a2ee17b347
Keep things portable (requested by joerg) by not depending on reallocarr
...
and instead doing the overflow check ourselves.
2018-06-18 18:33:31 +00:00
759529ae76
PR/53368: Thomas Barabosch: Potential integer overflow in usr.bin/patch/inp.c
2018-06-16 00:40:14 +00:00
019d365a3a
Pass -S to ed(1) so that patches containing ! commands don't run commands.
...
Real cause of CVS-2018-0492:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=894667 )
2018-04-05 18:50:10 +00:00
01869ca4d2
Remove workaround for ancient HTML generation code.
2017-07-03 21:28:48 +00:00
a08385b414
Explain what "-V none" does.
2015-11-07 18:26:55 +00:00
3f2a71f305
paths.h is not used.
2015-11-07 18:11:21 +00:00
282efd00b8
Fix two typos.
2015-08-02 12:19:12 +00:00
429d474275
remove reference to SCCS which is not supported anymore.
...
XXX: pullup-7
2015-08-02 03:19:02 +00:00
e139f8ac36
from bitrieg:
...
Substitution commands might contain a newline in the replacement pattern
(escaped with a backslash before it), causing patch's understanding of
the state the ed child process is in to diverge from reality. This can
lead to patch unwillingly feeding '!' (execute shell command) lines to
ed. Finding out how to do this is left as an exercise to the reader.
XXX: pullup-7
2015-07-30 21:47:51 +00:00
2703f4af80
Use absolute paths for RCS commands (Martin Natano)
2015-07-24 18:56:44 +00:00
8bfaa68572
From Martin Natano @bitrig: Use execve(2) instead of system to apply patches
...
that require rcs command execution instead system(3) to avoid malicious
filenames in patches causing bad things to happen. In the process, lose SCCS
support. It is not like we are shipping sccs commands for that to work.
2015-07-24 18:56:00 +00:00
938d39fd7f
PR/49422: Ryo ONODERA: patch(1) cannot handle context diffs
2014-11-27 15:07:23 +00:00
59de200486
Avoid coredumps when the linenumbers in the hunks turn negative.
...
From: http://marc.info/?l=openbsd-tech&m=141693055412785
XXX: pullup 7
2014-11-26 00:31:32 +00:00
e9d34c2f4d
mention -V none
2014-04-11 17:41:09 +00:00
0b1777bf87
recognize -V none
2014-04-11 17:30:03 +00:00
b5609ab012
Add comma after i.e.
...
From Bug Hunting.
2013-01-29 09:30:11 +00:00
1ac719d147
Remove unnecessary Bk/Ek pairs from SYNOPSIS.
...
No effective change except where I used the opportunity to sort options
and/or option descriptions.
2012-04-08 22:00:37 +00:00
d5bd94639a
Consistently use __dead and __printflike.
2011-09-06 18:25:14 +00:00
51ab93526e
Exit with EXIT_FAILURE for invalid arguments. PR 43517.
2011-03-26 21:45:48 +00:00
2724492d78
Print version string to stdout instead of stderr;
...
it is user-requested and not an error.
2010-10-02 19:31:14 +00:00
06c042e9bc
Plug a memory leak. From Igor Zinovik.
2010-01-10 01:10:16 +00:00
6e3add81a1
Do not try to mmap a zero length file.
2009-10-21 17:16:11 +00:00
57b5269dad
Explicitly include fcntl.h for O_CREAT to not depend on namespace
...
pollution.
2009-06-05 19:55:43 +00:00
d96ab870e3
Attempt to fix previous to address a coredump.
2009-04-18 03:25:34 +00:00
1e4c32c4f5
Fix sign-compare issue
2009-04-13 00:07:26 +00:00
7e69823738
Don't workaround ancient macro argumentent limitation with .Xo/.Xc.
2009-03-10 13:57:08 +00:00
d5b2c9a825
Update patch to the version used by DragonFly and derived from OpenBSD.
...
Major changes are:
- better detection of double applied patches
- rejects remain unified diffs for unified patches
- far less limitations, e.g. patch lines may be arbitrary long
This addresses PR standards/11220 by changing patch -b behavior to be
POSIX compliant. Old behavior can be obtained using --suffix, which
works since NetBSD 1.4. pkgsrc has been adjusted accordingly.
2008-09-19 18:33:34 +00:00
69bdd817cc
Add error checking to use of fputs in creating output file in
...
temprorary directory. Previously, fputs was used without checking the
return value, leading to silent truncation when the temporary
filesystem was full (such as can easily happen when migrating to
tmpfs).
2008-08-10 22:36:23 +00:00
bd560fde31
Convert to using raise_default_signal(3).
2007-10-14 04:54:34 +00:00
1e1f0dd3ab
PR/34627: Lubomir Kundrak: patch(1) segfaults, when fed from stdin in batch
...
or force mode
2006-09-26 16:36:07 +00:00
04e2c897d2
Coverity CID 3510: Don't leak memory.
2006-05-24 16:43:35 +00:00
bd21f84917
Coverity CID 1253, 1254: Don't assign the return of read or write to size_t.
...
It will never be negative.
2006-04-09 19:12:17 +00:00
f5a5211062
It is silly to creat() and close a file just to stat it. open/fstat/close
...
instead checking for errors. From Coverity CID 1276.
2006-04-09 19:06:34 +00:00
5800a5519e
Coverity CID 2078, 2079, 2080: Plug memory leaks.
2006-04-09 19:03:32 +00:00
de60b309dc
Sort options. Add -C to usage.
2005-03-25 23:55:02 +00:00
eb0439126c
Fix typo and drop trailing whitespace.
2005-03-25 23:49:28 +00:00
4f9669ed8d
add --dry-run like gnu patch.
2005-03-25 23:00:55 +00:00
33131ab95f
Fix an array overrun in option parsing.
...
Fixes PR 26732, PR 26775, PR 28416 and PR 28589.
2004-12-09 18:06:10 +00:00
58d3abc94f
Add (unsigned char) cast to ctype functions
...
Stop infinite loop if get lower case char with no upper case equiv.
2004-10-30 21:52:09 +00:00
224d89c7db
Remove debug printf (hi mycroft!).
2004-08-14 12:53:35 +00:00
ee776b9f85
Parse multiple options that are specified together; e.g. "-fs".
...
XXX It is not clear to me why patch(1) doesn't get getopt_long(3).
2004-08-06 14:54:26 +00:00
38b27406f9
remove unused macros
2003-07-30 08:51:55 +00:00
75c32b46eb
use bounded string op
2003-07-30 08:51:04 +00:00
c0970755a6
pedantic check around unsafe strncpy.
...
XXX this code needs a serious rewrite
2003-07-12 13:53:08 +00:00
ef4b3d620c
copyright missing in files, copy them from README
2003-07-12 13:47:42 +00:00
3e51d2b759
Const poisoning.
2003-07-08 01:55:35 +00:00
158e4448b0
Shorten long lines.
...
Some whitespace changes to get closer to KNF.
2003-06-01 22:04:29 +00:00
f4b0e187e9
Fix some lint warnings.
2003-05-30 23:08:12 +00:00
7e5434fe39
Remove stuff that are not needed any longer.
2003-05-30 22:33:58 +00:00
b86c323d39
Simplify handling of memory allocation, and make sure all return values
...
from malloc etc. are handled.
This removes the old behavior to retry the operation with a less memory-
consuming method in case malloc failed (this mechanism has never really
worked, and is hard to test. Besides, it is less useful now than it was
20 years ago when the code was written...)
2003-05-30 18:14:13 +00:00
christos