aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* Clean up recent fix for #1892Gravatar ridiculousfish2015-01-17
| | | | Restore 906d235 and simplify how __fish_restore_status works
* Introduce tests for #1892Gravatar ridiculousfish2015-01-17
|
* Ignore error message when not on a Gentoo systemGravatar Jelte Fennema2015-01-17
|
* Don't define unnamespaced internal function.Gravatar Konrad Borowski2015-01-17
|
* Fix job_or_process_extent with command substitutionGravatar Daniel K2015-01-17
|
* Fix: eval should preserve previous $status if the evaluated block does not ↵Gravatar Jorge Bucaran2015-01-17
| | | | | | change it Empty functions may return 1 when eval is used due to the $status not being correctly preserved inside the function definition.
* Use svg instead of png to get better image qualityGravatar Peter Dave Hello2015-01-17
|
* Unexport CMD_DURATIONGravatar Ben Hamilton2015-01-16
| | | | | | | | | | | | | | Valid uses of this environment variable don't really include passing it to subsequent child processes. I confirmed the fix with: function fish_prompt echo "cmd duration [$CMD_DURATION] " end cmd duration [0] sleep 2 cmd duration [2002]
* Make octal/hex escapes in printf and echo output literal bytesGravatar ridiculousfish2015-01-15
| | | | Fixes #1894
* Use iothread_perform variant that doesn't take a completion callbackGravatar ridiculousfish2015-01-14
| | | | Removes some ugly NULL casts
* Reap jobs before calling select() in job_continue()Gravatar Kevin Ballard2015-01-12
| | | | | | | | | | Prior to b0e09303a, simple jobs like `printf "%s\n" $line | read word _` never hit the call to select() because they were reaped in the SIGCHLD signal handler. With that commit, the signal handler no longer reaps children, and a job like that would enter select() and hit the 10000μs timeout before discovering that the job was already complete. Fixes #1884.
* Ignore user-supplied fd redirections above 2 for builtinsGravatar ridiculousfish2015-01-08
| | | | | Prevents e.g. specifying an fd which corresponds to the history file as the stdin for builtin_source
* Rework file descriptor handlingGravatar ridiculousfish2015-01-07
| | | | | | | | Remove global array of file descriptors, in favor of relying on CLO_EXEC exclusively. Also correctly implement "pipe avoidance" so that fd redirections do not conflict with pipes.
* Use scoped_push to save and restore real_ioGravatar ridiculousfish2015-01-07
| | | | Fixes a dangling pointer
* Add more expository comments to eval, and remove a useless parameterGravatar ridiculousfish2015-01-07
|
* Add tests to verify pipes do not conflict with fd redirectionsGravatar ridiculousfish2015-01-04
|
* Ignore xccheckout filesGravatar ridiculousfish2015-01-04
|
* Mark libnotify FDs as CLO_EXECGravatar ridiculous_fish2015-01-04
| | | | Fixes a fd leak on OS X
* Remove useless signal-checking loop in job_continueGravatar ridiculousfish2014-12-29
| | | | This loop has always been nonsense.
* Stop blocking signals within job_promoteGravatar ridiculousfish2014-12-29
| | | | | | Since we no longer inspect the job list from a signal handler, we don't need to muck with signals when modifying the job list
* Rename events and simplify signal event handlingGravatar ridiculousfish2014-12-29
| | | | | | | | | - Rename 'events' to 's_event_handlers' - Stop inspecting the s_event_handlers list upon receiving a signal. Instead, maintain the set of signals that are observed in a separate static array. This lets us avoid mucking with STL data structures in a signal handler, and so avoid blocking signals in event.cpp
* Change fish_paginate to paginate both stdout and stderr.Gravatar Andreas Heiduk2014-12-28
|
* Update pythonista.fish: make virtualenv aware that prompt is setGravatar volnt2014-12-28
| | | | | | Closes #1823, closes #1795. Signed-off-by: David Adam <zanchey@ucc.gu.uwa.edu.au>
* test: note that the square bracket form is availableGravatar David Adam2014-12-28
| | | | As requested in https://github.com/fish-shell/fish-shell/issues/1850
* Close <code> before <pre> in fish_indentGravatar ridiculousfish2014-12-24
|
* Use <pre><code> for fish_indent HTML outputGravatar ridiculousfish2014-12-24
| | | | As suggested in #1827
* Update completions for fish_indentGravatar ridiculousfish2014-12-24
|
* Clarify some documentation regarding fish_indentGravatar ridiculousfish2014-12-24
|
* Fix for assertion failure in fish_indent when setting a non-RGB colorGravatar ridiculousfish2014-12-24
|
* Tmuxinator completionsGravatar Jonathan Arnett2014-12-24
|
* Document new flags to fish_indentGravatar ridiculousfish2014-12-23
|
* Improve indentation of blocks inside if/while headersGravatar ridiculousfish2014-12-23
| | | | Fixes #1665
* Update fish_indent Xcode build targetGravatar ridiculousfish2014-12-23
|
* Correct path in fish_indent testsGravatar ridiculousfish2014-12-23
|
* Implement tests for fish_indentGravatar ridiculousfish2014-12-23
|
* Rewrite of fish_indentGravatar ridiculousfish2014-12-23
| | | | | Changes fish_indent to leverage new parse tree Also supports colorizing output via --html and --ansi flags.
* Allow running specific high-level tests by passing the name to test.fishGravatar ridiculousfish2014-12-23
| | | | | | Example: tests/test.fish expansion This will run the tests in expansion.in only
* Include fallback.h in output.h to avoid a compile errorGravatar ridiculousfish2014-12-23
|
* Clean up the last token when receiving TOK_ENDGravatar ridiculousfish2014-12-23
|
* Tweak comment handling. Mark a parse node with a new flag when it has commentsGravatar ridiculousfish2014-12-23
|
* Fix to avoid generating multiple error nodes on parse errorGravatar ridiculousfish2014-12-23
|
* Replace an expensive expand_one call with the cheaper unescape_stringGravatar ridiculousfish2014-12-23
| | | | Makes parsing a little faster
* Support nixos command-not-found handlerGravatar Max Gonzih2014-12-22
|
* osx/config.h: add some missed definesGravatar David Adam2014-12-21
|
* Merge a handful of old and new fixes for the build on SolarisGravatar David Adam2014-12-21
|\ | | | | | | | | Note that it does not successfully compile yet due to missing flock() on this platform.
| * Solaris build fixes: use _sys_errs if availableGravatar David Adam2014-12-21
| |
| * Solaris build fixes: detect <termios.h> in configureGravatar David Adam2014-12-21
| | | | | | | | The header is already used appropriately in the code.
| * Solaris build fixes: use memset instead of bzeroGravatar David Adam2014-12-21
| |
| * Solaris build fixes: fallback to _XOPEN_NAME_MAX if definedGravatar David Adam2014-12-21
| |
| * use configure to detect dirent.d_typeGravatar David Adam2014-12-21
| |