summaryrefslogtreecommitdiff
path: root/usr.bin/patch
AgeCommit message (Collapse)Author
2007-09-02use calloc() to avoid malloc(n * m) overflows; checked by djm canacar jsgTheo de Raadt
2007-05-31convert to new .Dd format;Jason McIntyre
2007-05-31- note some apps which are xpg4, not posixJason McIntyre
- note that some apps are optional to posix after some discussion w/ otto
2007-04-18adds a detailed synopsis to the man page; sorts options inIgor Sobrado
the description list and adds missing arguments; syncs usage() written with a lot of useful advice by jmc@ ok by jmc@
2006-03-11type police and assorted cleanup. From Lionel Fourquaux; ok ray@Otto Moerbeek
2005-11-14constuct -> constructMiod Vallat
2005-06-20umask juggling not needed; with Lionel Fourquaux.Otto Moerbeek
ok millert@ espie@
2005-05-16Do not call out mkdir -p, but reuse the code from mkdir(1).Marc Espie
Removes the possibility of nasty stuff happening thanks to unquoted arguments in system(1). Also plug a small memory leak. Problems noticed by Lionel Fourquaud. okay millert@, deraadt@
2004-11-19Write reject file in unified format if the original diff was inOtto Moerbeek
unified format. ok deraadt@
2004-11-19Allow for path names containing spaces and other funny chars (exceptOtto Moerbeek
tab) by scanning for the tab that separates the date from the pathname. If no tab is found, revert back to the old behaviour. Posix says there should be a space between the pathname and the date, but all known implementations of diff(1) use a tab. ok kurt@ deraadt@
2004-09-14remove unused variableTheo de Raadt
2004-08-05spacingTheo de Raadt
2004-07-09Properly detect if a patch already has been applied, even if thereOtto Moerbeek
are no eols at eofs involved. Tested by quite some people. ok deraadt@
2004-06-18Unbreak handling of standard diffs (which have an empty context).Otto Moerbeek
ok @millert
2004-06-14Teach patch(1) to detect if a diff creating a file (or one addingOtto Moerbeek
lines to an empty file) already has been applied; be a bit less chatty about empty contexts. Testing hshoexer@ ok millert@
2004-01-28Fix a segv when patch cannot find a file and the user indicatesOtto Moerbeek
to use the file anyway. Problem found by david@ ok david@ millert@
2003-12-08MAP_FILE is the default and MAP_PRIVATE has to be given or produces evil ↵Michael Shalayeff
warnings in debug kernel
2003-11-21madvise(sequential) the mmap()ed file and plug an fd leak on mmap() failure; ↵Michael Shalayeff
millert@ ok
2003-10-31o fairly major man page reorganizationTodd C. Miller
o add --posix option for strict POSIX conformance o change filename determination to match GNU patch Tested by various people, including a full ports build by naddy@
2003-09-28realloc(3) cleanup.Otto Moerbeek
ok cloder@ tedu@
2003-08-15Add license from patch.c to original source files missing a license.Otto Moerbeek
ok deraadt@
2003-08-12Fix no newline at end of file case for Plan B.Otto Moerbeek
ok millert@ tedu@
2003-08-10Do not add an extra newline at the end if the last line of the inputOtto Moerbeek
file contains no newline and the diff does not touch the last line. Contributions from millert@. ok millert@ tedu@
2003-08-10Warn if the diff is a context or unified diff and the contextOtto Moerbeek
is empty. In this case, a previously applied patch cannot be detected. ok millert@ tedu@
2003-08-08Avoid scanning the input file twice.Otto Moerbeek
ok millert@ tedu@
2003-08-05spacingTheo de Raadt
2003-08-05- Use mmap(2) instead of malloc(3) and read(2) to get an image of the inputOtto Moerbeek
file into memory. Some suggestions by tedu@. - Add a debug option to force using Plan B. ok millert@ deraadt@
2003-08-01- use stdbool.h instead of roll-your-own booleansOtto Moerbeek
- fix some -Wall warnings - fix asserts: in some cases remove them, in other cases they have become Internal errors or detection of malformed patch files. - fix some free() related code ok millert@ tedu@
2003-07-31Historically, patch would treat a bare -p as -p0. This contradictsTodd C. Miller
POSIX and GNU patch has also removed this, so we will too. No objections on icb (no one even seemed to know about this "feature").
2003-07-31Unbreak relative directory handling. Tweak from millert@Otto Moerbeek
ok millert@
2003-07-31Print a maximum of one invalid line number warning per patch in a patch file.Otto Moerbeek
Thanks to espie@ for spotting the problem. ok millert@ henning@ espie@
2003-07-30Add POSIX -i option; tedu@ OKTodd C. Miller
2003-07-30Make prompting POSIX-compliant. POSIX states "The patch utilityTodd C. Miller
shall write a prompt to standard output and request a filename interactively from the controlling terminal (for example, /dev/tty)." OK deraadt@ and otto@
2003-07-29o add pathnames.hTodd C. Miller
o ignore empty TMPDIR environment variable o strip any trailing slashes from TMPDIR otto@ OK
2003-07-28minor knf and cleanups; otto okTheo de Raadt
2003-07-28Don't treat consecutive slashes as path components; matches POSIXTodd C. Miller
OK otto@ and deraadt@
2003-07-28More cleanup: sprinkled some const, removed ugly EXTERN/INTERN.h,Otto Moerbeek
some rewriting of code to make it more readable and more KNF. ok millert@ tedu@ deraadt@
2003-07-28Be more exact on how backup files are handled.Otto Moerbeek
ok millert@ tedu@
2003-07-28Be more exact on how backup files are handled.Otto Moerbeek
ok millert@ tedu@
2003-07-28Make patch(1) exit value match POSIX and be consistent with diff.Todd C. Miller
Comments and OK from otto@
2003-07-25Add POSIX -b option but don't change the default beahvior wrt backupsTodd C. Miller
yet. The old -b option is now -z (matches GNU patch).
2003-07-24"options" not "switches"; jmc@ OKTodd C. Miller
also update Larry's email address while I am at it ;-)
2003-07-24Don't talk about ~ and # suffixes for orig and reject files since weTodd C. Miller
use .orig and .rej.
2003-07-24mdoc version;Jason McIntyre
ok millert@
2003-07-23New version of invalid line number fix. Passes patch(1) regressions.Otto Moerbeek
ok millert@ deraadt@
2003-07-22Missed this in the rejname commit.Todd C. Miller
2003-07-22Make rejname[] static to patch.c and crank its size to NAME_MAX+1Todd C. Miller
2003-07-22Use getopt_long() to parse options instead of rolling our own.Todd C. Miller
OK deraadt@ and otto@
2003-07-22just about nothingTheo de Raadt
2003-07-22nicerTheo de Raadt