Actually there were two bugs:
- Add REG_NOTBOL after the first substitution.
- Handle the rm_so == rm_eo == 0 case, where in a substitution such
as 's/bzzzt/z*/g' the first time z* matches nothing.
The appropriate entry in /etc/group as returned by getgrnam() is
used to determine if 'su root' may be permitted, rather than
checking if membership exists in the result of getgroups().
The following changes were made regarding the behaviour of the special
group for 'su root'
* allow for definition of SUGROUP (defaults to "wheel") to override group name.
* use getgrnam(SUGROUP) instead of getgrgid(0).
* only scan getgrnam(SUGROUP)->gr_mem when checking for group membership.
* be more specific as to why 'su root' failed
NOTE: If a user's primary group is SUGROUP, and they're not a member
of SUGROUP in /etc/group, they will not be able to su.
auto-login ftp URL is used when $ftp_proxy is defined. It now prints:
Auto-login using ftp URLs isn't supported when using $ftp_proxy
Should solve rest of [bin/3643].
Whilst this is inconsistant with the behaviour when $ftp_proxy isn't
defined, the following constraints apply:
* it's not possible to support ftp URL auto-login when $ftp_proxy is
defined, since it uses http not ftp, and you can't `login' to http
servers; fudging this would require a major rewrite of ftp anyway)
* silently ignoring $ftp_proxy and not using it if an ftp auto-login
URL is given is bad user interface design)
* mrg & others will harrass me if I remove support for autologin ftp URLs
when $ftp_proxy isn't defined, even though it made the behaviour
consistant whether $ftp_proxy was set or not.
Only information leaks now are:
* if '-s -s' is used (only allow s/key users, and force s/key use),
then "login incorrect" will be given if a non-s/key user (or
non-existant user) attempts to login; no password will be prompted
for.
XXX: maybe this should be fixed, but further analysis is required.
* an s/key user will be reminded in the "Password" prompt that they
have an s/key. Therefore it would be possible to determine if a user
is active on the machine if they have an s/key.
XXX: maybe an option is required to control this behaviour
_PASSWORD_WARNDAYS from <pwd.h>). For non-root users, enforce expiry when
it happens. From Simon Gerraty <sjg@zen.void.oz.au> in [bin/935].
* Check for group 0 in process's current group membership (as returned by
getgroups(2)), instead of just looking at the entry for wheel in /etc/group.
Based on code by Dan Caresone <dan@oink.geek.com.au> in [bin/792], and
also solves [bin/2466].
* Clean up to pass -Wall
* if the user has an s/key, provide a reminder in the password prompt
* if '-s' is given once, force a user that has an s/key to use it
* if '-s' is given more than once, only permit s/key logins
- rsh/rcmd combinations don't die sometimes, and spin in poll loops
+ detect errors from read/write etc, don't ignore them in some cases
+ use INFTIM instead of 0 in poll
+ detect invalid file descriptors in poll
+ use varargs/stdarg as appropriate
+ use posix signal calls
+ EWOULDBLOCK -> EAGAIN
directory specified, and add it to sysIncPath only if it exists.
However, afterwards make tested for the presence of a -m option by
checking to see if sysIncPath was an empty list, and assumed that
the -m option was not used if it was empty. This obviously breaks
if -m specified a non-existent directory. So I have added a flag
that is set if the -m option is used, and I test that instead.
Make used to only use the search path for nodes that were pure
sources (not targets of other sources). This has been corrected
and now gnu-autoconf generated Makefiles work in directories other
than the source one.
- Suffix transformation rescanning:
Suffix transformations (.c.o:; cc ...) were only recognized in
the past when both suffixes were members of the suffix list.
Thus a sequence like:
.z.b:
echo ${.TARGET}
.SUFFIXES: .z
would cause .z.b: to be inserted as a regular target (and the main
target in this case). Other make programs always add rules that
start with a period in the transformation list and never consider
them as targets. We cannot do that (consider .depend files) so we
resort to scanning the list of the current targets every time a
suffix gets added, and we mutate existing targets that are now
valid transformation rules into transformation rules. If the
transformed target was also the main target, we set the main target
to be the next target in the targets list.
I.e. if you had a line in your Makefile:
../foo.o: foo.c
`..' would be added in the search path. The addition of such paths has
been now disabled. If a pathname contains a slash, then the directory
where such a file is found is not added to the search path. Of course
this eliminates most (all?) use of this function.
* differentiate between being connected, and being logged in
* cleanup some text messages
* support username & password ftp URLs (ftp://user:pass@host/) in non-proxy
situations; assume proxy supports it for proxy situations.
* cd to / before performing any autofetch transfers
* use strncasecmp in URL parsing. fix from <Todd.Miller@courtesan.com>
("MSDOSFS mount" is too big for %12s).
Note that this breaks naive parser scripts, but they would not parse
the earlier "MSDOSFS mount" line correctly, anyway.
* add "more" & "less" as synonyms for "page"
* move editline setup code into controlediting(), and call appropriately.
only setup setup terminal if going into interactive mode. inspired
by Todd Miller <todd.miller@courtesan.com>
- Fix globbing so that patterns that don't have a matching number of [] or {}
don't get expanded. (before the [ case got expanded to nothing!) This is
disabled.
* only echo "Passive mode" in verbose mode; scripts that use ftp
may get unwanted output otherwise
* disable progress bar and modification time preservation when
retreiving to a non-regular files. fixes progress bar getting in
way of "get file /dev/tty"
* setup el_init() et al if editing is set, not if fromatty.
TODO: migrate this to a function, and call if editing is turned on later
in the session. also implement edit_cleanup if editing is turned off
* call el_set() after setting SIGWINCH handler. This fixes the problem
when suspending in a non-cbreak shell (e.g, csh) would trash your tty mode.
* reset interactive mode correctly in auto_fetch() mget mode
* support remglobbing of auto_fetch arguments
* new flag - '-e'; disable editing
* "page file" == "get file |${PAGER-less}"
Bugfixes/cleanup:
* consistently use a trailing '.' on messages
* code cleanup, including buffer overrun fixes, use puts
and putchar in places, etc (inspired by OpenBSD mods)
* disable progress bar when local-file is a pipe or '-'
* skip \r in http headers
* fix remote ftpd slash bug more elegantly (so it works with ////)
* abort_remote(): check if cout==NULL before using it. should fix [bin/3273]
* fixed up cosmetic problems when complete_remote() generated errors from the
remote server (such as "no files found", "login with user and pass", ...)
done by adding extra argument to remglob(), which is a pointer to an error
buffer to put messages in rather than printing to stdout.
rwho: no hosts in /var/rwho.
message like ruptime does, for consistency. (Similarly, exit with exit
code 0, though i'm not sure that's "correct" for either.)
Add code to netstat to use libkvm to for kernel variables "mclbytes"
and "msize', and if found, use those for netstat -m rather than
compiled-in defaults.
1. ${.*} variables did not get expanded in dependencies.
2. expanded ${.*} variables in .USE dependencies can cause tree
restructuring; handle it.
3. in compat mode, expand .USE before evaluating the list of targets,
instead of doing .USE expansions on demand, because they can cause
tree restructuring.
EXPORTABLE_SYSTEM. Additionally, since it doesn't actually use libtelnet
for any of the 'interesting' features, it shouldn't be using libtelnet
at all (because that adds some serious headaches and problems, on ELF
systems with shared libraries, which cannot be easily addressed and
don't actually _have_ to be). Now it simply pulls in genget.c from
the libtelnet source, and builds it locally.
quotes. E.g. if left quote was "``" and right quote was "''", then
"``foo ``bar'' baz''" would yield "foo `bar' baz" when it should yield
"foo ``bar'' baz".
it uses pbstr() instead of doing the characters individually (in
reverse order) with putback().
Fix dodiv() so that divert(<out-of-range stream>) doesn't cause "divnum" to
return 0, but rather the stream number specified to divert(). Macros
like PUSHDIVERT/POPDIVERT rely on this. The new behaviour is compatible
with vendor m4 implementations.
barring any more little things people want added ...]
New features:
* progressmeter is now asynchronous, so "stalled" transfers can be
detected. "- stalled -" is displayed instead of the ETA in this case.
When the xfer resumes, the time that the xfer was stalled for is
factored out of the ETA. It is debatable whether this is better than
not factoring it out, but I like it this way (I.e, if it stalls for 8
seconds and the ETA was 30 seconds, when it resumes the ETA will still
be 30 seconds).
* verbosity can be disabled on the command line (-V), so that in auto-fetch
mode the only lines displayed will be a description of the file, and
the progress bar (if possible)
* if the screen is resized (and detected via the SIGWINCH signal), the
progress bar will rescale automatically.
Bugs fixed:
* progress bar will not use the last character on the line, as this can
cause problems on some terminals
* screen dimensions (via ioctl(TIOCWINSZ)) should use stdout not stdin
* progressmeter() used some vars before initialising them
* ^D will quit now. [fixes bin/3162]
* use hstrerror() to generate error message for host name lookup failure.
* use getcwd instead of getwd (it should have been OK, but why tempt fate?)
* auto-fetch transfers will always return a positive exit value upon failure
or interruption, relative to the file's position in argv[].
* remote completion of / will work, without putting a leading "///".
This is actually a bug in ftpd(1), where "NLST /" prefixes all names
with "//", but fixing every ftpd(1) is not an option...
probably never be completely supported here (since its internal vs.
external symbol handling is so annoying), but at least we can recognize
the ECOFF binaries and print something useful, on machines that use
them.
* Command line editing via editline(3) library.
* Context sensitive command and file completion, including remote files.
Enhancements to auto-fetch feature:
* Support for http:// URLs using the http protocol, including proxy HTTP
support via $htty_proxy if it's defined.
* The connection is kept open between successive files on the same host.
(obviously, this does not count for http requests.)
* Return value of ftp is 0 on no error, or the offset in argv[] of the file
which failed (i.e., argv[x] failed, ftp returns x).
* If the path in an ftp URL or classic format line has a trailing '/',
cd to the path and enter interactive mode. Fixes [bin/3011], albiet
requiring the user to help ftp in determining the operation.
Other changes:
* '-P port' works for normal ftp, and is the default for all classic style
auto-fetch transfers and for ftp URLs that don't specify the port.
(previously it would just work for the first xfer.)
* Some code moved into separated files along logical divisions.
* Editing and completion can be compiled out with -DSMALLFTP.
submitted by Heiko W.Rupp.
closes pr-2448
Also, moved Alan Turing's birthday from calendar.computer to
calendar.birthday, removed duplicate Thoreau entry and cleaned up date
ordering in calendar.birthday
of the old zzhack-access macros, which with the current macro definitions
would cause a cast value to be used as an lvalue, which is a gcc extension,
but _not_ what we want.
and address() macros with new macros of the same names (to get values),
and macros with those names but with "set" prepended and which take a
second argument (to set values). The new macros use type casts to do
type conversion, which is much cleaner than using a union and making
endianness-related hacks to get the right 'short's and 'char's from
the right array indices. (YUCK! again, for good measure.)
* in autofetch mode, don't retry a file if the connection failed - skip it
and move to the next file. [bin/3051, Matt Green]
* only print error messages from SIZE or MDTM if the user used 'size' or
'modt' (respectively). prevents extraneous warnings during normal transfers
when connected to a site which doesn't support these (behaviour prior to
last commit), but still allows error feedback to specific user requests
for this info (which the last commit broke).
* 'account': only accept one optional argument
* if invoked as pftp, default to passive mode on (from FreeBSD)
* remove leading '0 ' in progress bar - looked ugly
* use warn instead of perror
* use strncpy when src isn't known to have safe length
* remglob(): use mkstemp() to prevent symlink games, and don't override
_PATH_TMP, use it as the prefix to the temp file
- fix the variable substitution code in make [PR/2748]
1. change s/a/b/ so that it substitutes the first occurance of the
pattern on each word, not only the first word.
2. add flag '1' to the variable substitution so that the substitutions
get performed only once.
***THIS IS AN INCOMPATIBLE CHANGE!***
Unfortunately there was no way to make things consistent without
modifying the current behavior. Fortunately none of our Makefiles
depended on this.
OLD:
VAR = aa1 aa2 aa3 aa4
S/a/b/ = ba1 aa2 aa3 aa4
S/a/b/g = bb1 bb2 bb3 bb4
NEW:
VAR = aa1 aa2 aa3 aa4
S/a/b/ = ba1 ba2 ba3 ba4
S/a/b/1 = ba1 aa2 aa3 aa4
S/a/b/g = bb1 bb2 bb3 bb4
S/a/b/1g = bb1 aa2 aa3 aa4
- add regexp variable substitution via 'C/foo/bar/' [PR/2752]
- add variable quoting via the ${VAR:Q} modifier. This is useful when running
recursive invocations of make(1):
make VAR=${VAR:Q}
will always work... (This may prove useful in the kernel builds...) [PR/2981]
used as lint1 input. That involves having lint pass the '-' through
to the cpp which preprocesses the lint1 input, and having lint1's
scanner recognize a cpp filename "" as "{standard input}".
convert them to pointers. If they're zero, they're converted (to
NULL pointers) regardless of size. If they're non-zero, they can't
be converted (without a cast). This matches the behavior of other
version of lint, e.g. the lints on Digital UNIX and HP-UX.
* recognize that pointers to identical unnamed and untyped structs,
unions, and enums are, in fact, identical. This is done by tagging
each of unnamed and untyped structure, union and enum with a unique
position of creation, which is used as a unique identifier that
when determine whether or not a pair of structures, unions, or enums
are identical.
unions, and enums are, in fact, identical. This is done by tagging
each of unnamed and untyped structure, union and enum with a unique
position of creation, which is used as a unique identifier that
when determine whether or not a pair of structures, unions, or enums
are identical.
* accept the file name '-' to indicate that standard input is to be
used as lint1 input. That involves having lint pass the '-' through
to the cpp which preprocesses the lint1 input, and having lint1's
scanner recognize a cpp filename "" as "{standard input}".
unions, and enums are, in fact, identical. This is done by tagging
each of unnamed and untyped structure, union and enum with a unique
position of creation, which is used as a unique identifier that
when determine whether or not a pair of structures, unions, or enums
are identical.
variable declaration, in addition to within function & variable
declarations and function bodies. I think this comes close enough
to what gcc does to be considered "correct enough." Certainly, it
fixes the problem for the couple of cases where this is a problem
in our tree.
kit, then hacked on by Matt Thomas <matt@3am-software.com>, then by me.
This runs, but it's in serious need of cleaning and/or a fair bit of
reworking. See the README file for more information, and a list of things
to do.
* implement 'progress bar/meter' (inspired by ncftp). use 'progress' to
toggle on. it will display current file size to 5 digits, automatically
determining suffix (up to 16384 P (petabytes) == 2^64).
* 'ls' now uses NLST (unadorned listing), a la older ftp clients. 'dir'
still does LIST (long listing). idea from John Nemeth <jnemeth@cue.bc.ca>
bug fixes:
* return first line of reply in reply_string[] from getreply(), instead
of last line. This fixes [bin/741] (parsing of SYST), and also means
that SIZE and MDTM messages will be parsed correctly if they're longer
than 1 line.
* parse URL-style auto-ftps that have no filename correctly
(e.g, ftp://host, ftp://host/, ftp://host/dir/). pointed out by
Jaromir Dolecek <dolecek@saruman.ics.muni.cz>
* pass the correct size array in 2nd arg of utimes() when setting the
modification time
friedman@gnu.ai.mit.edu (Noah Friedman) made to his modified ftp:
- implement "lpwd" - local pwd
- implement "preserve" - toggle preserving of file modification
times on retrieved files
- allow for explicit "on" or "off" arg to toggle commands
- "exit" synonym for "quit", "msend" synonym for "mput"
- in confirmation mode, allow 'a' (yes to rest of current command),
and 'p' (turn off prompt mode, as if 'prompt off' was done,
effective immediately)
- "modtime" returns time formatted as localtime, not GMT
Bug fixes:
- check for extraneous args on commands
- cleanup const usage, line formatting
- create 0 length temporary file in remglob() to prevent symlink games
(from OpenBSD)
- check length of filename of ~/.netrc (from OpenBSD)
If a user wishes to change a password on a system running YP, and
the master server is not running rpc.yppasswdd, chpass(1) would fail,
even if the user had a local entry. Fix this by checking for local
entry if master is not running rpc.yppasswd iff we defaulted to using
YP (not invoked with "-y").
XXX Unlike the similar change to passwd(1), this one duplicates some
XXX code (makes an attempt to contact rpc.yppasswdd early). This is
XXX a side-effect of the structure of this program. chpass(1) could
XXX use a re-write.
If a user wishes to change a password on a system running YP, and
the master server is not running rpc.yppasswdd, passwd(1) would fail,
even if the user had a local entry. Fix this by checking for local
entry if master is not running rpc.yppasswd iff we defaulted to using
YP (not invoked as "yppasswd" or with "-y").
Printing only those with the name "intr" is too restrictive, because it
means that devices can only have one interrupt, which makes little
sense for many devices and absolutely no sense for several common busses.
- don't echo 'ACCT' parameter when debugging (a la 'PASS')
- Fix checking of directory access for "/foo", the parent
directory is "/", not "" (from FreeBSD)
- remove trailing whitespace on lines
- add any missing NetBSD tags
- cleanups to man page, includinging sorting options description
feature additions:
- variable sized hash marks (from [bin/683], but done in the hash command
as an optional arg)
- more user-friendly transfer time printing (from FreeBSD, with mods)
- '-p' command line option to jump into PASV mode (closes [bin/2857],
but with an option rather than checking argv[0])
- SIGINFO support for printing xfer stats when sending/receiving requests
- '-P port' for changing the port to connect to (from thorpej@netbsd.org)
- '-a': bypass normal login, and try anonymous login (from OpenBSD
via thorpej)
- autofetch files via url (ftp://...) or "classic" (host:/file)
(from OpenBSD via thorpej)
- 'ftp' synonymous with 'open' (from FreeBSD)
-o (the reverse of this, also the default)
- use vis(3) in vputc() instead of handcrufted function (from OpenBSD).
- move gecos expansion into expandusername() (a la sendmail's buildfname).
A generic version of this last bit in libutil would be useful...
- cleanup the code, fix prototypes, etc.
- don't allow quotes in terminal names (for -s and -S) as they confuse
the shell. whitespace is currently forbidden for the same reason.
as there isn't any orthoganol way of escaping this, forbid it all,
which makes [bin/306] unfixable.
- use str(r)char instead of (r)index
Fix PR/2839: su will not build with Kerberos.
- Also:
-Don't coredump when $TERM is not set.
-Add prototypes, remove local old style declarations of system
functions.
-Recognize shells that contain "csh" as being csh alike.
-Don't build with SKEY unconditionally. Obey bsd.own.mk.
make that fail on some machines. Rather, since the code wants to access
the data both ways, use a structure and treat that structure as an array
of chars where appropriate.
referenced only by their basename and not by their full pathname. This
breaks when .PATH or MAKEOBJDIR are used. There might be Makefiles around
that try to work around this bug by prepending ${.CURDIR} to the sources,
and they should be found and fixed. Also a lot of the gunk in suff.c that
was attempting to work around the same problem could be removed.
YP server. This is required if the passwd database is to stay in sync
if this program is run on the YP server. Note, local passwd database
operations can still be performed by passing the -l flag.
YP server. This is required if the passwd database is to stay in sync
if this program is run on the YP server. Note, local passwd database
operations can still be performed by passing the -l flag.
Also, some minor cleanup and RCS id police.
by Mike Grupenhoff <kashmir@umiacs.umd.edu> in PR #2485.
While I'm here, update for modern libkvm interfaces, and be more
careful to check for error conditions reported by kvm_read(3).
for 'netstat -w <delay>') to 100, from 10. 10 was definitely not sufficient
for many hosts; 100 should be for most if not all. This code really should
dynamically allocate the information structures, based on the number of
interfaces in the kernel, account for interfaces that are added or removed,
etc., but given its current structure that would require substantial changes.
interface of the given name, print an error message and exit.
This whole section of code needs to be re-thought, if interfaces
can be dynamically added or removed.
efficient machine independent external format. The new format is
like what is used on svr4 systems, except it uses fixed sized
fields and a constant byte order.
- avoid divide-by-zero trap when specifying small widths
- do not overrun entry array when printing output tables
- cleanup storage allocation for entries
- use err/warn etc.
of long-standing bugs:
- Actually deal with the fact that the kernel ifnet list is
a TAILQ; it just happened to work before.
- Use kvm_openfiles() instead of kvm_open(). The code passed
arguments to kvm_open() as if it were kvm_openfiles(), but
apparently went unnoticed since the prototypes are the same.
Amusing bit: there were XXX's in the code which seemed to
apologize for a verbose libkvm, when it happened to be a
bug in netstat!
From PR #2308 by Stephen J. Roznowski <sjr@zombie.ncsc.mil>.
This change is not strictly necessary, as these files are only used to
test yacc. But I could not think of any compelling reason not to make
these change when everything else in the source tree was updated.
order to match the function prototype and to work with compilers that
default to unsigned chars.
Compare the getopt() return value with -1 instead of EOF.
mikel