summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Set version to 0.65.v0.65Jean Delvare2016-11-093-10/+41
|
* Update German translationHolger Wansing2016-11-081-3/+4
|
* Update Russian translationSergey Basalaev2016-11-081-3/+4
|
* Resync translation filesJean Delvare2016-11-072-20/+26
|
* Update the Japanese translation for Quilt 0.65Yasushi SHOJI2016-11-041-14/+19
|
* Update the French translationJean Delvare2016-11-032-21/+27
|
* mail: Describe the --charset option betterJean Delvare2016-11-031-1/+3
|
* Update German translationHolger Wansing2016-10-241-8/+11
|
* Update Russian translationSergey Basalaev2016-10-241-4/+6
|
* Update the French translationJean Delvare2016-10-211-1/+7
|
* Resync translation files + trivial updatesJean Delvare2016-10-215-72/+100
|
* bash_completion: Improve the handling of the push commandJean Delvare2016-10-211-1/+7
| | | | Some options need a parameter, invite the user to provide it.
* bash_completion: Add support for command "new"Jean Delvare2016-10-211-1/+11
| | | | Command "new" accepts options now, add support for them.
* bash_completion: Improve the handling of the mail commandJean Delvare2016-10-211-1/+14
| | | | Some options need a parameter, invite the user to provide it.
* bash_completion: Add new command optionsJean Delvare2016-10-211-15/+17
| | | | | Enable completion with the new options of commands diff, fold, import, mail, patches, pop, push, refresh, series and setup.
* bash_completion: Update the list of commandsJean Delvare2016-10-211-1/+1
| | | | Command "upgrade" was missing.
* push: Fix the synopsisJean Delvare2016-10-211-1/+1
| | | | Add missing options -m and --fuzz to the synopsis.
* mail: Fix the help textJean Delvare2016-10-211-1/+4
| | | | | Add options --charset and --signature to the synopsis, and describe the --charset option.
* bash_completion: Drop wrapping conditionalJean Delvare2016-10-211-4/+0
| | | | | | Drop the check for quilt being present. No other completion file has such a check, because it's useless. If the program isn't installed then its completion file isn't installed either.
* bash_completion: Drop duplicate functionsJean Delvare2016-10-211-66/+1
| | | | | | | | | | Completing long options of the grep command is not a fundamental feature. I can't remember ever needing it. The most useful grep options are short ones by design. Additionally, our internal implementation of _longopt is broken as far as I can tell. So drop our implementations of _expand, _filedir and _longopt, and only call _longopt if it is already available from bash-completion.
* bash_completion: Use _quilt_comfile for setupJean Delvare2016-10-211-3/+3
| | | | | Add support for directories to _quilt_comfile, and use that for the setup command. The drops the last direct dependency to _filedir.
* bash_completion: Use _quilt_comfile for diffJean Delvare2016-10-211-1/+1
| | | | | All other commands use _quilt_comfile so do the same for diff for consistency.
* bash_completion: Handle spaces in file namesJean Delvare2016-10-211-11/+12
| | | | | | | | Set IFS in _quilt_comfile so that file names containing spaces are properly handled. Also quote ${COMPREPLY[@]} everywhere so that such file names are preserved. This closes Debian bug #640551.
* bash_completion: Delete dead codeJean Delvare2016-10-211-6/+3
| | | | | | | | | | You can't change the caller's environment from the right hand side of a pipe. So that code was doing nothing, which means we can just delete it. What the code was supposedly doing, is actually implemented by complete's option "filenames". So move the comment to where this option is set, to clear the confusion.
* Update the French translationJean Delvare2016-10-211-1/+1
|
* Resync translation filesJean Delvare2016-10-215-135/+155
|
* Make one error message translatableJean Delvare2016-10-211-1/+1
|
* Fix a race condition in diff_fileJean Delvare2016-10-191-5/+7
| | | | | | Fix a race condition where we may check the return status of diff before the command has actually terminated. Make sure that we read all of diff's output to ensure this can't happen.
* diff: Report diff failuresJean Delvare2016-10-193-1/+2
| | | | | "quilt refresh" reports diff failures to the user. "quilt diff" should do the same.
* Reject binary files in patchesJean Delvare2016-09-151-1/+7
| | | | | | Since diffutils version 3.4, diff no longer returns an error code for binary files. Parse the first line of the output to detect this case and raise our own error.
* Project settings have priorityJean Delvare2015-12-042-2/+24
| | | | | | | | If QUILT_PATCHES or QUILT_SERIES are set both in ~/.quiltrc and .pc/.quilt_patches (respectively .pc/.quilt_series), the former currently takes precedence. This pretty much voids the value of per-project settings. Project settings should have the highest priority.
* setup: Drop variable $opt_fastJean Delvare2015-07-021-6/+5
| | | | | Variables $opt_fast and $QUILT_SETUP_FAST have the same meaning and value, so we can drop the former.
* setup: Reorder code in function inspect()Jean Delvare2015-07-021-12/+8
| | | | | Reorder the code in function inspect() to avoid testing for the same condition twice.
* setup: Delay the computation of md5sumsJean Delvare2015-07-022-3/+16
| | | | | | | | In fast mode, don't compute the md5sum of the files in the source directory before we need them. In most cases we will never actually need them, so this avoids computing them for nothing. Speed gain from this change can reach 12 %.
* inspect-wrapper: Identify files passed as stdinJean Delvare2015-07-021-0/+7
| | | | | | | | | | When the input file is passed through stdin, we can usually get its name from procfs. This avoids having to look up the file by md5sum. Not only this makes things faster, but this also helps if the patch file was modified by additional commands in the spec file before being applied (yes, I've seen this happen.) Speed gain from this change can reach 20 %.
* inspect-wrapper: Skip file copy in fast modeJean Delvare2015-06-161-3/+11
| | | | | | | | In fast mode, we only process calls to patch, and we don't let it actually apply the patches, so we never need to make a temporary copy of files read from the standard input. This change results in a 3 % performance gain in my test case.
* inspect-wrapper: Change original_file() parameterJean Delvare2015-06-161-4/+3
| | | | | Pass the md5sum instead of the file name to original_file(). This will allow for further optimization in fast mode.
* inspect-wrapper: Don't look up already known filesJean Delvare2015-06-162-9/+16
| | | | | | | If the input file is passed as a command line parameter, we already know its name, don't waste time looking it up. This change results in a 3 to 5 % performance gain in my test case.
* guards: Clarify user messageJean Delvare2015-04-291-1/+1
| | | | | | | | | | When --config=series is used and a listed file is not found on the disk, guards prints the following message: Not found in series: file This message is confusing, as the file was found in series, it is missing on the disk. Change the message to make this more obvious.
* Use standard sysconfdir as an etcdirNikolay Orlyuk2015-04-251-1/+1
|
* guards: Documentation updateJean Delvare2015-04-241-3/+5
| | | | | * Fix synopsis format. * Document the new --path=@<file> option.
* guards: Report which config file has problem in --check modeMichal Marek2015-04-241-2/+8
|
* guards: Include the file name also in the "Not found" errorMichal Marek2015-04-241-1/+5
|
* guards: Add support for an external filelist in --check modeMichal Marek2015-04-241-2/+14
| | | | This will allow us to run --check without a kernel-source.git work tree.
* guards: Simplify help textMichal Marek2015-04-241-2/+1
|
* inspect-wrapper: Compute our name only onceJean Delvare2015-04-151-10/+13
| | | | | | There are way too many occurrences of ${0##*/}. Store the result in a variable and use it as needed. Also hard-code the name when it is known, to save the cost of variable substitution altogether.
* inspect-wrapper: Report failed look-upsJean Delvare2015-03-242-19/+28
| | | | | The inspect-wrapper script looks up files by md5sum. If that fails, we should notify the user that something went wrong.
* Let cat_to_new_file handle the file backupJean Delvare2015-03-194-12/+52
| | | | | | | | | | | | | Let function cat_to_new_file optionally backup the destination file before overwriting it. This has two advantages: * This avoids duplicating code on the caller's side. Functions header and refresh both need to perform a backup. * This guarantees the consistency of the backup logic. At the moment, the refresh command would handle the backup of a symlinked file just fine while the header command would misbehave in that case. Now that the header command handles symlinked patches properly, also test it so that we don't break it accidentally later.
* hard-linked-patch.test: Test the header commandJean Delvare2015-03-191-0/+24
| | | | | Extend the hard-linked patch test case to also check the behavior when updating the header of such a patch.
* Test backup failureJean Delvare2015-03-191-0/+23
| | | | Add a test case for backup failure to the test suite.
generated by cgit (git 2.34.1) at 2024-09-19 17:14:19 +0000