne 0 ] # loop over arguments
do
case $1 in
--) shift; break;;
-W) shift
set -- -W"$@@"
continue;;
-[vF]) opts="$opts $1 '$2'"
shift;;
-[vF]*) opts="$opts '$1'" ;;
-f) echo @@include "$2" >> /tmp/ig.s.$$
shift;;
@group
-f*) f=`echo "$1" | sed 's/-f//'`
echo @@include "$f" >> /tmp/ig.s.$$ ;;
@end group
-?file=*) # -Wfile or --file
f=`echo "$1" | sed 's/-.file=//'`
echo @@include "$f" >> /tmp/ig.s.$$ ;;
-?file) # get arg, $2
echo @@include "$2" >> /tmp/ig.s.$$
shift;;
-?source=*) # -Wsource or --source
t=`echo "$1" | sed 's/-.source=//'`
echo "$t" >> /tmp/ig.s.$$ ;;
-?source) # get arg, $2
echo "$2" >> /tmp/ig.s.$$
shift;;
-?version)
echo igawk: version 1.0 1>&2
gawk --version
exit 0 ;;
-[W-]*) opts="$opts '$1'" ;;
*) break;;
esac
shift
done
if [ ! -s /tmp/ig.s.$$ ]
then
if [ -z "$1" ]
then
echo igawk: no program! 1>&2
exit 1
else
echo "$1" > /tmp/ig.s.$$
shift
fi
fi
# at this point, /tmp/ig.s.$$ has the program
@c endfile
@c @end group
@end example
The @code{awk} program to process @samp{@@include} directives reads through
the program, one line at a time using @code{getline}
(@pxref{Getline, ,Explicit Input with @code{getline}}).
The input file names and @samp{@@include} statements are managed using a
stack. As each @samp{@@include} is encountered, the current file name is
``pushed'' onto the stack, and the file named in the @samp{@@include}
directive becomes
the current file name. As each file is finished, the stack is ``popped,''
and the previous input file becomes the current input file again.
The process is started by making the original file the first one on the
stack.
The @code{pathto} function does the work of finding the full path to a
file. It simulates @code{gawk}'s behavior when searching the @code{AWKPATH}
environment variable
(@pxref{AWKPATH Variable, ,The @code{AWKPATH} Environment Variable}).
If a file name has a @samp{/} in it, no path search
is done. Otherwise, the file name is concatenated with the name of each
directory in the path, and an attempt is made to open the generated file
name. The only way in @code{awk} to test if a file can be read is to go
ahead and try to read it with @code{getline}; that is what @code{pathto}
does.@footnote{On some very old versions of @code{awk}, the test
@samp{getline junk < t} can loop forever if the file exists but is empty.
Caveat Emptor.}
If the file can be read, it is closed, and the file name is
returned.
@ignore
An alternative way to test for the file's existence would be to call
@samp{system("test -r " t)}, which uses the @code{test} utility to
see if the file exists and is readable. The disadvantage to this method
is that it requires creating an extra process, and can thus be slightly
slower.
@end ignore
@example
@c @group
@c file eg/prog/igawk.sh
gawk -- '
# process @@include directives
function pathto(file, i, t, junk)
@{
if (index(file, "/") != 0)
return file
for (i = 1; i <= ndirs; i++) @{
t = (pathlist[i] "/" file)
if ((getline junk < t) > 0) @{
# found it
close(t)
return t
@}
@}
return ""
@}
@c endfile
@c @end group
@end example
The main program is contained inside one @code{BEGIN} rule. The first thing it
does is set up the @code{pathlist} array that @code{pathto} uses. After
splitting the path on @samp{:}, null elements are replaced with @code{"."},
which represents the current directory.
@example
@group
@c file eg/prog/igawk.sh
BEGIN @{
path = ENVIRON["AWKPATH"]
ndirs = split(path, pathlist, ":")
for (i = 1; i <= ndirs; i++) @{
if (pathlist[i] == "")
pathlist[i] = "."
@}
@c endfile
@end group
@end example
The stack is initialized with @code{ARGV[1]}, which will be @file{/tmp/ig.s.$$}.
The main loop comes next. Input lines are read in succession. Lines that
do not start with @samp{@@include} are printed verbatim.
If the line does start with @samp{@@include}, the file name is in @code{$2}.
@code{pathto} is called to generate the full path. If it could not, then we
print an error message and continue.
The next thing to check is if the file has been included already. The
@code{processed} array is indexed by the full file name of each included
file, and it tracks this information for us. If the file has been
seen, a warning message is printed. Otherwise, the new file name is
pushed onto the stack and processing continues.
Finally, when @code{getline} encounters the end of the input file, the file
is closed and the stack is popped. When @code{stackptr} is less than zero,
the program is done.
@example
@c @group
@c file eg/prog/igawk.sh
stackptr = 0
input[stackptr] = ARGV[1] # ARGV[1] is first file
for (; stackptr >= 0; stackptr--) @{
while ((getline < input[stackptr]) > 0) @{
if (tolower($1) != "@@include") @{
print
continue
@}
fpath = pathto($2)
if (fpath == "") @{
printf("igawk:%s:%d: cannot find %s\n", \
input[stackptr], FNR, $2) > "/dev/stderr"
continue
@}
@group
if (! (fpath in processed)) @{
processed[fpath] = input[stackptr]
input[++stackptr] = fpath
@} else
print $2, "included in", input[stackptr], \
"already included in", \
processed[fpath] > "/dev/stderr"
@}
@end group
@group
close(input[stackptr])
@}
@}' /tmp/ig.s.$$ > /tmp/ig.e.$$
@end group
@c endfile
@c @end group
@end example
The last step is to call @code{gawk} with the expanded program and the original
options and command line arguments that the user supplied. @code{gawk}'s
exit status is passed back on to @code{igawk}'s calling program.
@c this causes more problems than it solves, so leave it out.
@ignore
The special file @file{/dev/null} is passed as a data file to @code{gawk}
to handle an interesting case. Suppose that the user's program only has
a @code{BEGIN} rule, and there are no data files to read. The program should exit without reading any data
files. However, suppose that an included library file defines an @code{END}
rule of its own. In this case, @code{gawk} will hang, reading standard
input. In order to avoid this, @file{/dev/null} is explicitly to the
command line. Reading from @file{/dev/null} always returns an immediate
end of file indication.
@c Hmm. Add /dev/null if $# is 0? Still messes up ARGV. Sigh.
@end ignore
@example
@c @group
@c file eg/prog/igawk.sh
eval gawk -f /tmp/ig.e.$$ $opts -- "$@@"
exit $?
@c endfile
@c @end group
@end example
This version of @code{igawk} represents my third attempt at this program.
There are three key simplifications that made the program work better.
@enumerate
@item
Using @samp{@@include} even for the files named with @samp{-f} makes building
the initial collected @code{awk} program much simpler; all the
@samp{@@include} processing can be done once.
@item
The @code{pathto} function doesn't try to save the line read with
@code{getline} when testing for the file's accessibility. Trying to save
this line for use with the main program complicates things considerably.
@c what problem does this engender though - exercise
@c answer, reading from "-" or /dev/stdin
@item
Using a @code{getline} loop in the @code{BEGIN} rule does it all in one
place. It is not necessary to call out to a separate loop for processing
nested @samp{@@include} statements.
@end enumerate
Also, this program illustrates that it is often worthwhile to combine
@code{sh} and @code{awk} programming together. You can usually accomplish
quite a lot, without having to resort to low-level programming in C or C++, and it
is frequently easier to do certain kinds of string and argument manipulation
using the shell than it is in @code{awk}.
Finally, @code{igawk} shows that it is not always necessary to add new
features to a program; they can often be layered on top. With @code{igawk},
there is no real reason to build @samp{@@include} processing into
@code{gawk} itself.
As an additional example of this, consider the idea of having two
files in a directory in the search path.
@table @file
@item default.awk
This file would contain a set of default library functions, such
as @code{getopt} and @code{assert}.
@item site.awk
This file would contain library functions that are specific to a site or
installation, i.e.@: locally developed functions.
Having a separate file allows @file{default.awk} to change with
new @code{gawk} releases, without requiring the system administrator to
update it each time by adding the local functions.
@end table
One user
@c Karl Berry, karl@ileaf.com, 10/95
suggested that @code{gawk} be modified to automatically read these files
upon startup. Instead, it would be very simple to modify @code{igawk}
to do this. Since @code{igawk} can process nested @samp{@@include}
directives, @file{default.awk} could simply contain @samp{@@include}
statements for the desired library functions.
@c Exercise: make this change
@node Language History, Gawk Summary, Sample Programs, Top
@chapter The Evolution of the @code{awk} Language
This @value{DOCUMENT} describes the GNU implementation of @code{awk}, which follows
the POSIX specification. Many @code{awk} users are only familiar
with the original @code{awk} implementation in Version 7 Unix.
(This implementation was the basis for @code{awk} in Berkeley Unix,
through 4.3--Reno. The 4.4 release of Berkeley Unix uses @code{gawk} 2.15.2
for its version of @code{awk}.) This chapter briefly describes the
evolution of the @code{awk} language, with cross references to other parts
of the @value{DOCUMENT} where you can find more information.
@menu
* V7/SVR3.1:: The major changes between V7 and System V
Release 3.1.
* SVR4:: Minor changes between System V Releases 3.1
and 4.
* POSIX:: New features from the POSIX standard.
* BTL:: New features from the Bell Laboratories
version of @code{awk}.
* POSIX/GNU:: The extensions in @code{gawk} not in POSIX
@code{awk}.
@end menu
@node V7/SVR3.1, SVR4, Language History, Language History
@section Major Changes between V7 and SVR3.1
The @code{awk} language evolved considerably between the release of
Version 7 Unix (1978) and the new version first made generally available in
System V Release 3.1 (1987). This section summarizes the changes, with
cross-references to further details.
@itemize @bullet
@item
The requirement for @samp{;} to separate rules on a line
(@pxref{Statements/Lines, ,@code{awk} Statements Versus Lines}).
@item
User-defined functions, and the @code{return} statement
(@pxref{User-defined, ,User-defined Functions}).
@item
The @code{delete} statement (@pxref{Delete, ,The @code{delete} Statement}).
@item
The @code{do}-@code{while} statement
(@pxref{Do Statement, ,The @code{do}-@code{while} Statement}).
@item
The built-in functions @code{atan2}, @code{cos}, @code{sin}, @code{rand} and
@code{srand} (@pxref{Numeric Functions, ,Numeric Built-in Functions}).
@item
The built-in functions @code{gsub}, @code{sub}, and @code{match}
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
@item
The built-in functions @code{close}, and @code{system}
(@pxref{I/O Functions, ,Built-in Functions for Input/Output}).
@item
The @code{ARGC}, @code{ARGV}, @code{FNR}, @code{RLENGTH}, @code{RSTART},
and @code{SUBSEP} built-in variables (@pxref{Built-in Variables}).
@item
The conditional expression using the ternary operator @samp{?:}
(@pxref{Conditional Exp, ,Conditional Expressions}).
@item
The exponentiation operator @samp{^}
(@pxref{Arithmetic Ops, ,Arithmetic Operators}) and its assignment operator
form @samp{^=} (@pxref{Assignment Ops, ,Assignment Expressions}).
@item
C-compatible operator precedence, which breaks some old @code{awk}
programs (@pxref{Precedence, ,Operator Precedence (How Operators Nest)}).
@item
Regexps as the value of @code{FS}
(@pxref{Field Separators, ,Specifying How Fields are Separated}), and as the
third argument to the @code{split} function
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
@item
Dynamic regexps as operands of the @samp{~} and @samp{!~} operators
(@pxref{Regexp Usage, ,How to Use Regular Expressions}).
@item
The escape sequences @samp{\b}, @samp{\f}, and @samp{\r}
(@pxref{Escape Sequences}).
(Some vendors have updated their old versions of @code{awk} to
recognize @samp{\r}, @samp{\b}, and @samp{\f}, but this is not
something you can rely on.)
@item
Redirection of input for the @code{getline} function
(@pxref{Getline, ,Explicit Input with @code{getline}}).
@item
Multiple @code{BEGIN} and @code{END} rules
(@pxref{BEGIN/END, ,The @code{BEGIN} and @code{END} Special Patterns}).
@item
Multi-dimensional arrays
(@pxref{Multi-dimensional, ,Multi-dimensional Arrays}).
@end itemize
@node SVR4, POSIX, V7/SVR3.1, Language History
@section Changes between SVR3.1 and SVR4
@cindex @code{awk} language, V.4 version
The System V Release 4 version of Unix @code{awk} added these features
(some of which originated in @code{gawk}):
@itemize @bullet
@item
The @code{ENVIRON} variable (@pxref{Built-in Variables}).
@item
Multiple @samp{-f} options on the command line
(@pxref{Options, ,Command Line Options}).
@item
The @samp{-v} option for assigning variables before program execution begins
(@pxref{Options, ,Command Line Options}).
@item
The @samp{--} option for terminating command line options.
@item
The @samp{\a}, @samp{\v}, and @samp{\x} escape sequences
(@pxref{Escape Sequences}).
@item
A defined return value for the @code{srand} built-in function
(@pxref{Numeric Functions, ,Numeric Built-in Functions}).
@item
The @code{toupper} and @code{tolower} built-in string functions
for case translation
(@pxref{String Functions, ,Built-in Functions for String Manipulation}).
@item
A cleaner specification for the @samp{%c} format-control letter in the
@code{printf} function
(@pxref{Control Letters, ,Format-Control Letters}).
@item
The ability to dynamically pass the field width and precision (@code{"%*.*d"})
in the argument list of the @code{printf} function
(@pxref{Control Letters, ,Format-Control Letters}).
@item
The use of regexp constants such as @code{/foo/} as expressions, where
they are equivalent to using the matching operator, as in @samp{$0 ~ /foo/}
(@pxref{Using Constant Regexps, ,Using Regular Expression Constants}).
@end itemize
@node POSIX, BTL, SVR4, Language History
@section Changes between SVR4 and POSIX @code{awk}
The POSIX Command Language and Utilities standard for @code{awk}
introduced the following changes into the language:
@itemize @bullet
@item
The use of @samp{-W} for implementation-specific options.
@item
The use of @code{CONVFMT} for controlling the conversion of numbers
to strings (@pxref{Conversion, ,Conversion of Strings and Numbers}).
@item
The concept of a numeric string, and tighter comparison rules to go
with it (@pxref{Typing and Comparison, ,Variable Typing and Comparison Expressions}).
@item
More complete documentation of many of the previously undocumented
features of the language.
@end itemize
The following common extensions are not permitted by the POSIX
standard:
@c IMPORTANT! Keep this list in sync with the one in node Options
@itemize @bullet
@item
@code{\x} escape sequences are not recognized
(@pxref{Escape Sequences}).
@item
Newlines do not act as whitespace to separate fields when @code{FS} is
equal to a single space.
@item
The synonym @code{func} for the keyword @code{function} is not
recognized (@pxref{Definition Syntax, ,Function Definition Syntax}).
@item
The operators @samp{**} and @samp{**=} cannot be used in
place of @samp{^} and @samp{^=} (@pxref{Arithmetic Ops, ,Arithmetic Operators},
and also @pxref{Assignment Ops, ,Assignment Expressions}).
@item
Specifying @samp{-Ft} on the command line does not set the value
of @code{FS} to be a single tab character
(@pxref{Field Separators, ,Specifying How Fields are Separated}).
@item
The @code{fflush} built-in function is not supported
(@pxref{I/O Functions, , Built-in Functions for Input/Output}).
@end itemize
@node BTL, POSIX/GNU, POSIX, Language History
@section Extensions in the Bell Laboratories @code{awk}
@cindex Kernighan, Brian
Brian Kernighan, one of the original designers of Unix @code{awk},
has made his version available via anonymous @code{ftp}
(@pxref{Other Versions, ,Other Freely Available @code{awk} Implementations}).
This section describes extensions in his version of @code{awk} that are
not in POSIX @code{awk}.
@itemize @bullet
@item
The @samp{-mf @var{NNN}} and @samp{-mr @var{NNN}} command line options
to set the maximum number of fields, and the maximum
record size, respectively
(@pxref{Options, ,Command Line Options}).
@item
The @code{fflush} built-in function for flushing buffered output
(@pxref{I/O Functions, ,Built-in Functions for Input/Output}).
@ignore
@item
The @code{SYMTAB} array, that allows access to the internal symbol
table of @code{awk}. This feature is not documented, largely because
it is somewhat shakily implemented. For instance, you cannot access arrays
or array elements through it.
@end ignore
@end itemize
@node POSIX/GNU, , BTL, Language History
@section Extensions in @code{gawk} Not in POSIX @code{awk}
@cindex compatibility mode
The GNU implementation, @code{gawk}, adds a number of features.
This sections lists them in the order they were added to @code{gawk}.
They can all be disabled with either the @samp{--traditional} or
@samp{--posix} options
(@pxref{Options, ,Command Line Options}).
Version 2.10 of @code{gawk} introduced these features:
@itemize @bullet
@item
The @code{AWKPATH} environment variable for specifying a path search for
the @samp{-f} command line option
(@pxref{Options, ,Command Line Options}).
@item
The @code{IGNORECASE} variable and its effects
(@pxref{Case-sensitivity, ,Case-sensitivity in Matching}).
@item
The @file{/dev/stdin}, @file{/dev/stdout}, @file{/dev/stderr}, and
@file{/dev/fd/@var{n}} file name interpretation
(@pxref{Special Files, ,Special File Names in @code{gawk}}).
@end itemize
Version 2.13 of @code{gawk} introduced these features:
@itemize @bullet
@item
The @code{FIELDWIDTHS} variable and its effects
(@pxref{Constant Size, ,Reading Fixed-width Data}).
@item
The @code{systime} and @code{strftime} built-in functions for obtaining
and printing time stamps
(@pxref{Time Functions, ,Functions for Dealing with Time Stamps}).
@item
The @samp{-W lint} option to provide source code and run time error
and portability checking
(@pxref{Options, ,Command Line Options}).
@item
The @samp{-W compat} option to turn off these extensions
(@pxref{Options, ,Command Line Options}).
@item
The @samp{-W posix} option for full POSIX compliance
(@pxref{Options, ,Command Line Options}).
@end itemize
Version 2.14 of @code{gawk} introduced these features:
@itemize @bullet
@item
The @code{next file} statement for skipping to the next data file
(@pxref{Nextfile Statement, ,The @code{nextfile} Statement}).
@end itemize
Version 2.15 of @code{gawk} introduced these features:
@itemize @bullet
@item
The @code{ARGIND} variable, that tracks the movement of @code{FILENAME}
through @code{ARGV} (@pxref{Built-in Variables}).
@item
The @code{ERRNO} variable, that contains the system error message when
@code{getline} returns @minus{}1, or when @code{close} fails
(@pxref{Built-in Variables}).
@item
The ability to use GNU-style long named options that start with @samp{--}
(@pxref{Options, ,Command Line Options}).
@item
The @samp{--source} option for mixing command line and library
file source code
(@pxref{Options, ,Command Line Options}).
@item
The @file{/dev/pid}, @file{/dev/ppid}, @file{/dev/pgrpid}, and
@file{/dev/user} file name interpretation
(@pxref{Special Files, ,Special File Names in @code{gawk}}).
@end itemize
Version 3.0 of @code{gawk} introduced these features:
@itemize @bullet
@item
The @code{next file} statement became @code{nextfile}
(@pxref{Nextfile Statement, ,The @code{nextfile} Statement}).
@item
The @samp{--lint-old} option to
warn about constructs that are not available in
the original Version 7 Unix version of @code{awk}
(@pxref{V7/SVR3.1, , Major Changes between V7 and SVR3.1}).
@item
The @samp{--traditional} option was added as a better name for
@samp{--compat} (@pxref{Options, ,Command Line Options}).
@item
The ability for @code{FS} to be a null string, and for the third
argument to @code{split} to be the null string
(@pxref{Single Character Fields, , Making Each Character a Separate Field}).
@item
The ability for @code{RS} to be a regexp
(@pxref{Records, , How Input is Split into Records}).
@item
The @code{RT} variable
(@pxref{Records, , How Input is Split into Records}).
@item
The @code{gensub} function for more powerful text manipulation
(@pxref{String Functions, , Built-in Functions for String Manipulation}).
@item
The @code{strftime} function acquired a default time format,
allowing it to be called with no arguments
(@pxref{Time Functions, , Functions for Dealing with Time Stamps}).
@item
Full support for both POSIX and GNU regexps
(@pxref{Regexp, , Regular Expressions}).
@item
The @samp{--re-interval} option to provide interval expressions in regexps
(@pxref{Regexp Operators, , Regular Expression Operators}).
@item
@code{IGNORECASE} changed, now applying to string comparison as well
as regexp operations
(@pxref{Case-sensitivity, ,Case-sensitivity in Matching}).
@item
The @samp{-m} option and the @code{fflush} function from the
Bell Labs research version of @code{awk}
(@pxref{Options, ,Command Line Options}; also
@pxref{I/O Functions, ,Built-in Functions for Input/Output}).
@item
The use of GNU Autoconf to control the configuration process
(@pxref{Quick Installation, , Compiling @code{gawk} for Unix}).
@item
Amiga support
(@pxref{Amiga Installation, ,Installing @code{gawk} on an Amiga}).
@c XXX ADD MORE STUFF HERE
@end itemize
@node Gawk Summary, Installation, Language History, Top
@appendix @code{gawk} Summary
This appendix provides a brief summary of the @code{gawk} command line and the
@code{awk} language. It is designed to serve as ``quick reference.'' It is
therefore terse, but complete.
@menu
* Command Line Summary:: Recapitulation of the command line.
* Language Summary:: A terse review of the language.
* Variables/Fields:: Variables, fields, and arrays.
* Rules Summary:: Patterns and Actions, and their component
parts.
* Actions Summary:: Quick overview of actions.
* Functions Summary:: Defining and calling functions.
* Historical Features:: Some undocumented but supported ``features''.
@end menu
@node Command Line Summary, Language Summary, Gawk Summary, Gawk Summary
@appendixsec Command Line Options Summary
The command line consists of options to @code{gawk} itself, the
@code{awk} program text (if not supplied via the @samp{-f} option), and
values to be made available in the @code{ARGC} and @code{ARGV}
predefined @code{awk} variables:
@example
gawk @r{[@var{POSIX or GNU style options}]} -f @var{source-file} @r{[@code{--}]} @var{file} @dots{}
gawk @r{[@var{POSIX or GNU style options}]} @r{[@code{--}]} '@var{program}' @var{file} @dots{}
@end example
The options that @code{gawk} accepts are:
@table @code
@item -F @var{fs}
@itemx --field-separator @var{fs}
Use @var{fs} for the input field separator (the value of the @code{FS}
predefined variable).
@item -f @var{program-file}
@itemx --file @var{program-file}
Read the @code{awk} program source from the file @var{program-file}, instead
of from the first command line argument.
@item -mf @var{NNN}
@itemx -mr @var{NNN}
The @samp{f} flag sets
the maximum number of fields, and the @samp{r} flag sets the maximum
record size. These options are ignored by @code{gawk}, since @code{gawk}
has no predefined limits; they are only for compatibility with the
Bell Labs research version of Unix @code{awk}.
@item -v @var{var}=@var{val}
@itemx --assign @var{var}=@var{val}
Assign the variable @var{var} the value @var{val} before program execution
begins.
@item -W traditional
@itemx -W compat
@itemx --traditional
@itemx --compat
Use compatibility mode, in which @code{gawk} extensions are turned
off.
@item -W copyleft
@itemx -W copyright
@itemx --copyleft
@itemx --copyright
Print the short version of the General Public License on the standard
output, and exit. This option may disappear in a future version of @code{gawk}.
@item -W help
@itemx -W usage
@itemx --help
@itemx --usage
Print a relatively short summary of the available options on the standard
output, and exit.
@item -W lint
@itemx --lint
Give warnings about dubious or non-portable @code{awk} constructs.
@item -W lint-old
@itemx --lint-old
Warn about constructs that are not available in
the original Version 7 Unix version of @code{awk}.
@item -W posix
@itemx --posix
Use POSIX compatibility mode, in which @code{gawk} extensions
are turned off and additional restrictions apply.
@item -W re-interval
@itemx --re-interval
Allow interval expressions
(@pxref{Regexp Operators, , Regular Expression Operators}),
in regexps.
@item -W source=@var{program-text}
@itemx --source @var{program-text}
Use @var{program-text} as @code{awk} program source code. This option allows
mixing command line source code with source code from files, and is
particularly useful for mixing command line programs with library functions.
@item -W version
@itemx --version
Print version information for this particular copy of @code{gawk} on the error
output.
@item --
Signal the end of options. This is useful to allow further arguments to the
@code{awk} program itself to start with a @samp{-}. This is mainly for
consistency with POSIX argument parsing conventions.
@end table
Any other options are flagged as invalid, but are otherwise ignored.
@xref{Options, ,Command Line Options}, for more details.
@node Language Summary, Variables/Fields, Command Line Summary, Gawk Summary
@appendixsec Language Summary
An @code{awk} program consists of a sequence of zero or more pattern-action
statements and optional function definitions. One or the other of the
pattern and action may be omitted.
@example
@var{pattern} @{ @var{action statements} @}
@var{pattern}
@{ @var{action statements} @}
function @var{name}(@var{parameter list}) @{ @var{action statements} @}
@end example
@code{gawk} first reads the program source from the
@var{program-file}(s), if specified, or from the first non-option
argument on the command line. The @samp{-f} option may be used multiple
times on the command line. @code{gawk} reads the program text from all
the @var{program-file} files, effectively concatenating them in the
order they are specified. This is useful for building libraries of
@code{awk} functions, without having to include them in each new
@code{awk} program that uses them. To use a library function in a file
from a program typed in on the command line, specify
@samp{--source '@var{program}'}, and type your program in between the single
quotes.
@xref{Options, ,Command Line Options}.
The environment variable @code{AWKPATH} specifies a search path to use
when finding source files named with the @samp{-f} option. The default
path, which is
@samp{.:/usr/local/share/awk}@footnote{The path may use a directory
other than @file{/usr/local/share/awk}, depending upon how @code{gawk}
was built and installed.} is used if @code{AWKPATH} is not set.
If a file name given to the @samp{-f} option contains a @samp{/} character,
no path search is performed.
@xref{AWKPATH Variable, ,The @code{AWKPATH} Environment Variable}.
@code{gawk} compiles the program into an internal form, and then proceeds to
read each file named in the @code{ARGV} array.
The initial values of @code{ARGV} come from the command line arguments.
If there are no files named
on the command line, @code{gawk} reads the standard input.
If a ``file'' named on the command line has the form
@samp{@var{var}=@var{val}}, it is treated as a variable assignment: the
variable @var{var} is assigned the value @var{val}.
If any of the files have a value that is the null string, that
element in the list is skipped.
For each record in the input, @code{gawk} tests to see if it matches any
@var{pattern} in the @code{awk} program. For each pattern that the record
matches, the associated @var{action} is executed.
@node Variables/Fields, Rules Summary, Language Summary, Gawk Summary
@appendixsec Variables and Fields
@code{awk} variables are not declared; they come into existence when they are
first used. Their values are either floating-point numbers or strings.
@code{awk} also has one-dimensional arrays; multiple-dimensional arrays
may be simulated. There are several predefined variables that
@code{awk} sets as a program runs; these are summarized below.
@menu
* Fields Summary:: Input field splitting.
* Built-in Summary:: @code{awk}'s built-in variables.
* Arrays Summary:: Using arrays.
* Data Type Summary:: Values in @code{awk} are numbers or strings.
@end menu
@node Fields Summary, Built-in Summary, Variables/Fields, Variables/Fields
@appendixsubsec Fields
As each input line is read, @code{gawk} splits the line into
@var{fields}, using the value of the @code{FS} variable as the field
separator. If @code{FS} is a single character, fields are separated by
that character. Otherwise, @code{FS} is expected to be a full regular
expression. In the special case that @code{FS} is a single space,
fields are separated by runs of spaces, tabs and/or newlines.@footnote{In
POSIX @code{awk}, newline does not separate fields.}
If @code{FS} is the null string (@code{""}), then each individual
character in the record becomes a separate field.
Note that the value
of @code{IGNORECASE} (@pxref{Case-sensitivity, ,Case-sensitivity in Matching})
also affects how fields are split when @code{FS} is a regular expression.
Each field in the input line may be referenced by its position, @code{$1},
@code{$2}, and so on. @code{$0} is the whole line. The value of a field may
be assigned to as well. Field numbers need not be constants:
@example
n = 5
print $n
@end example
@noindent
prints the fifth field in the input line. The variable @code{NF} is set to
the total number of fields in the input line.
References to non-existent fields (i.e.@: fields after @code{$NF}) return
the null string. However, assigning to a non-existent field (e.g.,
@code{$(NF+2) = 5}) increases the value of @code{NF}, creates any
intervening fields with the null string as their value, and causes the
value of @code{$0} to be recomputed, with the fields being separated by
the value of @code{OFS}.
Decrementing @code{NF} causes the values of fields past the new value to
be lost, and the value of @code{$0} to be recomputed, with the fields being
separated by the value of @code{OFS}.
@xref{Reading Files, ,Reading Input Files}.
@node Built-in Summary, Arrays Summary, Fields Summary, Variables/Fields
@appendixsubsec Built-in Variables
@code{gawk}'s built-in variables are:
@table @code
@item ARGC
The number of elements in @code{ARGV}. See below for what is actually
included in @code{ARGV}.
@item ARGIND
The index in @code{ARGV} of the current file being processed.
When @code{gawk} is processing the input data files,
it is always true that @samp{FILENAME == ARGV[ARGIND]}.
@item ARGV
The array of command line arguments. The array is indexed from zero to
@code{ARGC} @minus{} 1. Dynamically changing @code{ARGC} and
the contents of @code{ARGV}
can control the files used for data. A null-valued element in
@code{ARGV} is ignored. @code{ARGV} does not include the options to
@code{awk} or the text of the @code{awk} program itself.
@item CONVFMT
The conversion format to use when converting numbers to strings.
@item FIELDWIDTHS
A space separated list of numbers describing the fixed-width input data.
@item ENVIRON
An array of environment variable values. The array
is indexed by variable name, each element being the value of that
variable. Thus, the environment variable @code{HOME} is
@code{ENVIRON["HOME"]}. One possible value might be @file{/home/arnold}.
Changing this array does not affect the environment seen by programs
which @code{gawk} spawns via redirection or the @code{system} function.
(This may change in a future version of @code{gawk}.)
Some operating systems do not have environment variables.
The @code{ENVIRON} array is empty when running on these systems.
@item ERRNO
The system error message when an error occurs using @code{getline}
or @code{close}.
@item FILENAME
The name of the current input file. If no files are specified on the command
line, the value of @code{FILENAME} is the null string.
@item FNR
The input record number in the current input file.
@item FS
The input field separator, a space by default.
@item IGNORECASE
The case-sensitivity flag for string comparisons and regular expression
operations. If @code{IGNORECASE} has a non-zero value, then pattern
matching in rules, record separating with @code{RS}, field splitting
with @code{FS}, regular expression matching with @samp{~} and
@samp{!~}, and the @code{gensub}, @code{gsub}, @code{index},
@code{match}, @code{split} and @code{sub} built-in functions all
ignore case when doing regular expression operations, and all string
comparisons are done ignoring case.
The value of @code{IGNORECASE} does @emph{not} affect array subscripting.
@item NF
The number of fields in the current input record.
@item NR
The total number of input records seen so far.
@item OFMT
The output format for numbers for the @code{print} statement,
@code{"%.6g"} by default.
@item OFS
The output field separator, a space by default.
@item ORS
The output record separator, by default a newline.
@item RS
The input record separator, by default a newline.
If @code{RS} is set to the null string, then records are separated by
blank lines. When @code{RS} is set to the null string, then the newline
character always acts as a field separator, in addition to whatever value
@code{FS} may have. If @code{RS} is set to a multi-character
string, it denotes a regexp; input text matching the regexp
separates records.
@item RT
The input text that matched the text denoted by @code{RS},
the record separator.
@item RSTART
The index of the first character last matched by @code{match}; zero if no match.
@item RLENGTH
The length of the string last matched by @code{match}; @minus{}1 if no match.
@item SUBSEP
The string used to separate multiple subscripts in array elements, by
default @code{"\034"}.
@end table
@xref{Built-in Variables}, for more information.
@node Arrays Summary, Data Type Summary, Built-in Summary, Variables/Fields
@appendixsubsec Arrays
Arrays are subscripted with an expression between square brackets
(@samp{[} and @samp{]}). Array subscripts are @emph{always} strings;
numbers are converted to strings as necessary, following the standard
conversion rules
(@pxref{Conversion, ,Conversion of Strings and Numbers}).
If you use multiple expressions separated by commas inside the square
brackets, then the array subscript is a string consisting of the
concatenation of the individual subscript values, converted to strings,
separated by the subscript separator (the value of @code{SUBSEP}).
The special operator @code{in} may be used in a conditional context
to see if an array has an index consisting of a particular value.
@example
if (val in array)
print array[val]
@end example
If the array has multiple subscripts, use @samp{(i, j, @dots{}) in @var{array}}
to test for existence of an element.
The @code{in} construct may also be used in a @code{for} loop to iterate
over all the elements of an array.
@xref{Scanning an Array, ,Scanning All Elements of an Array}.
You can remove an element from an array using the @code{delete} statement.
You can clear an entire array using @samp{delete @var{array}}.
@xref{Arrays, ,Arrays in @code{awk}}.
@node Data Type Summary, , Arrays Summary, Variables/Fields
@appendixsubsec Data Types
The value of an @code{awk} expression is always either a number
or a string.
Some contexts (such as arithmetic operators) require numeric
values. They convert strings to numbers by interpreting the text
of the string as a number. If the string does not look like a
number, it converts to zero.
Other contexts (such as concatenation) require string values.
They convert numbers to strings by effectively printing them
with @code{sprintf}.
@xref{Conversion, ,Conversion of Strings and Numbers}, for the details.
To force conversion of a string value to a number, simply add zero
to it. If the value you start with is already a number, this
does not change it.
To force conversion of a numeric value to a string, concatenate it with
the null string.
Comparisons are done numerically if both operands are numeric, or if
one is numeric and the other is a numeric string. Otherwise one or
both operands are converted to strings and a string comparison is
performed. Fields, @code{getline} input, @code{FILENAME}, @code{ARGV}
elements, @code{ENVIRON} elements and the elements of an array created
by @code{split} are the only items that can be numeric strings. String
constants, such as @code{"3.1415927"} are not numeric strings, they are
string constants. The full rules for comparisons are described in
@ref{Typing and Comparison, ,Variable Typing and Comparison Expressions}.
Uninitialized variables have the string value @code{""} (the null, or
empty, string). In contexts where a number is required, this is
equivalent to zero.
@xref{Variables}, for more information on variable naming and initialization;
@pxref{Conversion, ,Conversion of Strings and Numbers}, for more information
on how variable values are interpreted.
@node Rules Summary, Actions Summary, Variables/Fields, Gawk Summary
@appendixsec Patterns
@menu
* Pattern Summary:: Quick overview of patterns.
* Regexp Summary:: Quick overview of regular expressions.
@end menu
An @code{awk} program is mostly composed of rules, each consisting of a
pattern followed by an action. The action is enclosed in @samp{@{} and
@samp{@}}. Either the pattern may be missing, or the action may be
missing, but not both. If the pattern is missing, the
action is executed for every input record. A missing action is
equivalent to @samp{@w{@{ print @}}}, which prints the entire line.
@c These paragraphs repeated for both patterns and actions. I don't
@c like this, but I also don't see any way around it. Update both copies
@c if they need fixing.
Comments begin with the @samp{#} character, and continue until the end of the
line. Blank lines may be used to separate statements. Statements normally
end with a newline; however, this is not the case for lines ending in a
@samp{,}, @samp{@{}, @samp{?}, @samp{:}, @samp{&&}, or @samp{||}. Lines
ending in @code{do} or @code{else} also have their statements automatically
continued on the following line. In other cases, a line can be continued by
ending it with a @samp{\}, in which case the newline is ignored.
Multiple statements may be put on one line by separating each one with
a @samp{;}.
This applies to both the statements within the action part of a rule (the
usual case), and to the rule statements.
@xref{Comments, ,Comments in @code{awk} Programs}, for information on
@code{awk}'s commenting convention;
@pxref{Statements/Lines, ,@code{awk} Statements Versus Lines}, for a
description of the line continuation mechanism in @code{awk}.
@node Pattern Summary, Regexp Summary, Rules Summary, Rules Summary
@appendixsubsec Pattern Summary
@code{awk} patterns may be one of the following:
@example
/@var{regular expression}/
@var{relational expression}
@var{pattern} && @var{pattern}
@var{pattern} || @var{pattern}
@var{pattern} ? @var{pattern} : @var{pattern}
(@var{pattern})
! @var{pattern}
@var{pattern1}, @var{pattern2}
BEGIN
END
@end example
@code{BEGIN} and @code{END} are two special kinds of patterns that are not
tested against the input. The action parts of all @code{BEGIN} rules are
concatenated as if all the statements had been written in a single @code{BEGIN}
rule. They are executed before any of the input is read. Similarly, all the
@code{END} rules are concatenated, and executed when all the input is exhausted (or
when an @code{exit} statement is executed). @code{BEGIN} and @code{END}
patterns cannot be combined with other patterns in pattern expressions.
@code{BEGIN} and @code{END} rules cannot have missing action parts.
For @code{/@var{regular-expression}/} patterns, the associated statement is
executed for each input record that matches the regular expression. Regular
expressions are summarized below.
A @var{relational expression} may use any of the operators defined below in
the section on actions. These generally test whether certain fields match
certain regular expressions.
The @samp{&&}, @samp{||}, and @samp{!} operators are logical ``and,''
logical ``or,'' and logical ``not,'' respectively, as in C. They do
short-circuit evaluation, also as in C, and are used for combining more
primitive pattern expressions. As in most languages, parentheses may be
used to change the order of evaluation.
The @samp{?:} operator is like the same operator in C. If the first
pattern matches, then the second pattern is matched against the input
record; otherwise, the third is matched. Only one of the second and
third patterns is matched.
The @samp{@var{pattern1}, @var{pattern2}} form of a pattern is called a
range pattern. It matches all input lines starting with a line that
matches @var{pattern1}, and continuing until a line that matches
@var{pattern2}, inclusive. A range pattern cannot be used as an operand
of any of the pattern operators.
@xref{Pattern Overview, ,Pattern Elements}.
@node Regexp Summary, , Pattern Summary, Rules Summary
@appendixsubsec Regular Expressions
Regular expressions are based on POSIX EREs (extended regular expressions).
The escape sequences allowed in string constants are also valid in
regular expressions (@pxref{Escape Sequences}).
Regexps are composed of characters as follows:
@table @code
@item @var{c}
matches the character @var{c} (assuming @var{c} is none of the characters
listed below).
@item \@var{c}
matches the literal character @var{c}.
@item .
matches any character, @emph{including} newline.
In strict POSIX mode, @samp{.} does not match the @sc{nul}
character, which is a character with all bits equal to zero.
@item ^
matches the beginning of a string.
@item $
matches the end of a string.
@item [@var{abc}@dots{}]
matches any of the characters @var{abc}@dots{} (character list).
@item [[:@var{class}:]]
matches any character in the character class @var{class}. Allowable classes
are @code{alnum}, @code{alpha}, @code{blank}, @code{cntrl},
@code{digit}, @code{graph}, @code{lower}, @code{print}, @code{punct},
@code{space}, @code{upper}, and @code{xdigit}.
@item [[.@var{symbol}.]]
matches the multi-character collating symbol @var{symbol}.
@code{gawk} does not currently support collating symbols.
@item [[=@var{classname}=]]
matches any of the equivalent characters in the current locale named by the
equivalence class @var{classname}.
@code{gawk} does not currently support equivalence classes.
@item [^@var{abc}@dots{}]
matches any character except @var{abc}@dots{} (negated
character list).
@item @var{r1}|@var{r2}
matches either @var{r1} or @var{r2} (alternation).
@item @var{r1r2}
matches @var{r1}, and then @var{r2} (concatenation).
@item @var{r}+
matches one or more @var{r}'s.
@item @var{r}*
matches zero or more @var{r}'s.
@item @var{r}?
matches zero or one @var{r}'s.
@item (@var{r})
matches @var{r} (grouping).
@item @var{r}@{@var{n}@}
@itemx @var{r}@{@var{n},@}
@itemx @var{r}@{@var{n},@var{m}@}
matches at least @var{n}, @var{n} to any number, or @var{n} to @var{m}
occurrences of @var{r} (interval expressions).
@item \y
matches the empty string at either the beginning or the
end of a word.
@item \B
matches the empty string within a word.
@item \<
matches the empty string at the beginning of a word.
@item \>
matches the empty string at the end of a word.
@item \w
matches any word-constituent character (alphanumeric characters and
the underscore).
@item \W
matches any character that is not word-constituent.
@item \`
matches the empty string at the beginning of a buffer (same as a string
in @code{gawk}).
@item \'
matches the empty string at the end of a buffer.
@end table
The various command line options
control how @code{gawk} interprets characters in regexps.
@c NOTE!!! Keep this in sync with the same table in the regexp chapter!
@table @asis
@item No options
In the default case, @code{gawk} provide all the facilities of
POSIX regexps and the GNU regexp operators described above.
However, interval expressions are not supported.
@item @code{--posix}
Only POSIX regexps are supported, the GNU operators are not special
(e.g., @samp{\w} matches a literal @samp{w}). Interval expressions
are allowed.
@item @code{--traditional}
Traditional Unix @code{awk} regexps are matched. The GNU operators
are not special, interval expressions are not available, and neither
are the POSIX character classes (@code{[[:alnum:]]} and so on).
Characters described by octal and hexadecimal escape sequences are
treated literally, even if they represent regexp metacharacters.
@item @code{--re-interval}
Allow interval expressions in regexps, even if @samp{--traditional}
has been provided.
@end table
@xref{Regexp, ,Regular Expressions}.
@node Actions Summary, Functions Summary, Rules Summary, Gawk Summary
@appendixsec Actions
Action statements are enclosed in braces, @samp{@{} and @samp{@}}.
A missing action statement is equivalent to @samp{@w{@{ print @}}}.
Action statements consist of the usual assignment, conditional, and looping
statements found in most languages. The operators, control statements,
and Input/Output statements available are similar to those in C.
@c These paragraphs repeated for both patterns and actions. I don't
@c like this, but I also don't see any way around it. Update both copies
@c if they need fixing.
Comments begin with the @samp{#} character, and continue until the end of the
line. Blank lines may be used to separate statements. Statements normally
end with a newline; however, this is not the case for lines ending in a
@samp{,}, @samp{@{}, @samp{?}, @samp{:}, @samp{&&}, or @samp{||}. Lines
ending in @code{do} or @code{else} also have their statements automatically
continued on the following line. In other cases, a line can be continued by
ending it with a @samp{\}, in which case the newline is ignored.
Multiple statements may be put on one line by separating each one with
a @samp{;}.
This applies to both the statements within the action part of a rule (the
usual case), and to the rule statements.
@xref{Comments, ,Comments in @code{awk} Programs}, for information on
@code{awk}'s commenting convention;
@pxref{Statements/Lines, ,@code{awk} Statements Versus Lines}, for a
description of the line continuation mechanism in @code{awk}.
@menu
* Operator Summary:: @code{awk} operators.
* Control Flow Summary:: The control statements.
* I/O Summary:: The I/O statements.
* Printf Summary:: A summary of @code{printf}.
* Special File Summary:: Special file names interpreted internally.
* Built-in Functions Summary:: Built-in numeric and string functions.
* Time Functions Summary:: Built-in time functions.
* String Constants Summary:: Escape sequences in strings.
@end menu
@node Operator Summary, Control Flow Summary, Actions Summary, Actions Summary
@appendixsubsec Operators
The operators in @code{awk}, in order of decreasing precedence, are:
@table @code
@item (@dots{})
Grouping.
@item $
Field reference.
@item ++ --
Increment and decrement, both prefix and postfix.
@item ^
Exponentiation (@samp{**} may also be used, and @samp{**=} for the assignment
operator, but they are not specified in the POSIX standard).
@item + - !
Unary plus, unary minus, and logical negation.
@item * / %
Multiplication, division, and modulus.
@item + -
Addition and subtraction.
@item @var{space}
String concatenation.
@item < <= > >= != ==
The usual relational operators.
@item ~ !~
Regular expression match, negated match.
@item in
Array membership.
@item &&
Logical ``and''.
@item ||
Logical ``or''.
@item ?:
A conditional expression. This has the form @samp{@var{expr1} ?
@var{expr2} : @var{expr3}}. If @var{expr1} is true, the value of the
expression is @var{expr2}; otherwise it is @var{expr3}. Only one of
@var{expr2} and @var{expr3} is evaluated.
@item = += -= *= /= %= ^=
Assignment. Both absolute assignment (@code{@var{var}=@var{value}})
and operator assignment (the other forms) are supported.
@end table
@xref{Expressions}.
@node Control Flow Summary, I/O Summary, Operator Summary, Actions Summary
@appendixsubsec Control Statements
The control statements are as follows:
@example
if (@var{condition}) @var{statement} @r{[} else @var{statement} @r{]}
while (@var{condition}) @var{statement}
do @var{statement} while (@var{condition})
for (@var{expr1}; @var{expr2}; @var{expr3}) @var{statement}
for (@var{var} in @var{array}) @var{statement}
break
continue
delete @var{array}[@var{index}]
delete @var{array}
exit @r{[} @var{expression} @r{]}
@{ @var{statements} @}
@end example
@xref{Statements, ,Control Statements in Actions}.
@node I/O Summary, Printf Summary, Control Flow Summary, Actions Summary
@appendixsubsec I/O Statements
The Input/Output statements are as follows:
@table @code
@item getline
Set @code{$0} from next input record; set @code{NF}, @code{NR}, @code{FNR}.
@xref{Getline, ,Explicit Input with @code{getline}}.
@item getline <@var{file}
Set @code{$0} from next record of @var{file}; set @code{NF}.
@item getline @var{var}
Set @var{var} from next input record; set @code{NR}, @code{FNR}.
@item getline @var{var} <@var{file}
Set @var{var} from next record of @var{file}.
@item @var{command} | getline
Run @var{command}, piping its output into @code{getline}; sets @code{$0},
@code{NF}, @code{NR}.
@item @var{command} | getline @code{var}
Run @var{command}, piping its output into @code{getline}; sets @var{var}.
@item next
Stop processing the current input record. The next input record is read and
processing starts over with the first pattern in the @code{awk} program.
If the end of the input data is reached, the @code{END} rule(s), if any,
are executed.
@xref{Next Statement, ,The @code{next} Statement}.
@item nextfile
Stop processing the current input file. The next input record read comes
from the next input file. @code{FILENAME} is updated, @code{FNR} is set to one,
@code{ARGIND} is incremented,
and processing starts over with the first pattern in the @code{awk} program.
If the end of the input data is reached, the @code{END} rule(s), if any,
are executed.
Earlier versions of @code{gawk} used @samp{next file}; this usage is still
supported, but is considered to be deprecated.
@xref{Nextfile Statement, ,The @code{nextfile} Statement}.
@item print
Prints the current record.
@xref{Printing, ,Printing Output}.
@item print @var{expr-list}
Prints expressions.
@item print @var{expr-list} > @var{file}
Prints expressions to @var{file}. If @var{file} does not exist, it is
created. If it does exist, its contents are deleted the first time the
@code{print} is executed.
@item print @var{expr-list} >> @var{file}
Prints expressions to @var{file}. The previous contents of @var{file}
are retained, and the output of @code{print} is appended to the file.
@item print @var{expr-list} | @var{command}
Prints expressions, sending the output down a pipe to @var{command}.
The pipeline to the command stays open until the @code{close} function
is called.
@item printf @var{fmt, expr-list}
Format and print.
@item printf @var{fmt, expr-list} > file
Format and print to @var{file}. If @var{file} does not exist, it is
created. If it does exist, its contents are deleted the first time the
@code{printf} is executed.
@item printf @var{fmt, expr-list} >> @var{file}
Format and print to @var{file}. The previous contents of @var{file}
are retained, and the output of @code{printf} is appended to the file.
@item printf @var{fmt, expr-list} | @var{command}
Format and print, sending the output down a pipe to @var{command}.
The pipeline to the command stays open until the @code{close} function
is called.
@end table
@code{getline} returns zero on end of file, and @minus{}1 on an error.
In the event of an error, @code{getline} will set @code{ERRNO} to
the value of a system-dependent string that describes the error.
@node Printf Summary, Special File Summary, I/O Summary, Actions Summary
@appendixsubsec @code{printf} Summary
Conversion specification have the form
@code{%}[@var{flag}][@var{width}][@code{.}@var{prec}]@var{format}.
@c whew!
Items in brackets are optional.
The @code{awk} @code{printf} statement and @code{sprintf} function
accept the following conversion specification formats:
@table @code
@item %c
An ASCII character. If the argument used for @samp{%c} is numeric, it is
treated as a character and printed. Otherwise, the argument is assumed to
be a string, and the only first character of that string is printed.
@item %d
@itemx %i
A decimal number (the integer part).
@item %e
@itemx %E
A floating point number of the form
@samp{@r{[}-@r{]}d.dddddde@r{[}+-@r{]}dd}.
The @samp{%E} format uses @samp{E} instead of @samp{e}.
@item %f
A floating point number of the form
@r{[}@code{-}@r{]}@code{ddd.dddddd}.
@item %g
@itemx %G
Use either the @samp{%e} or @samp{%f} formats, whichever produces a shorter
string, with non-significant zeros suppressed.
@samp{%G} will use @samp{%E} instead of @samp{%e}.
@item %o
An unsigned octal number (again, an integer).
@item %s
A character string.
@item %x
@itemx %X
An unsigned hexadecimal number (an integer).
The @samp{%X} format uses @samp{A} through @samp{F} instead of
@samp{a} through @samp{f} for decimal 10 through 15.
@item %%
A single @samp{%} character; no argument is converted.
@end table
There are optional, additional parameters that may lie between the @samp{%}
and the control letter:
@table @code
@item -
The expression should be left-justified within its field.
@item @var{space}
For numeric conversions, prefix positive values with a space, and
negative values with a minus sign.
@item +
The plus sign, used before the width modifier (see below),
says to always supply a sign for numeric conversions, even if the data
to be formatted is positive. The @samp{+} overrides the space modifier.
@item #
Use an ``alternate form'' for certain control letters.
For @samp{o}, supply a leading zero.
For @samp{x}, and @samp{X}, supply a leading @samp{0x} or @samp{0X} for
a non-zero result.
For @samp{e}, @samp{E}, and @samp{f}, the result will always contain a
decimal point.
For @samp{g}, and @samp{G}, trailing zeros are not removed from the result.
@item 0
A leading @samp{0} (zero) acts as a flag, that indicates output should be
padded with zeros instead of spaces.
This applies even to non-numeric output formats.
This flag only has an effect when the field width is wider than the
value to be printed.
@item @var{width}
The field should be padded to this width. The field is normally padded
with spaces. If the @samp{0} flag has been used, it is padded with zeros.
@item .@var{prec}
A number that specifies the precision to use when printing.
For the @samp{e}, @samp{E}, and @samp{f} formats, this specifies the
number of digits you want printed to the right of the decimal point.
For the @samp{g}, and @samp{G} formats, it specifies the maximum number
of significant digits. For the @samp{d}, @samp{o}, @samp{i}, @samp{u},
@samp{x}, and @samp{X} formats, it specifies the minimum number of
digits to print. For the @samp{s} format, it specifies the maximum number of
characters from the string that should be printed.
@end table
Either or both of the @var{width} and @var{prec} values may be specified
as @samp{*}. In that case, the particular value is taken from the argument
list.
@xref{Printf, ,Using @code{printf} Statements for Fancier Printing}.
@node Special File Summary, Built-in Functions Summary, Printf Summary, Actions Summary
@appendixsubsec Special File Names
When doing I/O redirection from either @code{print} or @code{printf} into a
file, or via @code{getline} from a file, @code{gawk} recognizes certain special
file names internally. These file names allow access to open file descriptors
inherited from @code{gawk}'s parent process (usually the shell). The
file names are:
@table @file
@item /dev/stdin
The standard input.
@item /dev/stdout
The standard output.
@item /dev/stderr
The standard error output.
@item /dev/fd/@var{n}
The file denoted by the open file descriptor @var{n}.
@end table
In addition, reading the following files provides process related information
about the running @code{gawk} program. All returned records are terminated
with a newline.
@table @file
@item /dev/pid
Returns the process ID of the current process.
@item /dev/ppid
Returns the parent process ID of the current process.
@item /dev/pgrpid
Returns the process group ID of the current process.
@item /dev/user
At least four space-separated fields, containing the return values of
the @code{getuid}, @code{geteuid}, @code{getgid}, and @code{getegid}
system calls.
If there are any additional fields, they are the group IDs returned by
@code{getgroups} system call.
(Multiple groups may not be supported on all systems.)
@end table
@noindent
These file names may also be used on the command line to name data files.
These file names are only recognized internally if you do not
actually have files with these names on your system.
@xref{Special Files, ,Special File Names in @code{gawk}}, for a longer description that
provides the motivation for this feature.
@node Built-in Functions Summary, Time Functions Summary, Special File Summary, Actions Summary
@appendixsubsec Built-in Functions
@code{awk} provides a number of built-in functions for performing
numeric operations, string related operations, and I/O related operations.
The built-in arithmetic functions are:
@table @code
@item atan2(@var{y}, @var{x})
the arctangent of @var{y/x} in radians.
@item cos(@var{expr})
the cosine of @var{expr}, which is in radians.
@item exp(@var{expr})
the exponential function (@code{e ^ @var{expr}}).
@item int(@var{expr})
truncates to integer.
@item log(@var{expr})
the natural logarithm of @code{expr}.
@item rand()
a random number between zero and one.
@item sin(@var{expr})
the sine of @var{expr}, which is in radians.
@item sqrt(@var{expr})
the square root function.
@item srand(@r{[}@var{expr}@r{]})
use @var{expr} as a new seed for the random number generator. If no @var{expr}
is provided, the time of day is used. The return value is the previous
seed for the random number generator.
@end table
@code{awk} has the following built-in string functions:
@table @code
@item gensub(@var{regex}, @var{subst}, @var{how} @r{[}, @var{target}@r{]})
If @var{how} is a string beginning with @samp{g} or @samp{G}, then
replace each match of @var{regex} in @var{target} with @var{subst}.
Otherwise, replace the @var{how}'th occurrence. If @var{target} is not
supplied, use @code{$0}. The return value is the changed string; the
original @var{target} is not modified. Within @var{subst},
@samp{\@var{n}}, where @var{n} is a digit from one to nine, can be used to
indicate the text that matched the @var{n}'th parenthesized
subexpression.
This function is @code{gawk}-specific.
@item gsub(@var{regex}, @var{subst} @r{[}, @var{target}@r{]})
for each substring matching the regular expression @var{regex} in the string
@var{target}, substitute the string @var{subst}, and return the number of
substitutions. If @var{target} is not supplied, use @code{$0}.
@item index(@var{str}, @var{search})
returns the index of the string @var{search} in the string @var{str}, or
zero if
@var{search} is not present.
@item length(@r{[}@var{str}@r{]})
returns the length of the string @var{str}. The length of @code{$0}
is returned if no argument is supplied.
@item match(@var{str}, @var{regex})
returns the position in @var{str} where the regular expression @var{regex}
occurs, or zero if @var{regex} is not present, and sets the values of
@code{RSTART} and @code{RLENGTH}.
@item split(@var{str}, @var{arr} @r{[}, @var{regex}@r{]})
splits the string @var{str} into the array @var{arr} on the regular expression
@var{regex}, and returns the number of elements. If @var{regex} is omitted,
@code{FS} is used instead. @var{regex} can be the null string, causing
each character to be placed into its own array element.
The array @var{arr} is cleared first.
@item sprintf(@var{fmt}, @var{expr-list})
prints @var{expr-list} according to @var{fmt}, and returns the resulting string.
@item sub(@var{regex}, @var{subst} @r{[}, @var{target}@r{]})
just like @code{gsub}, but only the first matching substring is replaced.
@item substr(@var{str}, @var{index} @r{[}, @var{len}@r{]})
returns the @var{len}-character substring of @var{str} starting at @var{index}.
If @var{len} is omitted, the rest of @var{str} is used.
@item tolower(@var{str})
returns a copy of the string @var{str}, with all the upper-case characters in
@var{str} translated to their corresponding lower-case counterparts.
Non-alphabetic characters are left unchanged.
@item toupper(@var{str})
returns a copy of the string @var{str}, with all the lower-case characters in
@var{str} translated to their corresponding upper-case counterparts.
Non-alphabetic characters are left unchanged.
@end table
The I/O related functions are:
@table @code
@item close(@var{expr})
Close the open file or pipe denoted by @var{expr}.
@item fflush(@r{[}@var{expr}@r{]})
Flush any buffered output for the output file or pipe denoted by @var{expr}.
If @var{expr} is omitted, standard output is flushed.
If @var{expr} is the null string (@code{""}), all output buffers are flushed.
@item system(@var{cmd-line})
Execute the command @var{cmd-line}, and return the exit status.
If your operating system does not support @code{system}, calling it will
generate a fatal error.
@samp{system("")} can be used to force @code{awk} to flush any pending
output. This is more portable, but less obvious, than calling @code{fflush}.
@end table
@node Time Functions Summary, String Constants Summary, Built-in Functions Summary, Actions Summary
@appendixsubsec Time Functions
The following two functions are available for getting the current
time of day, and for formatting time stamps.
They are specific to @code{gawk}.
@table @code
@item systime()
returns the current time of day as the number of seconds since a particular
epoch (Midnight, January 1, 1970 UTC, on POSIX systems).
@item strftime(@r{[}@var{format}@r{[}, @var{timestamp}@r{]]})
formats @var{timestamp} according to the specification in @var{format}.
The current time of day is used if no @var{timestamp} is supplied.
A default format equivalent to the output of the @code{date} utility is used if
no @var{format} is supplied.
@xref{Time Functions, ,Functions for Dealing with Time Stamps}, for the
details on the conversion specifiers that @code{strftime} accepts.
@end table
@iftex
@xref{Built-in, ,Built-in Functions}, for a description of all of
@code{awk}'s built-in functions.
@end iftex
@node String Constants Summary, , Time Functions Summary, Actions Summary
@appendixsubsec String Constants
String constants in @code{awk} are sequences of characters enclosed
in double quotes (@code{"}). Within strings, certain @dfn{escape sequences}
are recognized, as in C. These are:
@table @code
@item \\
A literal backslash.
@item \a
The ``alert'' character; usually the ASCII BEL character.
@item \b
Backspace.
@item \f
Formfeed.
@item \n
Newline.
@item \r
Carriage return.
@item \t
Horizontal tab.
@item \v
Vertical tab.
@item \x@var{hex digits}
The character represented by the string of hexadecimal digits following
the @samp{\x}. As in ANSI C, all following hexadecimal digits are
considered part of the escape sequence. E.g., @code{"\x1B"} is a
string containing the ASCII ESC (escape) character. (The @samp{\x}
escape sequence is not in POSIX @code{awk}.)
@item \@var{ddd}
The character represented by the one, two, or three digit sequence of octal
digits. Thus, @code{"\033"} is also a string containing the ASCII ESC
(escape) character.
@item \@var{c}
The literal character @var{c}, if @var{c} is not one of the above.
@end table
The escape sequences may also be used inside constant regular expressions
(e.g., the regexp @code{@w{/[@ \t\f\n\r\v]/}} matches whitespace
characters).
@xref{Escape Sequences}.
@node Functions Summary, Historical Features, Actions Summary, Gawk Summary
@appendixsec User-defined Functions
Functions in @code{awk} are defined as follows:
@example
function @var{name}(@var{parameter list}) @{ @var{statements} @}
@end example
Actual parameters supplied in the function call are used to instantiate
the formal parameters declared in the function. Arrays are passed by
reference, other variables are passed by value.
If there are fewer arguments passed than there are names in @var{parameter-list},
the extra names are given the null string as their value. Extra names have the
effect of local variables.
The open-parenthesis in a function call of a user-defined function must
immediately follow the function name, without any intervening white space.
This is to avoid a syntactic ambiguity with the concatenation operator.
The word @code{func} may be used in place of @code{function} (but not in
POSIX @code{awk}).
Use the @code{return} statement to return a value from a function.
@xref{User-defined, ,User-defined Functions}.
@node Historical Features, , Functions Summary, Gawk Summary
@appendixsec Historical Features
@cindex historical features
There are two features of historical @code{awk} implementations that
@code{gawk} supports.
First, it is possible to call the @code{length} built-in function not only
with no arguments, but even without parentheses!
@example
a = length
@end example
@noindent
is the same as either of
@example
a = length()
a = length($0)
@end example
@noindent
For example:
@example
$ echo abcdef | awk '@{ print length @}'
@print{} 6
@end example
@noindent
This feature is marked as ``deprecated'' in the POSIX standard, and
@code{gawk} will issue a warning about its use if @samp{--lint} is
specified on the command line.
(The ability to use @code{length} this way was actually an accident of the
original Unix @code{awk} implementation. If any built-in function used
@code{$0} as its default argument, it was possible to call that function
without the parentheses. In particular, it was common practice to use
the @code{length} function in this fashion, and this usage was documented
in the @code{awk} manual page.)
The other historical feature is the use of either the @code{break} statement,
or the @code{continue} statement
outside the body of a @code{while}, @code{for}, or @code{do} loop. Traditional
@code{awk} implementations have treated such usage as equivalent to the
@code{next} statement. More recent versions of Unix @code{awk} do not allow
it. @code{gawk} supports this usage if @samp{--traditional} has been
specified.
@xref{Options, ,Command Line Options}, for more information about the
@samp{--posix} and @samp{--lint} options.
@node Installation, Notes, Gawk Summary, Top
@appendix Installing @code{gawk}
This appendix provides instructions for installing @code{gawk} on the
various platforms that are supported by the developers. The primary
developers support Unix (and one day, GNU), while the other ports were
contributed. The file @file{ACKNOWLEDGMENT} in the @code{gawk}
distribution lists the electronic mail addresses of the people who did
the respective ports, and they are also provided in
@ref{Bugs, , Reporting Problems and Bugs}.
@menu
* Gawk Distribution:: What is in the @code{gawk} distribution.
* Unix Installation:: Installing @code{gawk} under various versions
of Unix.
* VMS Installation:: Installing @code{gawk} on VMS.
* PC Installation:: Installing and Compiling @code{gawk} on MS-DOS
and OS/2
* Atari Installation:: Installing @code{gawk} on the Atari ST.
* Amiga Installation:: Installing @code{gawk} on an Amiga.
* Bugs:: Reporting Problems and Bugs.
* Other Versions:: Other freely available @code{awk}
implementations.
@end menu
@node Gawk Distribution, Unix Installation, Installation, Installation
@appendixsec The @code{gawk} Distribution
This section first describes how to get the @code{gawk}
distribution, how to extract it, and then what is in the various files and
subdirectories.
@menu
* Getting:: How to get the distribution.
* Extracting:: How to extract the distribution.
* Distribution contents:: What is in the distribution.
@end menu
@node Getting, Extracting, Gawk Distribution, Gawk Distribution
@appendixsubsec Getting the @code{gawk} Distribution
@cindex getting @code{gawk}
@cindex anonymous @code{ftp}
@cindex @code{ftp}, anonymous
@cindex Free Software Foundation
There are three ways you can get GNU software.
@enumerate
@item
You can copy it from someone else who already has it.
@cindex Free Software Foundation
@item
You can order @code{gawk} directly from the Free Software Foundation.
Software distributions are available for Unix, MS-DOS, and VMS, on
tape and CD-ROM. The address is:
@quotation
Free Software Foundation @*
59 Temple Place---Suite 330 @*
Boston, MA 02111-1307 USA @*
Phone: +1-617-542-5942 @*
Fax (including Japan): +1-617-542-2652 @*
E-mail: @code{gnu@@prep.ai.mit.edu} @*
@end quotation
@noindent
Ordering from the FSF directly contributes to the support of the foundation
and to the production of more free software.
@item
You can get @code{gawk} by using anonymous @code{ftp} to the Internet host
@code{ftp.gnu.ai.mit.edu}, in the directory @file{/pub/gnu}.
Here is a list of alternate @code{ftp} sites from which you can obtain GNU
software. When a site is listed as ``@var{site}@code{:}@var{directory}'' the
@var{directory} indicates the directory where GNU software is kept.
You should use a site that is geographically close to you.
@table @asis
@item Asia:
@table @code
@item cair-archive.kaist.ac.kr:/pub/gnu
@itemx ftp.cs.titech.ac.jp
@itemx ftp.nectec.or.th:/pub/mirrors/gnu
@itemx utsun.s.u-tokyo.ac.jp:/ftpsync/prep
@end table
@item Australia:
@table @code
@item archie.au:/gnu
(@code{archie.oz} or @code{archie.oz.au} for ACSnet)
@end table
@item Africa:
@table @code
@item ftp.sun.ac.za:/pub/gnu
@end table
@item Middle East:
@table @code
@item ftp.technion.ac.il:/pub/unsupported/gnu
@end table
@item Europe:
@table @code
@item archive.eu.net
@itemx ftp.denet.dk
@itemx ftp.eunet.ch
@itemx ftp.funet.fi:/pub/gnu
@itemx ftp.ieunet.ie:pub/gnu
@itemx ftp.informatik.rwth-aachen.de:/pub/gnu
@itemx ftp.informatik.tu-muenchen.de
@itemx ftp.luth.se:/pub/unix/gnu
@itemx ftp.mcc.ac.uk
@itemx ftp.stacken.kth.se
@itemx ftp.sunet.se:/pub/gnu
@itemx ftp.univ-lyon1.fr:pub/gnu
@itemx ftp.win.tue.nl:/pub/gnu
@itemx irisa.irisa.fr:/pub/gnu
@itemx isy.liu.se
@itemx nic.switch.ch:/mirror/gnu
@itemx src.doc.ic.ac.uk:/gnu
@itemx unix.hensa.ac.uk:/pub/uunet/systems/gnu
@end table
@item South America:
@table @code
@item ftp.inf.utfsm.cl:/pub/gnu
@itemx ftp.unicamp.br:/pub/gnu
@end table
@item Western Canada:
@table @code
@item ftp.cs.ubc.ca:/mirror2/gnu
@end table
@item USA:
@table @code
@item col.hp.com:/mirrors/gnu
@itemx f.ms.uky.edu:/pub3/gnu
@itemx ftp.cc.gatech.edu:/pub/gnu
@itemx ftp.cs.columbia.edu:/archives/gnu/prep
@itemx ftp.digex.net:/pub/gnu
@itemx ftp.hawaii.edu:/mirrors/gnu
@itemx ftp.kpc.com:/pub/mirror/gnu
@end table
@c NEEDED
@page
@item USA (continued):
@table @code
@itemx ftp.uu.net:/systems/gnu
@itemx gatekeeper.dec.com:/pub/GNU
@itemx jaguar.utah.edu:/gnustuff
@itemx labrea.stanford.edu
@itemx mrcnext.cso.uiuc.edu:/pub/gnu
@itemx vixen.cso.uiuc.edu:/gnu
@itemx wuarchive.wustl.edu:/systems/gnu
@end table
@end table
@end enumerate
@node Extracting, Distribution contents, Getting, Gawk Distribution
@appendixsubsec Extracting the Distribution
@code{gawk} is distributed as a @code{tar} file compressed with the
GNU Zip program, @code{gzip}.
Once you have the distribution (for example,
@file{gawk-@value{VERSION}.@value{PATCHLEVEL}.tar.gz}), first use @code{gzip} to expand the
file, and then use @code{tar} to extract it. You can use the following
pipeline to produce the @code{gawk} distribution:
@example
# Under System V, add 'o' to the tar flags
gzip -d -c gawk-@value{VERSION}.@value{PATCHLEVEL}.tar.gz | tar -xvpf -
@end example
@noindent
This will create a directory named @file{gawk-@value{VERSION}.@value{PATCHLEVEL}} in the current
directory.
The distribution file name is of the form
@file{gawk-@var{V}.@var{R}.@var{n}.tar.gz}.
The @var{V} represents the major version of @code{gawk},
the @var{R} represents the current release of version @var{V}, and
the @var{n} represents a @dfn{patch level}, meaning that minor bugs have
been fixed in the release. The current patch level is @value{PATCHLEVEL},
but when
retrieving distributions, you should get the version with the highest
version, release, and patch level. (Note that release levels greater than
or equal to 90 denote ``beta,'' or non-production software; you may not wish
to retrieve such a version unless you don't mind experimenting.)
If you are not on a Unix system, you will need to make other arrangements
for getting and extracting the @code{gawk} distribution. You should consult
a local expert.
@node Distribution contents, , Extracting, Gawk Distribution
@appendixsubsec Contents of the @code{gawk} Distribution
The @code{gawk} distribution has a number of C source files,
documentation files,
subdirectories and files related to the configuration process
(@pxref{Unix Installation, ,Compiling and Installing @code{gawk} on Unix}),
and several subdirectories related to different, non-Unix,
operating systems.
@table @asis
@item various @samp{.c}, @samp{.y}, and @samp{.h} files
These files are the actual @code{gawk} source code.
@end table
@table @file
@item README
@itemx README_d/README.*
Descriptive files: @file{README} for @code{gawk} under Unix, and the
rest for the various hardware and software combinations.
@item INSTALL
A file providing an overview of the configuration and installation process.
@item PORTS
A list of systems to which @code{gawk} has been ported, and which
have successfully run the test suite.
@item ACKNOWLEDGMENT
A list of the people who contributed major parts of the code or documentation.
@item ChangeLog
A detailed list of source code changes as bugs are fixed or improvements made.
@item NEWS
A list of changes to @code{gawk} since the last release or patch.
@item COPYING
The GNU General Public License.
@item FUTURES
A brief list of features and/or changes being contemplated for future
releases, with some indication of the time frame for the feature, based
on its difficulty.
@item LIMITATIONS
A list of those factors that limit @code{gawk}'s performance.
Most of these depend on the hardware or operating system software, and
are not limits in @code{gawk} itself.
@item POSIX.STD
A description of one area where the POSIX standard for @code{awk} is
incorrect, and how @code{gawk} handles the problem.
@item PROBLEMS
A file describing known problems with the current release.
@cindex artificial intelligence, using @code{gawk}
@cindex AI programming, using @code{gawk}
@item doc/awkforai.txt
A short article describing why @code{gawk} is a good language for
AI (Artificial Intelligence) programming.
@item doc/README.card
@itemx doc/ad.block
@itemx doc/awkcard.in
@itemx doc/cardfonts
@itemx doc/colors
@itemx doc/macros
@itemx doc/no.colors
@itemx doc/setter.outline
The @code{troff} source for a five-color @code{awk} reference card.
A modern version of @code{troff}, such as GNU Troff (@code{groff}) is
needed to produce the color version. See the file @file{README.card}
for instructions if you have an older @code{troff}.
@item doc/gawk.1
The @code{troff} source for a manual page describing @code{gawk}.
This is distributed for the convenience of Unix users.
@item doc/gawk.texi
The Texinfo source file for this @value{DOCUMENT}.
It should be processed with @TeX{} to produce a printed document, and
with @code{makeinfo} to produce an Info file.
@item doc/gawk.info
The generated Info file for this @value{DOCUMENT}.
@item doc/igawk.1
The @code{troff} source for a manual page describing the @code{igawk}
program presented in
@ref{Igawk Program, ,An Easy Way to Use Library Functions}.
@item doc/Makefile.in
The input file used during the configuration process to generate the
actual @file{Makefile} for creating the documentation.
@item Makefile.in
@itemx acconfig.h
@itemx aclocal.m4
@itemx configh.in
@itemx configure.in
@itemx configure
@itemx custom.h
@itemx missing/*
These files and subdirectory are used when configuring @code{gawk}
for various Unix systems. They are explained in detail in
@ref{Unix Installation, ,Compiling and Installing @code{gawk} on Unix}.
@item awklib/extract.awk
@itemx awklib/Makefile.in
The @file{awklib} directory contains a copy of @file{extract.awk}
(@pxref{Extract Program, ,Extracting Programs from Texinfo Source Files}),
which can be used to extract the sample programs from the Texinfo
source file for this @value{DOCUMENT}, and a @file{Makefile.in} file, which
@code{configure} uses to generate a @file{Makefile}.
As part of the process of building @code{gawk}, the library functions from
@ref{Library Functions, , A Library of @code{awk} Functions},
and the @code{igawk} program from
@ref{Igawk Program, , An Easy Way to Use Library Functions},
are extracted into ready to use files.
They are installed as part of the installation process.
@item atari/*
Files needed for building @code{gawk} on an Atari ST.
@xref{Atari Installation, ,Installing @code{gawk} on the Atari ST}, for details.
@item pc/*
Files needed for building @code{gawk} under MS-DOS and OS/2.
@xref{PC Installation, ,MS-DOS and OS/2 Installation and Compilation}, for details.
@item vms/*
Files needed for building @code{gawk} under VMS.
@xref{VMS Installation, ,How to Compile and Install @code{gawk} on VMS}, for details.
@item test/*
A test suite for
@code{gawk}. You can use @samp{make check} from the top level @code{gawk}
directory to run your version of @code{gawk} against the test suite.
If @code{gawk} successfully passes @samp{make check} then you can
be confident of a successful port.
@end table
@node Unix Installation, VMS Installation, Gawk Distribution, Installation
@appendixsec Compiling and Installing @code{gawk} on Unix
Usually, you can compile and install @code{gawk} by typing only two
commands. However, if you do use an unusual system, you may need
to configure @code{gawk} for your system yourself.
@menu
* Quick Installation:: Compiling @code{gawk} under Unix.
* Configuration Philosophy:: How it's all supposed to work.
@end menu
@node Quick Installation, Configuration Philosophy, Unix Installation, Unix Installation
@appendixsubsec Compiling @code{gawk} for Unix
@cindex installation, unix
After you have extracted the @code{gawk} distribution, @code{cd}
to @file{gawk-@value{VERSION}.@value{PATCHLEVEL}}. Like most GNU software,
@code{gawk} is configured
automatically for your Unix system by running the @code{configure} program.
This program is a Bourne shell script that was generated automatically using
GNU @code{autoconf}.
@iftex
(The @code{autoconf} software is
described fully in
@cite{Autoconf---Generating Automatic Configuration Scripts},
which is available from the Free Software Foundation.)
@end iftex
@ifinfo
(The @code{autoconf} software is described fully starting with
@ref{Top, , Introduction, autoconf, Autoconf---Generating Automatic Configuration Scripts}.)
@end ifinfo
To configure @code{gawk}, simply run @code{configure}:
@example
sh ./configure
@end example
This produces a @file{Makefile} and @file{config.h} tailored to your system.
The @file{config.h} file describes various facts about your system.
You may wish to edit the @file{Makefile} to
change the @code{CFLAGS} variable, which controls
the command line options that are passed to the C compiler (such as
optimization levels, or compiling for debugging).
Alternatively, you can add your own values for most @code{make}
variables, such as @code{CC} and @code{CFLAGS}, on the command line when
running @code{configure}:
@example
CC=cc CFLAGS=-g sh ./configure
@end example
@noindent
See the file @file{INSTALL} in the @code{gawk} distribution for
all the details.
After you have run @code{configure}, and possibly edited the @file{Makefile},
type:
@example
make
@end example
@noindent
and shortly thereafter, you should have an executable version of @code{gawk}.
That's all there is to it!
(If these steps do not work, please send in a bug report;
@pxref{Bugs, ,Reporting Problems and Bugs}.)
@node Configuration Philosophy, , Quick Installation, Unix Installation
@appendixsubsec The Configuration Process
@cindex configuring @code{gawk}
(This section is of interest only if you know something about using the
C language and the Unix operating system.)
The source code for @code{gawk} generally attempts to adhere to formal
standards wherever possible. This means that @code{gawk} uses library
routines that are specified by the ANSI C standard and by the POSIX
operating system interface standard. When using an ANSI C compiler,
function prototypes are used to help improve the compile-time checking.
Many Unix systems do not support all of either the ANSI or the
POSIX standards. The @file{missing} subdirectory in the @code{gawk}
distribution contains replacement versions of those subroutines that are
most likely to be missing.
The @file{config.h} file that is created by the @code{configure} program
contains definitions that describe features of the particular operating
system where you are attempting to compile @code{gawk}. The three things
described by this file are what header files are available, so that
they can be correctly included,
what (supposedly) standard functions are actually available in your C
libraries, and
other miscellaneous facts about your
variant of Unix. For example, there may not be an @code{st_blksize}
element in the @code{stat} structure. In this case @samp{HAVE_ST_BLKSIZE}
would be undefined.
@cindex @code{custom.h} configuration file
It is possible for your C compiler to lie to @code{configure}. It may
do so by not exiting with an error when a library function is not
available. To get around this, you can edit the file @file{custom.h}.
Use an @samp{#ifdef} that is appropriate for your system, and either
@code{#define} any constants that @code{configure} should have defined but
didn't, or @code{#undef} any constants that @code{configure} defined and
should not have. @file{custom.h} is automatically included by
@file{config.h}.
It is also possible that the @code{configure} program generated by
@code{autoconf}
will not work on your system in some other fashion. If you do have a problem,
the file
@file{configure.in} is the input for @code{autoconf}. You may be able to
change this file, and generate a new version of @code{configure} that will
work on your system. @xref{Bugs, ,Reporting Problems and Bugs}, for
information on how to report problems in configuring @code{gawk}. The same
mechanism may be used to send in updates to @file{configure.in} and/or
@file{custom.h}.
@node VMS Installation, PC Installation, Unix Installation, Installation
@appendixsec How to Compile and Install @code{gawk} on VMS
@c based on material from Pat Rankin <rankin@eql.caltech.edu>
@cindex installation, vms
This section describes how to compile and install @code{gawk} under VMS.
@menu
* VMS Compilation:: How to compile @code{gawk} under VMS.
* VMS Installation Details:: How to install @code{gawk} under VMS.
* VMS Running:: How to run @code{gawk} under VMS.
* VMS POSIX:: Alternate instructions for VMS POSIX.
@end menu
@node VMS Compilation, VMS Installation Details, VMS Installation, VMS Installation
@appendixsubsec Compiling @code{gawk} on VMS
To compile @code{gawk} under VMS, there is a @code{DCL} command procedure that
will issue all the necessary @code{CC} and @code{LINK} commands, and there is
also a @file{Makefile} for use with the @code{MMS} utility. From the source
directory, use either
@example
$ @@[.VMS]VMSBUILD.COM
@end example
@noindent
or
@example
$ MMS/DESCRIPTION=[.VMS]DESCRIP.MMS GAWK
@end example
Depending upon which C compiler you are using, follow one of the sets
of instructions in this table:
@table @asis
@item VAX C V3.x
Use either @file{vmsbuild.com} or @file{descrip.mms} as is. These use
@code{CC/OPTIMIZE=NOLINE}, which is essential for Version 3.0.
@item VAX C V2.x
You must have Version 2.3 or 2.4; older ones won't work. Edit either
@file{vmsbuild.com} or @file{descrip.mms} according to the comments in them.
For @file{vmsbuild.com}, this just entails removing two @samp{!} delimiters.
Also edit @file{config.h} (which is a copy of file @file{[.config]vms-conf.h})
and comment out or delete the two lines @samp{#define __STDC__ 0} and
@samp{#define VAXC_BUILTINS} near the end.
@item GNU C
Edit @file{vmsbuild.com} or @file{descrip.mms}; the changes are different
from those for VAX C V2.x, but equally straightforward. No changes to
@file{config.h} should be needed.
@item DEC C
Edit @file{vmsbuild.com} or @file{descrip.mms} according to their comments.
No changes to @file{config.h} should be needed.
@end table
@code{gawk} has been tested under VAX/VMS 5.5-1 using VAX C V3.2,
GNU C 1.40 and 2.3. It should work without modifications for VMS V4.6 and up.
@node VMS Installation Details, VMS Running, VMS Compilation, VMS Installation
@appendixsubsec Installing @code{gawk} on VMS
To install @code{gawk}, all you need is a ``foreign'' command, which is
a @code{DCL} symbol whose value begins with a dollar sign. For example:
@example
$ GAWK :== $disk1:[gnubin]GAWK
@end example
@noindent
(Substitute the actual location of @code{gawk.exe} for
@samp{$disk1:[gnubin]}.) The symbol should be placed in the
@file{login.com} of any user who wishes to run @code{gawk},
so that it will be defined every time the user logs on.
Alternatively, the symbol may be placed in the system-wide
@file{sylogin.com} procedure, which will allow all users
to run @code{gawk}.
Optionally, the help entry can be loaded into a VMS help library:
@example
$ LIBRARY/HELP SYS$HELP:HELPLIB [.VMS]GAWK.HLP
@end example
@noindent
(You may want to substitute a site-specific help library rather than
the standard VMS library @samp{HELPLIB}.) After loading the help text,
@example
$ HELP GAWK
@end example
@noindent
will provide information about both the @code{gawk} implementation and the
@code{awk} programming language.
The logical name @samp{AWK_LIBRARY} can designate a default location
for @code{awk} program files. For the @samp{-f} option, if the specified
filename has no device or directory path information in it, @code{gawk}
will look in the current directory first, then in the directory specified
by the translation of @samp{AWK_LIBRARY} if the file was not found.
If after searching in both directories, the file still is not found,
then @code{gawk} appends the suffix @samp{.awk} to the filename and the
file search will be re-tried. If @samp{AWK_LIBRARY} is not defined, that
portion of the file search will fail benignly.
@node VMS Running, VMS POSIX, VMS Installation Details, VMS Installation
@appendixsubsec Running @code{gawk} on VMS
Command line parsing and quoting conventions are significantly different
on VMS, so examples in this @value{DOCUMENT} or from other sources often need minor
changes. They @emph{are} minor though, and all @code{awk} programs
should run correctly.
Here are a couple of trivial tests:
@example
$ gawk -- "BEGIN @{print ""Hello, World!""@}"
$ gawk -"W" version
! could also be -"W version" or "-W version"
@end example
@noindent
Note that upper-case and mixed-case text must be quoted.
The VMS port of @code{gawk} includes a @code{DCL}-style interface in addition
to the original shell-style interface (see the help entry for details).
One side-effect of dual command line parsing is that if there is only a
single parameter (as in the quoted string program above), the command
becomes ambiguous. To work around this, the normally optional @samp{--}
flag is required to force Unix style rather than @code{DCL} parsing. If any
other dash-type options (or multiple parameters such as data files to be
processed) are present, there is no ambiguity and @samp{--} can be omitted.
The default search path when looking for @code{awk} program files specified
by the @samp{-f} option is @code{"SYS$DISK:[],AWK_LIBRARY:"}. The logical
name @samp{AWKPATH} can be used to override this default. The format
of @samp{AWKPATH} is a comma-separated list of directory specifications.
When defining it, the value should be quoted so that it retains a single
translation, and not a multi-translation @code{RMS} searchlist.
@node VMS POSIX, , VMS Running, VMS Installation
@appendixsubsec Building and Using @code{gawk} on VMS POSIX
Ignore the instructions above, although @file{vms/gawk.hlp} should still
be made available in a help library. The source tree should be unpacked
into a container file subsystem rather than into the ordinary VMS file
system. Make sure that the two scripts, @file{configure} and
@file{vms/posix-cc.sh}, are executable; use @samp{chmod +x} on them if
necessary. Then execute the following two commands:
@example
@group
psx> CC=vms/posix-cc.sh configure
psx> make CC=c89 gawk
@end group
@end example
@noindent
The first command will construct files @file{config.h} and @file{Makefile} out
of templates, using a script to make the C compiler fit @code{configure}'s
expectations. The second command will compile and link @code{gawk} using
the C compiler directly; ignore any warnings from @code{make} about being
unable to redefine @code{CC}. @code{configure} will take a very long
time to execute, but at least it provides incremental feedback as it
runs.
This has been tested with VAX/VMS V6.2, VMS POSIX V2.0, and DEC C V5.2.
Once built, @code{gawk} will work like any other shell utility. Unlike
the normal VMS port of @code{gawk}, no special command line manipulation is
needed in the VMS POSIX environment.
@c Rewritten by Scott Deifik <scottd@amgen.com>
@c and Darrel Hankerson <hankedr@mail.auburn.edu>
@node PC Installation, Atari Installation, VMS Installation, Installation
@appendixsec MS-DOS and OS/2 Installation and Compilation
@cindex installation, MS-DOS and OS/2
If you have received a binary distribution prepared by the DOS
maintainers, then @code{gawk} and the necessary support files will appear
under the @file{gnu} directory, with executables in @file{gnu/bin},
libraries in @file{gnu/lib/awk}, and manual pages under @file{gnu/man}.
This is designed for easy installation to a @file{/gnu} directory on your
drive, but the files can be installed anywhere provided @code{AWKPATH} is
set properly. Regardless of the installation directory, the first line of
@file{igawk.cmd} and @file{igawk.bat} (in @file{gnu/bin}) may need to be
edited.
The binary distribution will contain a separate file describing the
contents. In particular, it may include more than one version of the
@code{gawk} executable. OS/2 binary distributions may have a
different arrangement, but installation is similar.
The OS/2 and MS-DOS versions of @code{gawk} search for program files as
described in @ref{AWKPATH Variable, ,The @code{AWKPATH} Environment Variable}.
However, semicolons (rather than colons) separate elements
in the @code{AWKPATH} variable. If @code{AWKPATH} is not set or is empty,
then the default search path is @code{@w{".;c:/lib/awk;c:/gnu/lib/awk"}}.
An @code{sh}-like shell (as opposed to @code{command.com} under MS-DOS
or @code{cmd.exe} under OS/2) may be useful for @code{awk} programming.
Ian Stewartson has written an excellent shell for MS-DOS and OS/2, and a
@code{ksh} clone and GNU Bash are available for OS/2. The file
@file{README_d/README.pc} in the @code{gawk} distribution contains
information on these shells. Users of Stewartson's shell on DOS should
examine its documentation on handling of command-lines. In particular,
the setting for @code{gawk} in the shell configuration may need to be
changed, and the @code{ignoretype} option may also be of interest.
@code{gawk} can be compiled for MS-DOS and OS/2 using the GNU development tools
from DJ Delorie (DJGPP, MS-DOS-only) or Eberhard Mattes (EMX, MS-DOS and OS/2).
Microsoft C can be used to build 16-bit versions for MS-DOS and OS/2. The file
@file{README_d/README.pc} in the @code{gawk} distribution contains additional
notes, and @file{pc/Makefile} contains important notes on compilation options.
To build @code{gawk}, copy the files in the @file{pc} directory (@emph{except}
for @file{ChangeLog}) to the
directory with the rest of the @code{gawk} sources. The @file{Makefile}
contains a configuration section with comments, and may need to be
edited in order to work with your @code{make} utility.
The @file{Makefile} contains a number of targets for building various MS-DOS
and OS/2 versions. A list of targets will be printed if the @code{make}
command is given without a target. As an example, to build @code{gawk}
using the DJGPP tools, enter @samp{make djgpp}.
Using @code{make} to run the standard tests and to install @code{gawk}
requires additional Unix-like tools, including @code{sh}, @code{sed}, and
@code{cp}. In order to run the tests, the @file{test/*.ok} files may need to
be converted so that they have the usual DOS-style end-of-line markers. Most
of the tests will work properly with Stewartson's shell along with the
companion utilities or appropriate GNU utilities. However, some editing of
@file{test/Makefile} is required. It is recommended that the file
@file{pc/Makefile.tst} be copied to @file{test/Makefile} as a
replacement. Details can be found in @file{README_d/README.pc}.
@node Atari Installation, Amiga Installation, PC Installation, Installation
@appendixsec Installing @code{gawk} on the Atari ST
@c based on material from Michal Jaegermann <michal@gortel.phys.ualberta.ca>
@cindex atari
@cindex installation, atari
There are no substantial differences when installing @code{gawk} on
various Atari models. Compiled @code{gawk} executables do not require
a large amount of memory with most @code{awk} programs and should run on all
Motorola processor based models (called further ST, even if that is not
exactly right).
In order to use @code{gawk}, you need to have a shell, either text or
graphics, that does not map all the characters of a command line to
upper-case. Maintaining case distinction in option flags is very
important (@pxref{Options, ,Command Line Options}).
These days this is the default, and it may only be a problem for some
very old machines. If your system does not preserve the case of option
flags, you will need to upgrade your tools. Support for I/O
redirection is necessary to make it easy to import @code{awk} programs
from other environments. Pipes are nice to have, but not vital.
@menu
* Atari Compiling:: Compiling @code{gawk} on Atari
* Atari Using:: Running @code{gawk} on Atari
@end menu
@node Atari Compiling, Atari Using, Atari Installation, Atari Installation
@appendixsubsec Compiling @code{gawk} on the Atari ST
A proper compilation of @code{gawk} sources when @code{sizeof(int)}
differs from @code{sizeof(void *)} requires an ANSI C compiler. An initial
port was done with @code{gcc}. You may actually prefer executables
where @code{int}s are four bytes wide, but the other variant works as well.
You may need quite a bit of memory when trying to recompile the @code{gawk}
sources, as some source files (@file{regex.c} in particular) are quite
big. If you run out of memory compiling such a file, try reducing the
optimization level for this particular file; this may help.
@cindex Linux
With a reasonable shell (Bash will do), and in particular if you run
Linux, MiNT or a similar operating system, you have a pretty good
chance that the @code{configure} utility will succeed. Otherwise
sample versions of @file{config.h} and @file{Makefile.st} are given in the
@file{atari} subdirectory and can be edited and copied to the
corresponding files in the main source directory. Even if
@code{configure} produced something, it might be advisable to compare
its results with the sample versions and possibly make adjustments.
Some @code{gawk} source code fragments depend on a preprocessor define
@samp{atarist}. This basically assumes the TOS environment with @code{gcc}.
Modify these sections as appropriate if they are not right for your
environment. Also see the remarks about @code{AWKPATH} and @code{envsep} in
@ref{Atari Using, ,Running @code{gawk} on the Atari ST}.
As shipped, the sample @file{config.h} claims that the @code{system}
function is missing from the libraries, which is not true, and an
alternative implementation of this function is provided in
@file{atari/system.c}. Depending upon your particular combination of
shell and operating system, you may wish to change the file to indicate
that @code{system} is available.
@node Atari Using, , Atari Compiling, Atari Installation
@appendixsubsec Running @code{gawk} on the Atari ST
An executable version of @code{gawk} should be placed, as usual,
anywhere in your @code{PATH} where your shell can find it.
While executing, @code{gawk} creates a number of temporary files. When
using @code{gcc} libraries for TOS, @code{gawk} looks for either of
the environment variables @code{TEMP} or @code{TMPDIR}, in that order.
If either one is found, its value is assumed to be a directory for
temporary files. This directory must exist, and if you can spare the
memory, it is a good idea to put it on a RAM drive. If neither
@code{TEMP} nor @code{TMPDIR} are found, then @code{gawk} uses the
current directory for its temporary files.
The ST version of @code{gawk} searches for its program files as described in
@ref{AWKPATH Variable, ,The @code{AWKPATH} Environment Variable}.
The default value for the @code{AWKPATH} variable is taken from
@code{DEFPATH} defined in @file{Makefile}. The sample @code{gcc}/TOS
@file{Makefile} for the ST in the distribution sets @code{DEFPATH} to
@code{@w{".,c:\lib\awk,c:\gnu\lib\awk"}}. The search path can be
modified by explicitly setting @code{AWKPATH} to whatever you wish.
Note that colons cannot be used on the ST to separate elements in the
@code{AWKPATH} variable, since they have another, reserved, meaning.
Instead, you must use a comma to separate elements in the path. When
recompiling, the separating character can be modified by initializing
the @code{envsep} variable in @file{atari/gawkmisc.atr} to another
value.
Although @code{awk} allows great flexibility in doing I/O redirections
from within a program, this facility should be used with care on the ST
running under TOS. In some circumstances the OS routines for file
handle pool processing lose track of certain events, causing the
computer to crash, and requiring a reboot. Often a warm reboot is
sufficient. Fortunately, this happens infrequently, and in rather
esoteric situations. In particular, avoid having one part of an
@code{awk} program using @code{print} statements explicitly redirected
to @code{"/dev/stdout"}, while other @code{print} statements use the
default standard output, and a calling shell has redirected standard
output to a file.
When @code{gawk} is compiled with the ST version of @code{gcc} and its
usual libraries, it will accept both @samp{/} and @samp{\} as path separators.
While this is convenient, it should be remembered that this removes one,
technically valid, character (@samp{/}) from your file names, and that
it may create problems for external programs, called via the @code{system}
function, which may not support this convention. Whenever it is possible
that a file created by @code{gawk} will be used by some other program,
use only backslashes. Also remember that in @code{awk}, backslashes in
strings have to be doubled in order to get literal backslashes
(@pxref{Escape Sequences}).
@node Amiga Installation, Bugs, Atari Installation, Installation
@appendixsec Installing @code{gawk} on an Amiga
@cindex amiga
@cindex installation, amiga
You can install @code{gawk} on an Amiga system using a Unix emulation
environment available via anonymous @code{ftp} from
@code{ftp.ninemoons.com} in the directory @file{pub/ade/current}.
This includes a shell based on @code{pdksh}. The primary component of
this environment is a Unix emulation library, @file{ixemul.lib}.
@c could really use more background here, who wrote this, etc.
A more complete distribution for the Amiga is available on
the Geek Gadgets CD-ROM from:
@quotation
CRONUS @*
1840 E. Warner Road #105-265 @*
Tempe, AZ 85284 USA @*
US Toll Free: (800) 804-0833 @*
Phone: +1-602-491-0442 @*
FAX: +1-602-491-0048 @*
Email: @code{info@@ninemoons.com} @*
WWW: @code{http://www.ninemoons.com} @*
Anonymous @code{ftp} site: @code{ftp.ninemoons.com} @*
@end quotation
Once you have the distribution, you can configure @code{gawk} simply by
running @code{configure}:
@example
configure -v m68k-amigaos
@end example
Then run @code{make}, and you should be all set!
(If these steps do not work, please send in a bug report;
@pxref{Bugs, ,Reporting Problems and Bugs}.)
@node Bugs, Other Versions, Amiga Installation, Installation
@appendixsec Reporting Problems and Bugs
@display
@i{There is nothing more dangerous than a bored archeologist.}
The Hitchhiker's Guide to the Galaxy
@c the radio show, not the book. :-)
@end display
@sp 1
If you have problems with @code{gawk} or think that you have found a bug,
please report it to the developers; we cannot promise to do anything
but we might well want to fix it.
Before reporting a bug, make sure you have actually found a real bug.
Carefully reread the documentation and see if it really says you can do
what you're trying to do. If it's not clear whether you should be able
to do something or not, report that too; it's a bug in the documentation!
Before reporting a bug or trying to fix it yourself, try to isolate it
to the smallest possible @code{awk} program and input data file that
reproduces the problem. Then send us the program and data file,
some idea of what kind of Unix system you're using, and the exact results
@code{gawk} gave you. Also say what you expected to occur; this will help
us decide whether the problem was really in the documentation.
Once you have a precise problem, there are two e-mail addresses you
can send mail to.
@table @asis
@item Internet:
@samp{bug-gnu-utils@@prep.ai.mit.edu}
@item UUCP:
@samp{uunet!prep.ai.mit.edu!bug-gnu-utils}
@end table
Please include the
version number of @code{gawk} you are using. You can get this information
with the command @samp{gawk --version}.
You should send a carbon copy of your mail to Arnold Robbins, who can
be reached at @samp{arnold@@gnu.ai.mit.edu}.
@cindex @code{comp.lang.awk}
@strong{Important!} Do @emph{not} try to report bugs in @code{gawk} by
posting to the Usenet/Internet newsgroup @code{comp.lang.awk}.
While the @code{gawk} developers do occasionally read this newsgroup,
there is no guarantee that we will see your posting. The steps described
above are the official, recognized ways for reporting bugs.
Non-bug suggestions are always welcome as well. If you have questions
about things that are unclear in the documentation or are just obscure
features, ask Arnold Robbins; he will try to help you out, although he
may not have the time to fix the problem. You can send him electronic
mail at the Internet address above.
If you find bugs in one of the non-Unix ports of @code{gawk}, please send
an electronic mail message to the person who maintains that port. They
are listed below, and also in the @file{README} file in the @code{gawk}
distribution. Information in the @file{README} file should be considered
authoritative if it conflicts with this @value{DOCUMENT}.
@c NEEDED for looks
@page
The people maintaining the non-Unix ports of @code{gawk} are:
@cindex Deifik, Scott
@cindex Fish, Fred
@cindex Hankerson, Darrel
@cindex Jaegermann, Michal
@cindex Rankin, Pat
@cindex Rommel, Kai Uwe
@table @asis
@item MS-DOS
Scott Deifik, @samp{scottd@@amgen.com}, and
Darrel Hankerson, @samp{hankedr@@mail.auburn.edu}.
@item OS/2
Kai Uwe Rommel, @samp{rommel@@ars.de}.
@item VMS
Pat Rankin, @samp{rankin@@eql.caltech.edu}.
@item Atari ST
Michal Jaegermann, @samp{michal@@gortel.phys.ualberta.ca}.
@item Amiga
Fred Fish, @samp{fnf@@ninemoons.com}.
@end table
If your bug is also reproducible under Unix, please send copies of your
report to the general GNU bug list, as well as to Arnold Robbins, at the
addresses listed above.
@node Other Versions, , Bugs, Installation
@appendixsec Other Freely Available @code{awk} Implementations
@cindex Brennan, Michael
@ignore
From: emory!amc.com!brennan (Michael Brennan)
Subject: C++ comments in awk programs
To: arnold@gnu.ai.mit.edu (Arnold Robbins)
Date: Wed, 4 Sep 1996 08:11:48 -0700 (PDT)
@end ignore
@display
@i{It's kind of fun to put comments like this in your awk code.}
@code{// Do C++ comments work? answer: yes! of course}
Michael Brennan
@end display
@sp 1
There are two other freely available @code{awk} implementations.
This section briefly describes where to get them.
@table @asis
@cindex Kernighan, Brian
@cindex anonymous @code{ftp}
@cindex @code{ftp}, anonymous
@item Unix @code{awk}
Brian Kernighan has been able to make his implementation of
@code{awk} freely available. You can get it via anonymous @code{ftp}
to the host @code{@w{netlib.att.com}}. Change directory to
@file{/netlib/research}. Use ``binary'' or ``image'' mode, and
retrieve @file{awk.bundle.Z}.
This is a shell archive that has been compressed with the @code{compress}
utility. It can be uncompressed with either @code{uncompress} or the
GNU @code{gunzip} utility.
This version requires an ANSI C compiler; GCC (the GNU C compiler)
works quite nicely.
@cindex Brennan, Michael
@cindex @code{mawk}
@item @code{mawk}
Michael Brennan has written an independent implementation of @code{awk},
called @code{mawk}. It is available under the GPL
(@pxref{Copying, ,GNU GENERAL PUBLIC LICENSE}),
just as @code{gawk} is.
You can get it via anonymous @code{ftp} to the host
@code{@w{ftp.whidbey.net}}. Change directory to @file{/pub/brennan}.
Use ``binary'' or ``image'' mode, and retrieve @file{mawk1.3.3.tar.gz}
(or the latest version that is there).
@code{gunzip} may be used to decompress this file. Installation
is similar to @code{gawk}'s
(@pxref{Unix Installation, , Compiling and Installing @code{gawk} on Unix}).
@end table
@node Notes, Glossary, Installation, Top
@appendix Implementation Notes
This appendix contains information mainly of interest to implementors and
maintainers of @code{gawk}. Everything in it applies specifically to
@code{gawk}, and not to other implementations.
@menu
* Compatibility Mode:: How to disable certain @code{gawk} extensions.
* Additions:: Making Additions To @code{gawk}.
* Future Extensions:: New features that may be implemented one day.
* Improvements:: Suggestions for improvements by volunteers.
@end menu
@node Compatibility Mode, Additions, Notes, Notes
@appendixsec Downward Compatibility and Debugging
@xref{POSIX/GNU, ,Extensions in @code{gawk} Not in POSIX @code{awk}},
for a summary of the GNU extensions to the @code{awk} language and program.
All of these features can be turned off by invoking @code{gawk} with the
@samp{--traditional} option, or with the @samp{--posix} option.
If @code{gawk} is compiled for debugging with @samp{-DDEBUG}, then there
is one more option available on the command line:
@table @code
@item -W parsedebug
@itemx --parsedebug
Print out the parse stack information as the program is being parsed.
@end table
This option is intended only for serious @code{gawk} developers,
and not for the casual user. It probably has not even been compiled into
your version of @code{gawk}, since it slows down execution.
@node Additions, Future Extensions, Compatibility Mode, Notes
@appendixsec Making Additions to @code{gawk}
If you should find that you wish to enhance @code{gawk} in a significant
fashion, you are perfectly free to do so. That is the point of having
free software; the source code is available, and you are free to change
it as you wish (@pxref{Copying, ,GNU GENERAL PUBLIC LICENSE}).
This section discusses the ways you might wish to change @code{gawk},
and any considerations you should bear in mind.
@menu
* Adding Code:: Adding code to the main body of @code{gawk}.
* New Ports:: Porting @code{gawk} to a new operating system.
@end menu
@node Adding Code, New Ports, Additions, Additions
@appendixsubsec Adding New Features
@cindex adding new features
@cindex features, adding
You are free to add any new features you like to @code{gawk}.
However, if you want your changes to be incorporated into the @code{gawk}
distribution, there are several steps that you need to take in order to
make it possible for me to include to your changes.
@enumerate 1
@item
Get the latest version.
It is much easier for me to integrate changes if they are relative to
the most recent distributed version of @code{gawk}. If your version of
@code{gawk} is very old, I may not be able to integrate them at all.
@xref{Getting, ,Getting the @code{gawk} Distribution},
for information on getting the latest version of @code{gawk}.
@item
@iftex
Follow the @cite{GNU Coding Standards}.
@end iftex
@ifinfo
See @inforef{Top, , Version, standards, GNU Coding Standards}.
@end ifinfo
This document describes how GNU software should be written. If you haven't
read it, please do so, preferably @emph{before} starting to modify @code{gawk}.
(The @cite{GNU Coding Standards} are available as part of the Autoconf
distribution, from the FSF.)
@cindex @code{gawk} coding style
@cindex coding style used in @code{gawk}
@item
Use the @code{gawk} coding style.
The C code for @code{gawk} follows the instructions in the
@cite{GNU Coding Standards}, with minor exceptions. The code is formatted
using the traditional ``K&R'' style, particularly as regards the placement
of braces and the use of tabs. In brief, the coding rules for @code{gawk}
are:
@itemize @bullet
@item
Use old style (non-prototype) function headers when defining functions.
@item
Put the name of the function at the beginning of its own line.
@item
Put the return type of the function, even if it is @code{int}, on the
line above the line with the name and arguments of the function.
@item
The declarations for the function arguments should not be indented.
@item
Put spaces around parentheses used in control structures
(@code{if}, @code{while}, @code{for}, @code{do}, @code{switch}
and @code{return}).
@item
Do not put spaces in front of parentheses used in function calls.
@item
Put spaces around all C operators, and after commas in function calls.
@item
Do not use the comma operator to produce multiple side-effects, except
in @code{for} loop initialization and increment parts, and in macro bodies.
@item
Use real tabs for indenting, not spaces.
@item
Use the ``K&R'' brace layout style.
@item
Use comparisons against @code{NULL} and @code{'\0'} in the conditions of
@code{if}, @code{while} and @code{for} statements, and in the @code{case}s
of @code{switch} statements, instead of just the
plain pointer or character value.
@item
Use the @code{TRUE}, @code{FALSE}, and @code{NULL} symbolic constants,
and the character constant @code{'\0'} where appropriate, instead of @code{1}
and @code{0}.
@item
Provide one-line descriptive comments for each function.
@item
Do not use @samp{#elif}. Many older Unix C compilers cannot handle it.
@item
Do not use the @code{alloca} function for allocating memory off the stack.
Its use causes more portability trouble than the minor benefit of not having
to free the storage. Instead, use @code{malloc} and @code{free}.
@end itemize
If I have to reformat your code to follow the coding style used in
@code{gawk}, I may not bother.
@item
Be prepared to sign the appropriate paperwork.
In order for the FSF to distribute your changes, you must either place
those changes in the public domain, and submit a signed statement to that
effect, or assign the copyright in your changes to the FSF.
Both of these actions are easy to do, and @emph{many} people have done so
already. If you have questions, please contact me
(@pxref{Bugs, , Reporting Problems and Bugs}),
or @code{gnu@@prep.ai.mit.edu}.
@item
Update the documentation.
Along with your new code, please supply new sections and or chapters
for this @value{DOCUMENT}. If at all possible, please use real
Texinfo, instead of just supplying unformatted ASCII text (although
even that is better than no documentation at all).
Conventions to be followed in @cite{@value{TITLE}} are provided
after the @samp{@@bye} at the end of the Texinfo source file.
If possible, please update the man page as well.
You will also have to sign paperwork for your documentation changes.
@item
Submit changes as context diffs or unified diffs.
Use @samp{diff -c -r -N} or @samp{diff -u -r -N} to compare
the original @code{gawk} source tree with your version.
(I find context diffs to be more readable, but unified diffs are
more compact.)
I recommend using the GNU version of @code{diff}.
Send the output produced by either run of @code{diff} to me when you
submit your changes.
@xref{Bugs, , Reporting Problems and Bugs}, for the electronic mail
information.
Using this format makes it easy for me to apply your changes to the
master version of the @code{gawk} source code (using @code{patch}).
If I have to apply the changes manually, using a text editor, I may
not do so, particularly if there are lots of changes.
@end enumerate
Although this sounds like a lot of work, please remember that while you
may write the new code, I have to maintain it and support it, and if it
isn't possible for me to do that with a minimum of extra work, then I
probably will not.
@node New Ports, , Adding Code, Additions
@appendixsubsec Porting @code{gawk} to a New Operating System
@cindex porting @code{gawk}
If you wish to port @code{gawk} to a new operating system, there are
several steps to follow.
@enumerate 1
@item
Follow the guidelines in
@ref{Adding Code, ,Adding New Features},
concerning coding style, submission of diffs, and so on.
@item
When doing a port, bear in mind that your code must co-exist peacefully
with the rest of @code{gawk}, and the other ports. Avoid gratuitous
changes to the system-independent parts of the code. If at all possible,
avoid sprinkling @samp{#ifdef}s just for your port throughout the
code.
If the changes needed for a particular system affect too much of the
code, I probably will not accept them. In such a case, you will, of course,
be able to distribute your changes on your own, as long as you comply
with the GPL
(@pxref{Copying, ,GNU GENERAL PUBLIC LICENSE}).
@item
A number of the files that come with @code{gawk} are maintained by other
people at the Free Software Foundation. Thus, you should not change them
unless it is for a very good reason. I.e.@: changes are not out of the
question, but changes to these files will be scrutinized extra carefully.
The files are @file{alloca.c}, @file{getopt.h}, @file{getopt.c},
@file{getopt1.c}, @file{regex.h}, @file{regex.c}, @file{dfa.h},
@file{dfa.c}, @file{install-sh}, and @file{mkinstalldirs}.
@item
Be willing to continue to maintain the port.
Non-Unix operating systems are supported by volunteers who maintain
the code needed to compile and run @code{gawk} on their systems. If no-one
volunteers to maintain a port, that port becomes unsupported, and it may
be necessary to remove it from the distribution.
@item
Supply an appropriate @file{gawkmisc.???} file.
Each port has its own @file{gawkmisc.???} that implements certain
operating system specific functions. This is cleaner than a plethora of
@samp{#ifdef}s scattered throughout the code. The @file{gawkmisc.c} in
the main source directory includes the appropriate
@file{gawkmisc.???} file from each subdirectory.
Be sure to update it as well.
Each port's @file{gawkmisc.???} file has a suffix reminiscent of the machine
or operating system for the port. For example, @file{pc/gawkmisc.pc} and
@file{vms/gawkmisc.vms}. The use of separate suffixes, instead of plain
@file{gawkmisc.c}, makes it possible to move files from a port's subdirectory
into the main subdirectory, without accidentally destroying the real
@file{gawkmisc.c} file. (Currently, this is only an issue for the MS-DOS
and OS/2 ports.)
@item
Supply a @file{Makefile} and any other C source and header files that are
necessary for your operating system. All your code should be in a
separate subdirectory, with a name that is the same as, or reminiscent
of, either your operating system or the computer system. If possible,
try to structure things so that it is not necessary to move files out
of the subdirectory into the main source directory. If that is not
possible, then be sure to avoid using names for your files that
duplicate the names of files in the main source directory.
@item
Update the documentation.
Please write a section (or sections) for this @value{DOCUMENT} describing the
installation and compilation steps needed to install and/or compile
@code{gawk} for your system.
@item
Be prepared to sign the appropriate paperwork.
In order for the FSF to distribute your code, you must either place
your code in the public domain, and submit a signed statement to that
effect, or assign the copyright in your code to the FSF.
@ifinfo
Both of these actions are easy to do, and @emph{many} people have done so
already. If you have questions, please contact me, or
@code{gnu@@prep.ai.mit.edu}.
@end ifinfo
@end enumerate
Following these steps will make it much easier to integrate your changes
into @code{gawk}, and have them co-exist happily with the code for other
operating systems that is already there.
In the code that you supply, and that you maintain, feel free to use a
coding style and brace layout that suits your taste.
@node Future Extensions, Improvements, Additions, Notes
@appendixsec Probable Future Extensions
@ignore
From emory!scalpel.netlabs.com!lwall Tue Oct 31 12:43:17 1995
Return-Path: <emory!scalpel.netlabs.com!lwall>
Message-Id: <9510311732.AA28472@scalpel.netlabs.com>
To: arnold@skeeve.atl.ga.us (Arnold D. Robbins)
Subject: Re: May I quote you?
In-Reply-To: Your message of "Tue, 31 Oct 95 09:11:00 EST."
<m0tAHPQ-00014MC@skeeve.atl.ga.us>
Date: Tue, 31 Oct 95 09:32:46 -0800
From: Larry Wall <emory!scalpel.netlabs.com!lwall>
: Greetings. I am working on the release of gawk 3.0. Part of it will be a
: thoroughly updated manual. One of the sections deals with planned future
: extensions and enhancements. I have the following at the beginning
: of it:
:
: @cindex PERL
: @cindex Wall, Larry
: @display
: @i{AWK is a language similar to PERL, only considerably more elegant.} @*
: Arnold Robbins
: @sp 1
: @i{Hey!} @*
: Larry Wall
: @end display
:
: Before I actually release this for publication, I wanted to get your
: permission to quote you. (Hopefully, in the spirit of much of GNU, the
: implied humor is visible... :-)
I think that would be fine.
Larry
@end ignore
@cindex PERL
@cindex Wall, Larry
@display
@i{AWK is a language similar to PERL, only considerably more elegant.}
Arnold Robbins
@i{Hey!}
Larry Wall
@end display
@sp 1
This section briefly lists extensions and possible improvements
that indicate the directions we are
currently considering for @code{gawk}. The file @file{FUTURES} in the
@code{gawk} distributions lists these extensions as well.
This is a list of probable future changes that will be usable by the
@code{awk} language programmer.
@c these are ordered by likelihood
@table @asis
@item Localization
The GNU project is starting to support multiple languages.
It will at least be possible to make @code{gawk} print its warnings and
error messages in languages other than English.
It may be possible for @code{awk} programs to also use the multiple
language facilities, separate from @code{gawk} itself.
@item Databases
It may be possible to map a GDBM/NDBM/SDBM file into an @code{awk} array.
@item A @code{PROCINFO} Array
The special files that provide process-related information
(@pxref{Special Files, ,Special File Names in @code{gawk}})
may be superseded by a @code{PROCINFO} array that would provide the same
information, in an easier to access fashion.
@item More @code{lint} warnings
There are more things that could be checked for portability.
@item Control of subprocess environment
Changes made in @code{gawk} to the array @code{ENVIRON} may be
propagated to subprocesses run by @code{gawk}.
@ignore
@item @code{RECLEN} variable for fixed length records
Along with @code{FIELDWIDTHS}, this would speed up the processing of
fixed-length records.
@item A @code{restart} keyword
After modifying @code{$0}, @code{restart} would restart the pattern
matching loop, without reading a new record from the input.
@item A @samp{|&} redirection
The @samp{|&} redirection, in place of @samp{|}, would open a two-way
pipeline for communication with a sub-process (via @code{getline} and
@code{print} and @code{printf}).
@item Function valued variables
It would be possible to assign the name of a user-defined or built-in
function to a regular @code{awk} variable, and then call the function
indirectly, by using the regular variable. This would make it possible
to write general purpose sorting and comparing routines, for example,
by simply passing the name of one function into another.
@item A built-in @code{stat} function
The @code{stat} function would provide an easy-to-use hook to the
@code{stat} system call so that @code{awk} programs could determine information
about files.
@item A built-in @code{ftw} function
Combined with function valued variables and the @code{stat} function,
@code{ftw} (file tree walk) would make it easy for an @code{awk} program
to walk an entire file tree.
@end ignore
@end table
This is a list of probable improvements that will make @code{gawk}
perform better.
@table @asis
@item An Improved Version of @code{dfa}
The @code{dfa} pattern matcher from GNU @code{grep} has some
problems. Either a new version or a fixed one will deal with some
important regexp matching issues.
@item Use of GNU @code{malloc}
The GNU version of @code{malloc} could potentially speed up @code{gawk},
since it relies heavily on the use of dynamic memory allocation.
@item Use of the @code{rx} regexp library
The @code{rx} regular expression library could potentially speed up
all regexp operations that require knowing the exact location of matches.
This includes record termination, field and array splitting,
and the @code{sub}, @code{gsub}, @code{gensub} and @code{match} functions.
@end table
@node Improvements, , Future Extensions, Notes
@appendixsec Suggestions for Improvements
Here are some projects that would-be @code{gawk} hackers might like to take
on. They vary in size from a few days to a few weeks of programming,
depending on which one you choose and how fast a programmer you are. Please
send any improvements you write to the maintainers at the GNU project.
@xref{Adding Code, , Adding New Features},
for guidelines to follow when adding new features to @code{gawk}.
@xref{Bugs, ,Reporting Problems and Bugs}, for information on
contacting the maintainers.
@enumerate
@item
Compilation of @code{awk} programs: @code{gawk} uses a Bison (YACC-like)
parser to convert the script given it into a syntax tree; the syntax
tree is then executed by a simple recursive evaluator. This method incurs
a lot of overhead, since the recursive evaluator performs many procedure
calls to do even the simplest things.
It should be possible for @code{gawk} to convert the script's parse tree
into a C program which the user would then compile, using the normal
C compiler and a special @code{gawk} library to provide all the needed
functions (regexps, fields, associative arrays, type coercion, and so
on).
An easier possibility might be for an intermediate phase of @code{awk} to
convert the parse tree into a linear byte code form like the one used
in GNU Emacs Lisp. The recursive evaluator would then be replaced by
a straight line byte code interpreter that would be intermediate in speed
between running a compiled program and doing what @code{gawk} does
now.
@item
The programs in the test suite could use documenting in this @value{DOCUMENT}.
@item
See the @file{FUTURES} file for more ideas. Contact us if you would
seriously like to tackle any of the items listed there.
@end enumerate
@node Glossary, Copying, Notes, Top
@appendix Glossary
@table @asis
@item Action
A series of @code{awk} statements attached to a rule. If the rule's
pattern matches an input record, @code{awk} executes the
rule's action. Actions are always enclosed in curly braces.
@xref{Action Overview, ,Overview of Actions}.
@item Amazing @code{awk} Assembler
Henry Spencer at the University of Toronto wrote a retargetable assembler
completely as @code{awk} scripts. It is thousands of lines long, including
machine descriptions for several eight-bit microcomputers.
It is a good example of a
program that would have been better written in another language.
@item Amazingly Workable Formatter (@code{awf})
Henry Spencer at the University of Toronto wrote a formatter that accepts
a large subset of the @samp{nroff -ms} and @samp{nroff -man} formatting
commands, using @code{awk} and @code{sh}.
@item ANSI
The American National Standards Institute. This organization produces
many standards, among them the standards for the C and C++ programming
languages.
@item Assignment
An @code{awk} expression that changes the value of some @code{awk}
variable or data object. An object that you can assign to is called an
@dfn{lvalue}. The assigned values are called @dfn{rvalues}.
@xref{Assignment Ops, ,Assignment Expressions}.
@item @code{awk} Language
The language in which @code{awk} programs are written.
@item @code{awk} Program
An @code{awk} program consists of a series of @dfn{patterns} and
@dfn{actions}, collectively known as @dfn{rules}. For each input record
given to the program, the program's rules are all processed in turn.
@code{awk} programs may also contain function definitions.
@item @code{awk} Script
Another name for an @code{awk} program.
@item Bash
The GNU version of the standard shell (the Bourne-Again shell).
See ``Bourne Shell.''
@item BBS
See ``Bulletin Board System.''
@item Boolean Expression
Named after the English mathematician Boole. See ``Logical Expression.''
@item Bourne Shell
The standard shell (@file{/bin/sh}) on Unix and Unix-like systems,
originally written by Steven R.@: Bourne.
Many shells (Bash, @code{ksh}, @code{pdksh}, @code{zsh}) are
generally upwardly compatible with the Bourne shell.
@item Built-in Function
The @code{awk} language provides built-in functions that perform various
numerical, time stamp related, and string computations. Examples are
@code{sqrt} (for the square root of a number) and @code{substr} (for a
substring of a string). @xref{Built-in, ,Built-in Functions}.
@item Built-in Variable
@code{ARGC}, @code{ARGIND}, @code{ARGV}, @code{CONVFMT}, @code{ENVIRON},
@code{ERRNO}, @code{FIELDWIDTHS}, @code{FILENAME}, @code{FNR}, @code{FS},
@code{IGNORECASE}, @code{NF}, @code{NR}, @code{OFMT}, @code{OFS}, @code{ORS},
@code{RLENGTH}, @code{RSTART}, @code{RS}, @code{RT}, and @code{SUBSEP},
are the variables that have special meaning to @code{awk}.
Changing some of them affects @code{awk}'s running environment.
Several of these variables are specific to @code{gawk}.
@xref{Built-in Variables}.
@item Braces
See ``Curly Braces.''
@item Bulletin Board System
A computer system allowing users to log in and read and/or leave messages
for other users of the system, much like leaving paper notes on a bulletin
board.
@item C
The system programming language that most GNU software is written in. The
@code{awk} programming language has C-like syntax, and this @value{DOCUMENT}
points out similarities between @code{awk} and C when appropriate.
@cindex ISO 8859-1
@cindex ISO Latin-1
@item Character Set
The set of numeric codes used by a computer system to represent the
characters (letters, numbers, punctuation, etc.) of a particular country
or place. The most common character set in use today is ASCII (American
Standard Code for Information Interchange). Many European
countries use an extension of ASCII known as ISO-8859-1 (ISO Latin-1).
@item CHEM
A preprocessor for @code{pic} that reads descriptions of molecules
and produces @code{pic} input for drawing them. It was written in @code{awk}
by Brian Kernighan and Jon Bentley, and is available from
@code{@w{netlib@@research.att.com}}.
@item Compound Statement
A series of @code{awk} statements, enclosed in curly braces. Compound
statements may be nested.
@xref{Statements, ,Control Statements in Actions}.
@item Concatenation
Concatenating two strings means sticking them together, one after another,
giving a new string. For example, the string @samp{foo} concatenated with
the string @samp{bar} gives the string @samp{foobar}.
@xref{Concatenation, ,String Concatenation}.
@item Conditional Expression
An expression using the @samp{?:} ternary operator, such as
@samp{@var{expr1} ? @var{expr2} : @var{expr3}}. The expression
@var{expr1} is evaluated; if the result is true, the value of the whole
expression is the value of @var{expr2}, otherwise the value is
@var{expr3}. In either case, only one of @var{expr2} and @var{expr3}
is evaluated. @xref{Conditional Exp, ,Conditional Expressions}.
@item Comparison Expression
A relation that is either true or false, such as @samp{(a < b)}.
Comparison expressions are used in @code{if}, @code{while}, @code{do},
and @code{for}
statements, and in patterns to select which input records to process.
@xref{Typing and Comparison, ,Variable Typing and Comparison Expressions}.
@item Curly Braces
The characters @samp{@{} and @samp{@}}. Curly braces are used in
@code{awk} for delimiting actions, compound statements, and function
bodies.
@item Dark Corner
An area in the language where specifications often were (or still
are) not clear, leading to unexpected or undesirable behavior.
Such areas are marked in this @value{DOCUMENT} with ``(d.c.)'' in the
text, and are indexed under the heading ``dark corner.''
@item Data Objects
These are numbers and strings of characters. Numbers are converted into
strings and vice versa, as needed.
@xref{Conversion, ,Conversion of Strings and Numbers}.
@item Double Precision
An internal representation of numbers that can have fractional parts.
Double precision numbers keep track of more digits than do single precision
numbers, but operations on them are more expensive. This is the way
@code{awk} stores numeric values. It is the C type @code{double}.
@item Dynamic Regular Expression
A dynamic regular expression is a regular expression written as an
ordinary expression. It could be a string constant, such as
@code{"foo"}, but it may also be an expression whose value can vary.
@xref{Computed Regexps, , Using Dynamic Regexps}.
@item Environment
A collection of strings, of the form @var{name@code{=}val}, that each
program has available to it. Users generally place values into the
environment in order to provide information to various programs. Typical
examples are the environment variables @code{HOME} and @code{PATH}.
@item Empty String
See ``Null String.''
@item Escape Sequences
A special sequence of characters used for describing non-printing
characters, such as @samp{\n} for newline, or @samp{\033} for the ASCII
ESC (escape) character. @xref{Escape Sequences}.
@item Field
When @code{awk} reads an input record, it splits the record into pieces
separated by whitespace (or by a separator regexp which you can
change by setting the built-in variable @code{FS}). Such pieces are
called fields. If the pieces are of fixed length, you can use the built-in
variable @code{FIELDWIDTHS} to describe their lengths.
@xref{Field Separators, ,Specifying How Fields are Separated},
and also see
@xref{Constant Size, , Reading Fixed-width Data}.
@item Floating Point Number
Often referred to in mathematical terms as a ``rational'' number, this is
just a number that can have a fractional part.
See ``Double Precision'' and ``Single Precision.''
@item Format
Format strings are used to control the appearance of output in the
@code{printf} statement. Also, data conversions from numbers to strings
are controlled by the format string contained in the built-in variable
@code{CONVFMT}. @xref{Control Letters, ,Format-Control Letters}.
@item Function
A specialized group of statements used to encapsulate general
or program-specific tasks. @code{awk} has a number of built-in
functions, and also allows you to define your own.
@xref{Built-in, ,Built-in Functions},
and @ref{User-defined, ,User-defined Functions}.
@item FSF
See ``Free Software Foundation.''
@item Free Software Foundation
A non-profit organization dedicated
to the production and distribution of freely distributable software.
It was founded by Richard M.@: Stallman, the author of the original
Emacs editor. GNU Emacs is the most widely used version of Emacs today.
@item @code{gawk}
The GNU implementation of @code{awk}.
@item General Public License
This document describes the terms under which @code{gawk} and its source
code may be distributed. (@pxref{Copying, ,GNU GENERAL PUBLIC LICENSE})
@item GNU
``GNU's not Unix''. An on-going project of the Free Software Foundation
to create a complete, freely distributable, POSIX-compliant computing
environment.
@item GPL
See ``General Public License.''
@item Hexadecimal
Base 16 notation, where the digits are @code{0}-@code{9} and
@code{A}-@code{F}, with @samp{A}
representing 10, @samp{B} representing 11, and so on up to @samp{F} for 15.
Hexadecimal numbers are written in C using a leading @samp{0x},
to indicate their base. Thus, @code{0x12} is 18 (one times 16 plus 2).
@item I/O
Abbreviation for ``Input/Output,'' the act of moving data into and/or
out of a running program.
@item Input Record
A single chunk of data read in by @code{awk}. Usually, an @code{awk} input
record consists of one line of text.
@xref{Records, ,How Input is Split into Records}.
@item Integer
A whole number, i.e.@: a number that does not have a fractional part.
@item Keyword
In the @code{awk} language, a keyword is a word that has special
meaning. Keywords are reserved and may not be used as variable names.
@code{gawk}'s keywords are:
@code{BEGIN},
@code{END},
@code{if},
@code{else},
@code{while},
@code{do@dots{}while},
@code{for},
@code{for@dots{}in},
@code{break},
@code{continue},
@code{delete},
@code{next},
@code{nextfile},
@code{function},
@code{func},
and @code{exit}.
@item Logical Expression
An expression using the operators for logic, AND, OR, and NOT, written
@samp{&&}, @samp{||}, and @samp{!} in @code{awk}. Often called Boolean
expressions, after the mathematician who pioneered this kind of
mathematical logic.
@item Lvalue
An expression that can appear on the left side of an assignment
operator. In most languages, lvalues can be variables or array
elements. In @code{awk}, a field designator can also be used as an
lvalue.
@item Null String
A string with no characters in it. It is represented explicitly in
@code{awk} programs by placing two double-quote characters next to
each other (@code{""}). It can appear in input data by having two successive
occurrences of the field separator appear next to each other.
@item Number
A numeric valued data object. The @code{gawk} implementation uses double
precision floating point to represent numbers.
Very old @code{awk} implementations use single precision floating
point.
@item Octal
Base-eight notation, where the digits are @code{0}-@code{7}.
Octal numbers are written in C using a leading @samp{0},
to indicate their base. Thus, @code{013} is 11 (one times 8 plus 3).
@item Pattern
Patterns tell @code{awk} which input records are interesting to which
rules.
A pattern is an arbitrary conditional expression against which input is
tested. If the condition is satisfied, the pattern is said to @dfn{match}
the input record. A typical pattern might compare the input record against
a regular expression. @xref{Pattern Overview, ,Pattern Elements}.
@item POSIX
The name for a series of standards being developed by the IEEE
that specify a Portable Operating System interface. The ``IX'' denotes
the Unix heritage of these standards. The main standard of interest for
@code{awk} users is
@cite{IEEE Standard for Information Technology, Standard 1003.2-1992,
Portable Operating System Interface (POSIX) Part 2: Shell and Utilities}.
Informally, this standard is often referred to as simply ``P1003.2.''
@item Private
Variables and/or functions that are meant for use exclusively by library
functions, and not for the main @code{awk} program. Special care must be
taken when naming such variables and functions.
@xref{Library Names, , Naming Library Function Global Variables}.
@item Range (of input lines)
A sequence of consecutive lines from the input file. A pattern
can specify ranges of input lines for @code{awk} to process, or it can
specify single lines. @xref{Pattern Overview, ,Pattern Elements}.
@item Recursion
When a function calls itself, either directly or indirectly.
If this isn't clear, refer to the entry for ``recursion.''
@item Redirection
Redirection means performing input from other than the standard input
stream, or output to other than the standard output stream.
You can redirect the output of the @code{print} and @code{printf} statements
to a file or a system command, using the @samp{>}, @samp{>>}, and @samp{|}
operators. You can redirect input to the @code{getline} statement using
the @samp{<} and @samp{|} operators.
@xref{Redirection, ,Redirecting Output of @code{print} and @code{printf}},
and @ref{Getline, ,Explicit Input with @code{getline}}.
@item Regexp
Short for @dfn{regular expression}. A regexp is a pattern that denotes a
set of strings, possibly an infinite set. For example, the regexp
@samp{R.*xp} matches any string starting with the letter @samp{R}
and ending with the letters @samp{xp}. In @code{awk}, regexps are
used in patterns and in conditional expressions. Regexps may contain
escape sequences. @xref{Regexp, ,Regular Expressions}.
@item Regular Expression
See ``regexp.''
@item Regular Expression Constant
A regular expression constant is a regular expression written within
slashes, such as @code{/foo/}. This regular expression is chosen
when you write the @code{awk} program, and cannot be changed doing
its execution. @xref{Regexp Usage, ,How to Use Regular Expressions}.
@item Rule
A segment of an @code{awk} program that specifies how to process single
input records. A rule consists of a @dfn{pattern} and an @dfn{action}.
@code{awk} reads an input record; then, for each rule, if the input record
satisfies the rule's pattern, @code{awk} executes the rule's action.
Otherwise, the rule does nothing for that input record.
@item Rvalue
A value that can appear on the right side of an assignment operator.
In @code{awk}, essentially every expression has a value. These values
are rvalues.
@item @code{sed}
See ``Stream Editor.''
@item Short-Circuit
The nature of the @code{awk} logical operators @samp{&&} and @samp{||}.
If the value of the entire expression can be deduced from evaluating just
the left-hand side of these operators, the right-hand side will not
be evaluated
(@pxref{Boolean Ops, ,Boolean Expressions}).
@item Side Effect
A side effect occurs when an expression has an effect aside from merely
producing a value. Assignment expressions, increment and decrement
expressions and function calls have side effects.
@xref{Assignment Ops, ,Assignment Expressions}.
@item Single Precision
An internal representation of numbers that can have fractional parts.
Single precision numbers keep track of fewer digits than do double precision
numbers, but operations on them are less expensive in terms of CPU time.
This is the type used by some very old versions of @code{awk} to store
numeric values. It is the C type @code{float}.
@item Space
The character generated by hitting the space bar on the keyboard.
@item Special File
A file name interpreted internally by @code{gawk}, instead of being handed
directly to the underlying operating system. For example, @file{/dev/stderr}.
@xref{Special Files, ,Special File Names in @code{gawk}}.
@item Stream Editor
A program that reads records from an input stream and processes them one
or more at a time. This is in contrast with batch programs, which may
expect to read their input files in entirety before starting to do
anything, and with interactive programs, which require input from the
user.
@item String
A datum consisting of a sequence of characters, such as @samp{I am a
string}. Constant strings are written with double-quotes in the
@code{awk} language, and may contain escape sequences.
@xref{Escape Sequences}.
@item Tab
The character generated by hitting the @kbd{TAB} key on the keyboard.
It usually expands to up to eight spaces upon output.
@item Unix
A computer operating system originally developed in the early 1970's at
AT&T Bell Laboratories. It initially became popular in universities around
the world, and later moved into commercial evnironments as a software
development system and network server system. There are many commercial
versions of Unix, as well as several work-alike systems whose source code
is freely available (such as Linux, NetBSD, and FreeBSD).
@item Whitespace
A sequence of space, tab, or newline characters occurring inside an input
record or a string.
@end table
@node Copying, Index, Glossary, Top
@unnumbered GNU GENERAL PUBLIC LICENSE
@center Version 2, June 1991
@display
Copyright @copyright{} 1989, 1991 Free Software Foundation, Inc.
59 Temple Place --- Suite 330, Boston, MA 02111-1307, USA
Everyone is permitted to copy and distribute verbatim copies
of this license document, but changing it is not allowed.
@end display
@c fakenode --- for prepinfo
@unnumberedsec Preamble
The licenses for most software are designed to take away your
freedom to share and change it. By contrast, the GNU General Public
License is intended to guarantee your freedom to share and change free
software---to make sure the software is free for all its users. This
General Public License applies to most of the Free Software
Foundation's software and to any other program whose authors commit to
using it. (Some other Free Software Foundation software is covered by
the GNU Library General Public License instead.) You can apply it to
your programs, too.
When we speak of free software, we are referring to freedom, not
price. Our General Public Licenses are designed to make sure that you
have the freedom to distribute copies of free software (and charge for
this service if you wish), that you receive source code or can get it
if you want it, that you can change the software or use pieces of it
in new free programs; and that you know you can do these things.
To protect your rights, we need to make restrictions that forbid
anyone to deny you these rights or to ask you to surrender the rights.
These restrictions translate to certain responsibilities for you if you
distribute copies of the software, or if you modify it.
For example, if you distribute copies of such a program, whether
gratis or for a fee, you must give the recipients all the rights that
you have. You must make sure that they, too, receive or can get the
source code. And you must show them these terms so they know their
rights.
We protect your rights with two steps: (1) copyright the software, and
(2) offer you this license which gives you legal permission to copy,
distribute and/or modify the software.
Also, for each author's protection and ours, we want to make certain
that everyone understands that there is no warranty for this free
software. If the software is modified by someone else and passed on, we
want its recipients to know that what they have is not the original, so
that any problems introduced by others will not reflect on the original
authors' reputations.
Finally, any free program is threatened constantly by software
patents. We wish to avoid the danger that redistributors of a free
program will individually obtain patent licenses, in effect making the
program proprietary. To prevent this, we have made it clear that any
patent must be licensed for everyone's free use or not licensed at all.
The precise terms and conditions for copying, distribution and
modification follow.
@iftex
@c fakenode --- for prepinfo
@unnumberedsec TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
@end iftex
@ifinfo
@center TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION
@end ifinfo
@enumerate 0
@item
This License applies to any program or other work which contains
a notice placed by the copyright holder saying it may be distributed
under the terms of this General Public License. The ``Program'', below,
refers to any such program or work, and a ``work based on the Program''
means either the Program or any derivative work under copyright law:
that is to say, a work containing the Program or a portion of it,
either verbatim or with modifications and/or translated into another
language. (Hereinafter, translation is included without limitation in
the term ``modification''.) Each licensee is addressed as ``you''.
Activities other than copying, distribution and modification are not
covered by this License; they are outside its scope. The act of
running the Program is not restricted, and the output from the Program
is covered only if its contents constitute a work based on the
Program (independent of having been made by running the Program).
Whether that is true depends on what the Program does.
@item
You may copy and distribute verbatim copies of the Program's
source code as you receive it, in any medium, provided that you
conspicuously and appropriately publish on each copy an appropriate
copyright notice and disclaimer of warranty; keep intact all the
notices that refer to this License and to the absence of any warranty;
and give any other recipients of the Program a copy of this License
along with the Program.
You may charge a fee for the physical act of transferring a copy, and
you may at your option offer warranty protection in exchange for a fee.
@item
You may modify your copy or copies of the Program or any portion
of it, thus forming a work based on the Program, and copy and
distribute such modifications or work under the terms of Section 1
above, provided that you also meet all of these conditions:
@enumerate a
@item
You must cause the modified files to carry prominent notices
stating that you changed the files and the date of any change.
@item
You must cause any work that you distribute or publish, that in
whole or in part contains or is derived from the Program or any
part thereof, to be licensed as a whole at no charge to all third
parties under the terms of this License.
@item
If the modified program normally reads commands interactively
when run, you must cause it, when started running for such
interactive use in the most ordinary way, to print or display an
announcement including an appropriate copyright notice and a
notice that there is no warranty (or else, saying that you provide
a warranty) and that users may redistribute the program under
these conditions, and telling the user how to view a copy of this
License. (Exception: if the Program itself is interactive but
does not normally print such an announcement, your work based on
the Program is not required to print an announcement.)
@end enumerate
These requirements apply to the modified work as a whole. If
identifiable sections of that work are not derived from the Program,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works. But when you
distribute the same sections as part of a whole which is a work based
on the Program, the distribution of the whole must be on the terms of
this License, whose permissions for other licensees extend to the
entire whole, and thus to each and every part regardless of who wrote it.
Thus, it is not the intent of this section to claim rights or contest
your rights to work written entirely by you; rather, the intent is to
exercise the right to control the distribution of derivative or
collective works based on the Program.
In addition, mere aggregation of another work not based on the Program
with the Program (or with a work based on the Program) on a volume of
a storage or distribution medium does not bring the other work under
the scope of this License.
@item
You may copy and distribute the Program (or a work based on it,
under Section 2) in object code or executable form under the terms of
Sections 1 and 2 above provided that you also do one of the following:
@enumerate a
@item
Accompany it with the complete corresponding machine-readable
source code, which must be distributed under the terms of Sections
1 and 2 above on a medium customarily used for software interchange; or,
@item
Accompany it with a written offer, valid for at least three
years, to give any third party, for a charge no more than your
cost of physically performing source distribution, a complete
machine-readable copy of the corresponding source code, to be
distributed under the terms of Sections 1 and 2 above on a medium
customarily used for software interchange; or,
@item
Accompany it with the information you received as to the offer
to distribute corresponding source code. (This alternative is
allowed only for non-commercial distribution and only if you
received the program in object code or executable form with such
an offer, in accord with Subsection b above.)
@end enumerate
The source code for a work means the preferred form of the work for
making modifications to it. For an executable work, complete source
code means all the source code for all modules it contains, plus any
associated interface definition files, plus the scripts used to
control compilation and installation of the executable. However, as a
special exception, the source code distributed need not include
anything that is normally distributed (in either source or binary
form) with the major components (compiler, kernel, and so on) of the
operating system on which the executable runs, unless that component
itself accompanies the executable.
If distribution of executable or object code is made by offering
access to copy from a designated place, then offering equivalent
access to copy the source code from the same place counts as
distribution of the source code, even though third parties are not
compelled to copy the source along with the object code.
@item
You may not copy, modify, sublicense, or distribute the Program
except as expressly provided under this License. Any attempt
otherwise to copy, modify, sublicense or distribute the Program is
void, and will automatically terminate your rights under this License.
However, parties who have received copies, or rights, from you under
this License will not have their licenses terminated so long as such
parties remain in full compliance.
@item
You are not required to accept this License, since you have not
signed it. However, nothing else grants you permission to modify or
distribute the Program or its derivative works. These actions are
prohibited by law if you do not accept this License. Therefore, by
modifying or distributing the Program (or any work based on the
Program), you indicate your acceptance of this License to do so, and
all its terms and conditions for copying, distributing or modifying
the Program or works based on it.
@item
Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the
original licensor to copy, distribute or modify the Program subject to
these terms and conditions. You may not impose any further
restrictions on the recipients' exercise of the rights granted herein.
You are not responsible for enforcing compliance by third parties to
this License.
@item
If, as a consequence of a court judgment or allegation of patent
infringement or for any other reason (not limited to patent issues),
conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License. If you cannot
distribute so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you
may not distribute the Program at all. For example, if a patent
license would not permit royalty-free redistribution of the Program by
all those who receive copies directly or indirectly through you, then
the only way you could satisfy both it and this License would be to
refrain entirely from distribution of the Program.
If any portion of this section is held invalid or unenforceable under
any particular circumstance, the balance of the section is intended to
apply and the section as a whole is intended to apply in other
circumstances.
It is not the purpose of this section to induce you to infringe any
patents or other property right claims or to contest validity of any
such claims; this section has the sole purpose of protecting the
integrity of the free software distribution system, which is
implemented by public license practices. Many people have made
generous contributions to the wide range of software distributed
through that system in reliance on consistent application of that
system; it is up to the author/donor to decide if he or she is willing
to distribute software through any other system and a licensee cannot
impose that choice.
This section is intended to make thoroughly clear what is believed to
be a consequence of the rest of this License.
@item
If the distribution and/or use of the Program is restricted in
certain countries either by patents or by copyrighted interfaces, the
original copyright holder who places the Program under this License
may add an explicit geographical distribution limitation excluding
those countries, so that distribution is permitted only in or among
countries not thus excluded. In such case, this License incorporates
the limitation as if written in the body of this License.
@item
The Free Software Foundation may publish revised and/or new versions
of the General Public License from time to time. Such new versions will
be similar in spirit to the present version, but may differ in detail to
address new problems or concerns.
Each version is given a distinguishing version number. If the Program
specifies a version number of this License which applies to it and ``any
later version'', you have the option of following the terms and conditions
either of that version or of any later version published by the Free
Software Foundation. If the Program does not specify a version number of
this License, you may choose any version ever published by the Free Software
Foundation.
@item
If you wish to incorporate parts of the Program into other free
programs whose distribution conditions are different, write to the author
to ask for permission. For software which is copyrighted by the Free
Software Foundation, write to the Free Software Foundation; we sometimes
make exceptions for this. Our decision will be guided by the two goals
of preserving the free status of all derivatives of our free software and
of promoting the sharing and reuse of software generally.
@iftex
@c fakenode --- for prepinfo
@heading NO WARRANTY
@end iftex
@ifinfo
@center NO WARRANTY
@end ifinfo
@item
BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW@. EXCEPT WHEN
OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
PROVIDE THE PROGRAM ``AS IS'' WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE@. THE ENTIRE RISK AS
TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU@. SHOULD THE
PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
REPAIR OR CORRECTION.
@item
IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES.
@end enumerate
@iftex
@c fakenode --- for prepinfo
@heading END OF TERMS AND CONDITIONS
@end iftex
@ifinfo
@center END OF TERMS AND CONDITIONS
@end ifinfo
@page
@c fakenode --- for prepinfo
@unnumberedsec How to Apply These Terms to Your New Programs
If you develop a new program, and you want it to be of the greatest
possible use to the public, the best way to achieve this is to make it
free software which everyone can redistribute and change under these terms.
To do so, attach the following notices to the program. It is safest
to attach them to the start of each source file to most effectively
convey the exclusion of warranty; and each file should have at least
the ``copyright'' line and a pointer to where the full notice is found.
@smallexample
@var{one line to give the program's name and an idea of what it does.}
Copyright (C) 19@var{yy} @var{name of author}
This program is free software; you can redistribute it and/or
modify it under the terms of the GNU General Public License
as published by the Free Software Foundation; either version 2
of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE@. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place --- Suite 330, Boston, MA 02111-1307, USA.
@end smallexample
Also add information on how to contact you by electronic and paper mail.
If the program is interactive, make it output a short notice like this
when it starts in an interactive mode:
@smallexample
Gnomovision version 69, Copyright (C) 19@var{yy} @var{name of author}
Gnomovision comes with ABSOLUTELY NO WARRANTY; for details
type `show w'. This is free software, and you are welcome
to redistribute it under certain conditions; type `show c'
for details.
@end smallexample
The hypothetical commands @samp{show w} and @samp{show c} should show
the appropriate parts of the General Public License. Of course, the
commands you use may be called something other than @samp{show w} and
@samp{show c}; they could even be mouse-clicks or menu items---whatever
suits your program.
You should also get your employer (if you work as a programmer) or your
school, if any, to sign a ``copyright disclaimer'' for the program, if
necessary. Here is a sample; alter the names:
@smallexample
@group
Yoyodyne, Inc., hereby disclaims all copyright
interest in the program `Gnomovision'
(which makes passes at compilers) written
by James Hacker.
@var{signature of Ty Coon}, 1 April 1989
Ty Coon, President of Vice
@end group
@end smallexample
This General Public License does not permit incorporating your program into
proprietary programs. If your program is a subroutine library, you may
consider it more useful to permit linking proprietary applications with the
library. If this is what you want to do, use the GNU Library General
Public License instead of this License.
@node Index, , Copying, Top
@unnumbered Index
@printindex cp
@summarycontents
@contents
@bye
Unresolved Issues:
------------------
1. From ADR.
Robert J. Chassell points out that awk programs should have some indication
of how to use them. It would be useful to perhaps have a "programming
style" section of the manual that would include this and other tips.
2. The default AWKPATH search path should be configurable via `configure'
The default and how this changes needs to be documented.
Consistency issues:
/.../ regexps are in @code, not @samp
".." strings are in @code, not @samp
no @print before @dots
values of expressions in the text (@code{x} has the value 15),
should be in roman, not @code
Use tab and not TAB
Use ESC and not ESCAPE
Use space and not blank to describe the space bar's character
The term "blank" is thus basically reserved for "blank lines" etc.
The `(d.c.)' should appear inside the closing `.' of a sentence
It should come before (pxref{...})
" " should have an @w{} around it
Use "non-" everywhere
Use @code{ftp} when talking about anonymous ftp
Use upper-case and lower-case, not "upper case" and "lower case"
Use alphanumeric, not alpha-numeric
Use --foo, not -Wfoo when describing long options
Use findex for all programs and functions in the example chapters
Use "Bell Laboratories", but not "Bell Labs".
Use "behavior" instead of "behaviour".
Use "zeros" instead of "zeroes".
Use "Input/Output", not "input/output". Also "I/O", not "i/o".
Use @code{do}, and not @code{do}-@code{while}, except where
actually discussing the do-while.
The words "a", "and", "as", "between", "for", "from", "in", "of",
"on", "that", "the", "to", "with", and "without",
should not be capitalized in @chapter, @section etc.
"Into" and "How" should.
Search for @dfn; make sure important items are also indexed.
"e.g." should always be followed by a comma.
"i.e." should never be followed by a comma, and should be followed
by `@:'.
The numbers zero through ten should be spelled out, except when
talking about file descriptor numbers. > 10 and < 0, it's
ok to use numbers.
In tables, put command line options in @code, while in the text,
put them in @samp.
When using @strong, use "Note:" or "Caution:" with colons and
not exclamation points. Do not surround the paragraphs
with @quotation ... @end quotation.
Date: Wed, 13 Apr 94 15:20:52 -0400
From: rsm@gnu.ai.mit.edu (Richard Stallman)
To: gnu-prog@gnu.ai.mit.edu
Subject: A reminder: no pathnames in GNU
It's a GNU convention to use the term "file name" for the name of a
file, never "pathname". We use the term "path" for search paths,
which are lists of file names. Using it for a single file name as
well is potentially confusing to users.
So please check any documentation you maintain, if you think you might
have used "pathname".
Note that "file name" should be two words when it appears as ordinary
text. It's ok as one word when it's a metasyntactic variable, though.
Suggestions:
------------
Enhance FIELDWIDTHS with some way to indicate "the rest of the record".
E.g., a length of 0 or -1 or something. May be "n"?
Make FIELDWIDTHS be an array?
What if FIELDWIDTHS has invalid values in it?
Reference in New Issue View Git Blame Copy Permalink