Clean up deleted files.

This commit is contained in:
mycroft 1993-04-23 02:09:22 +00:00
parent 20d732999d
commit efcd17acc4
4 changed files with 0 additions and 1479 deletions

File diff suppressed because it is too large Load Diff

View File

@ -1,57 +0,0 @@
This version of ed is not strictly POSIX compliant, as described in the
POSIX 1003.2 Draft 11.2 document (see the file `ed-1003.2'). BSD commands
have been implemented wherever they do not conflict with the POSIX
standard. For backwards compatibility, the POSIX rule that says a range
of addresses cannot be used where only a single address is expected has
been relaxed (see the file `legal-addrs').
The BSD commands included are:
1) the `W' command for appending text to an existing file,
2) the `s' command (i.e., s[rgp]*) to repeat a previous substitution,
3) the `z' command for scrolling through the buffer, and
4) the `wq' command for exiting after a write.
BSD line addressing syntax (i.e., `^' and `%' - see the file
`legal.addrs') is also recognized.
The POSIX interactive global commands `G' and `V' are extended to support
multiple commands, including `a', `i' and `c'. The command format is the
same as for the global commands `g' and `v', i.e., one command per line
with each line, except for the last, ending in a backslash (\).
If crypt is available, files can be read and written using DES encryption.
The `x' command prompts the user to enter a key used for encrypting/
decrypting subsequent reads and writes. If only a newline is entered as
the key, then encryption is disabled. Otherwise, a key is read in the
same manner as a password entry. The key remains in effect until
encryption is disabled. For more information on the encryption algorithm,
see the bdes(1) man page. Encryption/decryption should be fully compatible
with SunOS DES.
An extension to the POSIX file commands `E', `e', `r', `W' and `w' is that
<file> arguments are processed for backslash escapes, i.e., any character
preceded by a backslash is interpreted literally. A trailing backslash
is ignored. If the first unescaped character of a <file> argument is a
bang (!), then the rest of the line is interpreted as a shell command,
and no escape processing is performed by ed. This rule does not apply to
the 'f' command (described below).
While the default file name cannot be set to a shell command (i.e.,
!<shell-cmd>) via any of the `r', `w', and `e' commands, POSIX evidently
does not preclude using the `f' command for this purpose. Upon reading
or writing the default file when its name begins with a bang, ed
interprets the rest of the line as a shell command and attempts to execute
it. Escape processing is not performed in this case.
If ed is invoked with a name argument prefixed by a bang, then the
remainder of the argument is interpreted as a shell command. To invoke
ed on a file whose name starts with bang, prefix the name with a backslash.
--
The following commands are neither part of POSIX 1003.2 D11/2 nor are
they supported in this version of ed.
The vi command: (addr1,addr2) !<shell-cmd> which replaces a range of
lines with the output of a shell command is not supported.
The vi command: [rwe] !!, where !! is replaced by the previous
!<shell-cmd> is not supported.

View File

@ -1,59 +0,0 @@
The following describes line addressing syntax. This is a combination of
both BSD and POSIX.
An address represents the number of a line in the editor buffer. It is
constructed from one of the bases in the table below, optionally followed
by an absolute offset. The offset may include any combination
of digits, operators (i.e., `+', `-' and `^') and whitespace.
Addresses are read from left to right, and their values are computed
relative to the current address (i.e., the number reported by the command
`.=').
One exception to the rule that addresses represent line numbers is the
address 0. This means "before the first line," and is legal wherever it
makes sense.
An address range is two addresses separated either by a comma or
semi-colon. The value of the first address in a range cannot exceed the
value of the the second. If an n-tuple of addresses is given where
n > 2, then the corresponding range is determined by the last two addresses
in the n-tuple. If only one address is expected, then then the last
address is used (see the examples below).
Each address in a comma-delimited range is interpreted relative to the
current address. In a semi-colon-delimited range, the first address is
used to set the current address, and the second address is interpreted
relative to the new current address.
The following address symbols are recognized:
. - current line
$ - last line
- - previous line; may be repeated with cumulative effect
-<n> - <n>th previous line, where <n> is a non-negative number
^ - previous line; may be repeated with cumulative effect (BSD)
^<n> - <n>th previous line, where <n> is a non-negative number
+ - next line; may be repeated with cumulative effect
+<n> - <n>th next line, where <n> is a non-negative number
% - first through last lines, i.e., equivalent to: 1,$ (BSD)
, - first through last lines, i.e., equivalent to: 1,$ (POSIX)
; - current through last lines, i.e., equivalent to: .,$ (POSIX)
/pat/ - next line containing pattern pat
?pat? - previous line containing pattern pat
[0-9]* - line <n>, where <n> is a number in the range [0,$]
<ws><n> - whitespace <ws> followed by a number <n> is interpreted as +<n>
'<lc> - line previously marked by k command, where <lc> is a
lower case letter
Examples:
The address tuple `4;+1' is equivalent to the address range `4,5'.
The address tuple `1;+2,+3' is equivalent to the range `3,4'.
The address tuple `1;+2;+3' is equivalent to the range `3,6'.
The command `1;m+' swaps the first and second lines.
The command `0;/pat/r file' reads `file' to after the line containing the
pattern pat.
The command `t0;/pat/' copies the current line to after the first line
containing the pattern pat.
The last two examples above violate the POSIX rule that says a range cannot
be used where at most a single address is expected. This rule has been
relaxed to enable such syntax.

View File

@ -1,49 +0,0 @@
The following commands are recognized. These are a combination of BSD
and POSIX. The POSIX version of a command always overrides any BSD
version. The commands are shown together with the number of addresses
that each accepts.
!<shell-cmd> - run <shell-cmd> via sh {1} (POSIX)
($)= - print line number
(.)a - append text to the buffer
(.,.)c - change lines in the buffer
(.,.)d - delete lines from the buffer
E <file> - edit a file unconditionally {1}
E !<shell-cmd> - edit the standard output of <shell-cmd> {1} (POSIX)
e <file> - edit a file {1}
e !<shell-cmd> - edit the standard output of shell-cmd {1} (POSIX)
f <file> - set the default file to <file> {1}
(1,$)G/pat/ - interactively edit lines matching a pattern (POSIX) {1}
(1,$)g/pat/<cmd-list> - apply <cmd-list> to lines matching a pattern
H - print explanations of all errors (POSIX)
h - print explanation of the last error (POSIX)
(.)i - insert text in the buffer
(.,.+1)j - join lines in the buffer
(.)k<lc> - mark a line for later '<lc> addressing
(.,.)l - print lines unambiguously to stdout
(.,.)m(.) - move lines in the buffer
(.,.)n - enumerate lines to stdout (POSIX)
P - toggle command prompt (POSIX)
(.,.)p - print lines to stdout
Q - quit ed unconditionally
q - quit ed
($)r <file> - read a file {1}
($)r !<shell-cmd> - read the stdout of <shell-cmd> {1} (POSIX)
(.,.)s/pat/sub/g - substitute matching text in a line via a template
(.,.)s/pat/sub/<n> - substitute the <n>th match in a line (POSIX)
(.,.)s[rgp]* - repeat the last substitution (BSD)
(.,.)t(.) - copy (transfer) lines in the buffer
u - undo the last command (POSIX)
(1,$)V/pat/ - interactively edit lines not matching a pattern (POSIX) {1}
(1,$)v/pat/<cmd-list> - apply <cmd-list> to lines not matching a pattern
(1,$)W <file> - append lines to a file (BSD)
(1,$)W !<shell-cmd> - write lines to the stdin of <shell-cmd> (POSIX)
(1,$)w <file> - write lines in the buffer to a file
(1,$)w !<shell-cmd> - write lines to the stdin of <shell-cmd> (POSIX)
x - prompt for an encryption key (SunOS) {1}
(.+1)z<rows> - page <rows> lines at a time to stdout (BSD)
(.)<newline> - print line to stdout (BSD)
--
{1} See the file `extensions' for additional information on these
commands.