summary refs log tree commit diff
Commit message (Collapse)AuthorAgeFilesLines
* 34926: _fc: Remove a spurious 2Mikael Magnusson2015-04-192-1/+5
|
* 34921: handle error in recursive par_event().Peter Stephenson2015-04-172-0/+5
| | | | Here documents in an interrupted list caused bad juju.
* 34920: info on git pullPeter Stephenson2015-04-172-2/+13
|
* 34919: update interrupt parse error for ZLE cases.Peter Stephenson2015-04-173-2/+5
| | | | Covers both send-break and push-input.
* 34899: completion: _vim: add nvim to #compdefDaniel Hahler2015-04-172-1/+6
| | | | nvim is used by NeoVim and is fairly compatible.
* 34905: no parse error after keyboard interrupt.Peter Stephenson2015-04-174-6/+14
| | | | | | | | Handled generally, though only showing up in special nested cases. Also fix ZLE so it doesn't cancel the interrupt flag when not actually returning from a local keymap.
* 34902: Doc: Explain the numeric argument and be consistent about what we call itMikael Magnusson2015-04-164-20/+37
|
* 34901: Test for the previous commitPeter Stephenson2015-04-152-0/+19
|
* 34900: assignment before an "exec".Peter Stephenson2015-04-152-7/+19
| | | | | | Without POSXIBUILTIN: restore after, so we only get side effects. With POSXIBUILTIN: keep set variable
* unposted: rewrite Nularg math handling to be like other casesPeter Stephenson2015-04-152-2/+9
|
* 34892 (slightly tweaked): math evaluation fixPeter Stephenson2015-04-153-3/+9
| | | | | An empty expression resulting from substitution includes a Nularg, which needs handling the same as an empty string.
* 34887: Fix POSIX_BUILTINS with assignment.Peter Stephenson2015-04-153-7/+40
| | | | | | | | | | | In the form var=val command special-builtin-or-func the var is restored after execution, unlike the case where "command" is absent. Clear up case in code that handles this. Add tests.
* users/20159: completion: zstyle: Add 'show-ambiguity'Daniel Shahaf2015-04-132-0/+4
|
* 34886: completion: git: Complete rebase --exec argumentDaniel Shahaf2015-04-132-1/+4
|
* 34885: completion: git: Sort "commit object" completions most recent firstDaniel Shahaf2015-04-132-1/+4
|
* 34885: completion: git: Fix another instance of the 34671 bug fixed in 34814Daniel Shahaf2015-04-132-1/+8
|
* 34880: POSIX_TRAPS docHan Pingtian2015-04-132-1/+3
|
* 34875: CSH_JUNKIE_HISTORY docHan Pingtian2015-04-132-2/+4
|
* users/20078: HIST_BEEP docHan Pingtian2015-04-132-1/+5
|
* 34874: completion for cabalClint Adams2015-04-102-0/+14
|
* 34748: completion: git: update list of builtintoolsDaniel Hahler2015-04-102-9/+21
|
* 34850: make URLs in html/pdf docs clickableJun-ichi Takimoto2015-04-079-32/+42
|
* 34851: fix thinko from 34093 that short-circuited some "whence -m" searchesBarton E. Schaefer2015-04-062-5/+12
|
* 34837: avoid loss of original file path when applying colon-modifiers in ↵Barton E. Schaefer2015-04-032-2/+8
| | | | glob qualifiers
* 34840: minor format fixes in prompt.yoJun-ichi Takimoto2015-04-032-4/+8
|
* 34836: fix description of '%-0<<' in prompt.yoJun-ichi Takimoto2015-04-032-1/+6
|
* 34814: completion: git: Fix bug introduced by 236da69Daniel Shahaf2015-04-012-1/+8
| | | | Based on a patch by Daniel Hahler <git@thequod.de>.
* 34822: correction to GLOB_SUBST docHan Pingtian2015-04-012-1/+6
|
* users/20058: improve device completion in ip completion.Peter Stephenson2015-03-314-39/+58
| | | | | We were too eager to regard strings already on the command line as possible devices.
* 34817: Catch some errors earlier when reading history.Peter Stephenson2015-03-293-1/+13
| | | | | | | | Mostly for the case of an interrupt. Don't try to process words when we know something's gone wrong. Also abort history reading earlier on an interrupt.
* 34792: langinfo: Fix pointer typeTheo Buehler2015-03-292-1/+6
|
* 34762: completion: git: add short option for '--dir-diff' (_git-difftool)Daniel Hahler2015-03-292-1/+6
|
* 34781: add implicit space after alias expansion of tokens that form words ↵Barton E. Schaefer2015-03-283-1/+15
| | | | | | without spacing fixes crash when using completion immediately following such an alias
* 34804: refine POSIX_ALIAS change to preserve old behavior of [[ ]] conditionalsBarton E. Schaefer2015-03-283-4/+10
|
* 34788: refine errflag handling in cmd_or_math()Barton E. Schaefer2015-03-272-1/+5
|
* 34785: fix typo in SHINSTDIN docHan Pingtian2015-03-272-1/+3
|
* 34787: Add more history testsPeter Stephenson2015-03-272-0/+45
|
* 34784: fix old bug with history word selectionPeter Stephenson2015-03-263-11/+26
|
* 34779: tests for 34776Barton E. Schaefer2015-03-253-1/+31
|
* 34769: incorrect doc of process time calculationHan Pingtian2015-03-252-1/+6
|
* 34776: improve suppression of alias expansions from history.Peter Stephenson2015-03-253-35/+17
| | | | | Now uses the mechanism in use for other forms of suppression of sections of input.
* 34774: fix %prep and use read from a no-op pipe to force wait for the timeoutBarton E. Schaefer2015-03-253-2/+7
|
* 34772: add new test suite file for history; fix "read" usage in comptestBarton E. Schaefer2015-03-233-1/+26
|
* 34759: improve implementation of last commitPeter Stephenson2015-03-223-7/+10
|
* 34758: fix yet more history / command subst interaction.Peter Stephenson2015-03-223-1/+38
| | | | | | In general we need to wind back over the history text input inside command substitution because there's no level of the input mechanism between history and the lexer.
* users/20034: adjust PREFIX and SUFFIX for unambiguous_cursor on second ↵Barton E. Schaefer2015-03-212-0/+16
| | | | completion attempt with GLOB_COMPLETE
* 34753: work around NFS problem in tests.Peter Stephenson2015-03-202-9/+32
| | | | | Use tmp directory if it helps us check file time comparisons. If it doesn't we're no worse off than we were.
* 34752: another fix for history expansion in cmd substPeter Stephenson2015-03-202-2/+20
|
* 34742: history expansion inside command substitution failed.Peter Stephenson2015-03-192-3/+11
| | | | Needs the case of alias expansion separating out.
* 34739: git completion: Fix regression in ce80a92Daniel Shahaf2015-03-192-1/+6
| | | | | This restores the message when completing a commit object name that isn't one of the most recent 20 commits. e.g., 'git checkout deadbeef<TAB>'.