summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2015-09-09Extend further-mitigate-test-suite-hangs.patch to mitigate a race condition ↵Axel Beckert1-3/+18
in E01options.ztst
2015-09-08Update changelog for new upstream release candidateAxel Beckert1-0/+7
2015-09-08Merge tag 'zsh-5.1-test-1' into debianAxel Beckert29-48/+323
2015-09-06unposted: make 5.1-test-1 with updates for 5.1.1Peter Stephenson5-4/+13
2015-09-06unposted: typo in commentMikael Magnusson1-1/+1
2015-09-05unposted: Minor documentation tweaks.Daniel Shahaf3-3/+7
2015-09-0536410: vcs_info git: Present merge heads as patchesDaniel Shahaf3-1/+24
This shows, during 'git merge', the revision hashes of the "remote" head (the one that will become second parent of the commit) in the %m expando. Review-by: Frank Terbeck
2015-09-04unposted: typo in change logPeter Stephenson1-1/+1
2015-09-0436421: fix breakage in pattern change in 36415Peter Stephenson2-1/+6
2015-09-0436417: fix/update top level texinfo menuJun-ichi Takimoto2-8/+22
2015-09-0436415: remap bytes from invalid multibyte characters.Peter Stephenson3-6/+47
These now go to 0xdc00 + index. If wchar_t is a Unicode code point, this is by construction an invalid character within the Unicode range. If it isn't, we would hope the result was no worse than the current fudge.
2015-09-0336416: ^C in getzlequery() just aborts query.Peter Stephenson3-1/+15
Also logical but possibly invisible fix for error propagated back from listing interface.
2015-09-0336403: type -w += suffix aliasDaniel Shahaf3-1/+12
This makes 'type -w' distinguish suffix aliases from regular aliases, like bare 'type' already does. Use-case: detecting programmatically whether the command word is indeed a valid command word (see <https://github.com/zsh-users/zsh-syntax-highlighting/issues/126>)
2015-09-02Add ChangeLog entry for tests from previous commitBarton E. Schaefer1-0/+2
2015-09-0236393: process queued signals during dotrap()Barton E. Schaefer3-0/+18
2015-09-0236378: skip directories when looking for files to autoloadPeter Stephenson2-1/+8
2015-09-02unposted: fix egregious typoPeter Stephenson1-1/+1
2015-09-0236399: Shell code text with multiple here documents wasn't shown properly.Peter Stephenson3-2/+38
2015-09-0136347: Fix appending empty array to associationsMikael Magnusson3-3/+21
2015-09-0136380: Avoid gcc case syntaxMikael Magnusson1-1/+2
2015-09-0136376: handle signals during read1char() so it is possible to interrupt ↵Barton E. Schaefer2-1/+11
correct/correctall prompts
2015-09-0136365: magic-space could delete the reset of the line.Peter Stephenson2-1/+6
If we got a LEXERR, which didn't necessarily mean an actual error, just something the rather nasty ctxtlex() got confused by, we didn't scan the reset of the line.
2015-09-01unposted: Update internal version to 5.1-dev-0.Peter Stephenson2-2/+7
Avoids installation clash of develpment build with 5.1 release.
2015-09-0136353: _git: Unbreak after previous.Daniel Shahaf2-1/+6
2015-09-01Upload to unstable as 5.1-1Axel Beckert1-3/+9
2015-08-3036328 + 36340: _git-cherry-pick: Complete other branches onlyMateusz Karbowy2-14/+57
2015-08-31Remove pre-#768937 legacy code from zsh{,-dev,-dbg,-doc}.postinstAxel Beckert4-32/+0
Thanks to lintian for making me aware of this via its (experimental) maintainer-script-may-use-dir_to_symlink_helper warning. Actually these (now removed) code snippets did even the opposite of what is defined in debian/*.maintscript for the according arch:any packages. Either the failing rmdir saved us from the symlink being set or the debian/*.maintscript generated snippets coming after the now removed snippets fixed it just immediately again. The only arch:all package in this list is zsh-doc which still has a symlink /usr/share/doc/zsh-doc to zsh-common. But since this change (in 5.0.2-3) predates Jessie as well as Trusty, we should be able to safely remove that code now anyways. If for some reason, this commit is starting troubles with zsh-doc upgrades, you might want to add a file named debian/zsh-doc.maintscript to the source package and put this single line (without the leading spaces) into it: dir_to_symlink /usr/share/doc/zsh-doc zsh-common 5.0.2-3~ This line is untested though.
2015-08-30Remove lintian overrides of spelling-error-in-binary false positivesAxel Beckert2-4/+0
Seem no more emitted, hence cause unused-override lintian warnings.
2015-08-3036338: Replace "5.0.9" with "5.1" in source commentsAxel Beckert3-2/+6
2015-08-30Update further-mitigate-test-suite-hangs.patch to remove additional ↵Axel Beckert1-10/+5
cherry-picked fixes
2015-08-30Update changelog for new upstream releaseAxel Beckert1-0/+6
2015-08-30Merge tag 'zsh-5.1' into debianAxel Beckert11-17/+119
2015-08-30Relase zsh 5.1Peter Stephenson2-2/+6
2015-08-2936321: _debcheckout: New completion.Daniel Shahaf2-0/+26
2015-08-2836306: _subversion: Complete --show-item valuesDaniel Shahaf2-0/+23
2015-08-28users/20475: document cdr stores unique entry of most recent usePeter Stephenson2-3/+10
2015-08-28users/20466 plus comment: change test for skipping strftime extensions.Peter Stephenson2-1/+12
This test is known to fail on Solaris, but seems to work in other common implementations. Hence there seems no point in having the release fall over on this test. Needs further investigation later.
2015-08-2836314: Avoid using short_loops syntax in distributed filesMikael Magnusson2-2/+5
2015-08-2836302: attr: Fix compilation with libcap 2.24Mikael Magnusson2-3/+7
2015-08-27unposted: _subversion: Avoid SHORT_LOOPSDaniel Shahaf2-1/+4
2015-08-2736304: _git-cat-file: Fix regression in 36237Daniel Shahaf2-0/+6
This broke 'git cat-file blob origin:Etc/<TAB>'.
2015-08-27Upload to experimental as 5.0.8-test-3-1Axel Beckert1-2/+3
2015-08-27cf. Axel Beckert 36292: discard even more extraneous stderr in "hang" testBarton E. Schaefer2-1/+6
2015-08-27Update changelog for new upstream release candidate 5.0.8-test-3Axel Beckert1-2/+2
2015-08-27Merge tag 'zsh-5.0.8-test-3' into debianAxel Beckert10-18/+120
2015-08-27Throw away even more errors in further-mitigate-test-suite-hangs.patchAxel Beckert2-6/+9
2015-08-2636288: refine 36825, slightly different test for redirect contextBarton E. Schaefer2-1/+4
2015-08-2636285: update 36025, context cannot be command and redirect at the same ↵Barton E. Schaefer2-4/+8
time, redirect wins
2015-08-2636282: discard stderr of "hang" testBarton E. Schaefer2-1/+5
2015-08-26Update patch to integrate Bart's suggestion on further test suite fixesAxel Beckert1-6/+9
Hopefully fixes test suite failure which so far only show up in the Jenkins builds. Source of the patch: http://www.zsh.org/mla/workers/2015/msg02205.html