aboutsummaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Fix chomping: only make it default on line-processing binariesLaurent Bercot2020-11-2414-33/+37
|
* Trivial bugfixLaurent Bercot2020-11-241-1/+1
|
* Better forstdinLaurent Bercot2020-11-241-8/+5
| | | | | | | No need for the complexity: the important distinction is between "eof after reading something" and "eof right away". 0 is a natural fit for eof after some data, and 1 is a natural fit for immediate eof. Anything else can be scripted around this.
* Make forstdin only split on newlines by defaultLaurent Bercot2020-11-241-1/+1
|
* New forstdin; new trap; prepare for 2.7.0.0Laurent Bercot2020-11-232-79/+79
|
* trap: fix off-by-one in argvs initLaurent Bercot2020-08-041-4/+2
| | | | | Oh hey, gcc 10.2 actually came with a useful new warning that helps catch off-by-ones!
* Add -I/-i options to envfileLaurent Bercot2020-05-261-9/+41
|
* New envfile format; prepare for 2.6.1.0Laurent Bercot2020-05-262-18/+136
|
* Simpler error management in el_substituteLaurent Bercot2020-05-161-3/+1
|
* bugfix: el_substitute could incorrectly share stralloc pointersLaurent Bercot2020-05-161-1/+1
|
* Extend runblock functionality (thanks Profpatsch)Laurent Bercot2020-01-271-5/+12
|
* Rename cd/umask to execline-cd/execline-umask, make symlinks, etc.Laurent Bercot2019-12-305-2/+12
|
* Add posix-umask; prepare for 2.6.0.0Laurent Bercot2019-12-243-0/+212
|
* Align dollarat option priority to rightmostLaurent Bercot2019-12-241-1/+1
|
* Don't depend on PATH_MAX being definedLaurent Bercot2019-10-201-0/+2
|
* bugfix: el_execsequence() needs to exec argv2 when posix_spawn() failsLaurent Bercot2019-10-191-5/+16
|
* Make execlineb itself honor EXECLINE_STRICTLaurent Bercot2019-10-121-0/+16
|
* Make posix-cd's CDPATH testing even more horribleLaurent Bercot2019-10-111-1/+1
|
* posix-cd: correct stralloc len after canonicalizationLaurent Bercot2019-10-091-1/+2
|
* Fix posix-cd's ENOMEM exit codeLaurent Bercot2019-10-091-1/+1
|
* Add --enable-pedantic-posix, posix-cd, prepare for 2.5.3.0Laurent Bercot2019-10-093-15/+195
| | | | Also make wait posix-compliant and update doc.
* Adapt to new stopwatch APIsLaurent Bercot2019-09-062-4/+2
|
* Make trap and wait use a stopwatchLaurent Bercot2019-09-042-0/+2
|
* Add envfile; prepare for 2.5.1.0Laurent Bercot2019-02-122-0/+94
|
* with pthread support for nsssLaurent Bercot2018-07-281-0/+1
|
* Add nsss support (for homeof)Laurent Bercot2018-07-281-0/+1
|
* Adapt to skalibs-2.7.0.0, prepare for 2.5.0.1Laurent Bercot2018-07-213-0/+3
|
* Use sig0_scan in trap.cLaurent Bercot2018-03-211-4/+3
|
* Remove import, prepare for 2.5.0.0Laurent Bercot2018-03-166-37/+9
|
* Remove env_getLaurent Bercot2017-12-252-5/+6
|
* bugfix: segfault in execlineb when both $@ and script are emptyLaurent Bercot2017-11-171-0/+1
|
* More conversion to xpathexec, make exit code change consistent across binariesLaurent Bercot2017-08-222-2/+4
|
* Propagate fd_close() changesLaurent Bercot2017-07-162-7/+0
|
* Use xpathexec_* functionsLaurent Bercot2017-05-1833-83/+49
|
* Duh... don't forget half the NSIG instancesLaurent Bercot2017-04-281-2/+2
|
* Use the new SKALIBS_NSIG in trap.c to accommodate OpenBSD snowflakinessLaurent Bercot2017-04-281-8/+8
|
* trap.c: explicitly include skalibs/nonposix.h for proper NSIG definition ↵Laurent Bercot2017-04-281-0/+1
| | | | on, you guessed it, the BSDs!
* trap.c: fix usage of NSIG (which is max+1, not max)Laurent Bercot2017-04-281-10/+10
|
* Spell it out for OpenBSD: no, we don't want to trap SIGKILL or SIGSTOP, ↵Laurent Bercot2017-04-281-0/+2
| | | | everything is going to be okay, now now, please don't cry
* Fix invalid variable name check in getpid (thanks jjk)Laurent Bercot2017-04-131-1/+1
|
* bugfix: catch empty blocks in multidefineLaurent Bercot2017-04-112-0/+2
|
* Replace memcpy with memmove in el_substitute (copying into the same string!)Laurent Bercot2017-04-051-1/+1
|
* More superfluous headers cleanupLaurent Bercot2017-03-135-6/+1
|
* Fix a few omissionsLaurent Bercot2017-03-1214-48/+42
|
* Fix typo in the import obsolescence warningLaurent Bercot2017-03-081-1/+1
|
* Make it build with skalibs 2.5.0.0Laurent Bercot2017-03-085-7/+9
|
* Adapt to skalibs 2.5.0.0 APILaurent Bercot2017-03-0849-252/+189
|
* Revert size_t *, put a temporary XXX until skalibs changesLaurent Bercot2017-01-101-1/+1
|
* Types fix: first passLaurent Bercot2017-01-0735-114/+144
| | | | | | | | This pass makes variable size_t-ready, so everything works when the prototypes are fixed in skalibs. Some code uses "unsigned int *" where it should be "size_t *"; it cannot be changed now, but it's been marked with XXX. It must change at the same time as the skalibs API.
* bugfix: runblock -r segfaulted on too few arguments (thanks jjk)Laurent Bercot2016-10-161-1/+1
|
generated by cgit v1.3.1 (git 2.54.0) at 2026-07-19 15:28:03 +0000