aboutsummaryrefslogtreecommitdiff
path: root/stagit.c
Commit message (Collapse)AuthorAgeFilesLines
* fix potential leak, found by clang-analyzerHiltjo Posthuma2017-03-121-0/+1
|
* fix memory leak of commitinfoHiltjo Posthuma2017-03-121-1/+1
|
* atom.xml: add newline before EOFHiltjo Posthuma2017-03-121-1/+1
|
* fix memory leak of commitinfo->parentHiltjo Posthuma2017-03-121-0/+1
|
* fix memory leak of blob objectHiltjo Posthuma2017-03-121-0/+1
|
* simplify/shorten pledge stubHiltjo Posthuma2017-03-121-5/+1
|
* reduce scope of repodir, remove unused branch parameterHiltjo Posthuma2017-03-121-6/+6
|
* also remove language and text direction attribute from <html>Hiltjo Posthuma2016-11-091-1/+1
|
* remove Content-Language: enHiltjo Posthuma2016-11-091-1/+1
| | | | thanks Aymeric Mansoux!
* treat an empty repo (without a HEAD) as validHiltjo Posthuma2016-05-291-5/+10
| | | | and create the standard files. Disable caching if there is no HEAD.
* refs_cmp: simplify check, functionally the sameHiltjo Posthuma2016-05-211-6/+3
|
* check if LICENSE, README and .gitmodules is a file blobHiltjo Posthuma2016-05-081-3/+8
| | | | don't link to directory trees, etc. Thanks 600360 for reporting it!
* simplify, id is unusedHiltjo Posthuma2016-05-071-3/+2
|
* simplify, no need for gotoHiltjo Posthuma2016-05-071-5/+3
|
* consistency for messages for special-case in diffstatHiltjo Posthuma2016-05-071-4/+4
|
* fix error in "code cleanup" commit 9328d2690e118127bcaa5cf9f665d8e7711f7a03Hiltjo Posthuma2016-05-071-1/+0
| | | | | ... the loop was simplified, but forgot to remove p++ (double used) while rebasing the change.
* fix and simplify time offsetHiltjo Posthuma2016-05-061-8/+7
|
* make diffstat files an anchor (jump to delta)Hiltjo Posthuma2016-05-061-4/+4
|
* fix joinpath(): use of global 'repodir', should be 'path'Hiltjo Posthuma2016-05-031-22/+17
| | | | use joinpath for another case.
* fix regression from last commitHiltjo Posthuma2016-05-031-1/+1
|
* code cleanupHiltjo Posthuma2016-05-031-43/+18
|
* remove config.h, add options to stagit.cHiltjo Posthuma2016-05-031-1/+5
|
* add OpenBSD pledge(2) support using -DUSE_PLEDGEHiltjo Posthuma2016-05-031-0/+11
|
* style: write atom feed end in one write, kills 2 linesHiltjo Posthuma2016-05-021-3/+1
|
* cache support (-c option)Hiltjo Posthuma2016-05-021-38/+113
| | | | | | | | "Optionally the -c cachefile option can be used to cache the entries of the log page up to the point of the last commit. The cachefile will store the last commit id and the entries in the HTML table." this caches the diffstat and commits, it is an expensive operation (twss).
* small code cleanup: no need for status variableHiltjo Posthuma2016-05-011-4/+3
|
* atom feed: use published for author commit date, updated for committer dateHiltjo Posthuma2016-05-011-1/+8
| | | | this is the same as cgit.
* fix times (and timezone)Hiltjo Posthuma2016-05-011-11/+27
| | | | | | - in the index and log show the short time (GMT). - in the Atom feed use GMT time. - for commits show the time + offset.
* fix path join '/'Hiltjo Posthuma2016-05-011-21/+18
| | | | add joinpath() function to handle this.
* code cleanup + style, functionally equivalentHiltjo Posthuma2016-05-011-19/+11
|
* layout: right align lines changed count in diffstatHiltjo Posthuma2016-04-301-1/+1
|
* optimization: suppress large diffsHiltjo Posthuma2016-04-301-0/+8
| | | | the values are arbitrary, these can be tweaked later if needed.
* optimization: read stats once and remember itHiltjo Posthuma2016-04-301-39/+147
| | | | | | | | | for an initial run and new commits this speeds stagit up a bit: on an initial run of sbase goes from about 4 seconds to 2.8 on my machine. now we can't use statsbuf, so create the stats string ourselves, while at it color the + and - using a style (can be disabled for the color-haters out there ;)).
* whoops, module is m---------Hiltjo Posthuma2016-04-291-1/+1
|
* submodule supportHiltjo Posthuma2016-04-291-15/+9
| | | | | | | - detect .gitmodules file for submodules, add a link "submodules" to the top menu. - link each submodule to the .gitmodules file in HEAD. - revert the special style for modules, but list it as a file "m---------", don't show the line/filesize though, it is not a file.
* suppress meaningless error codes when a repo can't be opened (similar to ↵Hiltjo Posthuma2016-04-291-1/+1
| | | | stagit-index)
* minor style fixHiltjo Posthuma2016-04-291-1/+1
|
* Fix null pointer access for submodules in writefilestreeQuentin Rameau2016-04-291-6/+11
| | | | | Do not try to set a link to submodules from bare repo as we can't get the actual url.
* tweak style of showing submodulesHiltjo Posthuma2016-04-291-4/+6
| | | | this will make it look similar to cgit. show mode as "m---------"
* Don't fail on submodulesz3bra2016-04-291-29/+40
| | | | | | | | | | Submodules will get listed prefixed with a '@', using a specific CSS class for styling. The href will be set to the url of the submodule. Filesize will appear as 0 and filemode will not be printed to avoid an awkward mode: "?---------". In writefilestree, we don't return anymore if an entry can't be categorized as an object, but rather, fail if we can't retrieve its name.
* fix: check gmtime, can return NULL on errorHiltjo Posthuma2016-04-271-1/+2
|
* resolve absolute paths to repodir, remove basename just use strrchr.Hiltjo Posthuma2016-04-271-24/+8
| | | | | | | - resolve repodir, for example: stagit-index ../ used to use ".." as the name, now it will resolve to the real directory name. - just use strrchr(path, '/') instead of basename, '/' path separator is now used.
* fix: link to line in hunk if a patch has multiple files and hunksHiltjo Posthuma2016-04-271-5/+5
| | | | thanks to lostd for reporting it!
* add titles to pages, summary for commit messages, filename for filesHiltjo Posthuma2016-03-011-6/+9
|
* fix right-alignment of "Files" columnHiltjo Posthuma2016-02-261-1/+1
|
* Use the recommended HTML5 document type.lostd2016-02-261-1/+1
|
* check path truncationHiltjo Posthuma2016-02-241-17/+38
| | | | be strict about it
* atom feed: add link, move content order to bottomHiltjo Posthuma2016-02-231-7/+11
| | | | | NOTE: using relative links violates the Atom specification, but relative links are convenient aswell.
* make printshowfile more generic and like other functionsHiltjo Posthuma2016-02-231-15/+14
|
* just use Name as first column, thanks quinq for the suggestionHiltjo Posthuma2016-01-211-3/+2
|
generated by cgit v1.2.3 (git 2.39.1) at 2025-06-28 18:42:04 +0200