gitweb: use filetest to allow ACLs
[git.git] / gitweb / gitweb.perl
blob6ac49eaf3e6477eb9eddaeec81a39db11f586551
1 #!/usr/bin/perl
3 # gitweb - simple web interface to track changes in git repositories
5 # (C) 2005-2006, Kay Sievers <kay.sievers@vrfy.org>
6 # (C) 2005, Christian Gierke
8 # This program is licensed under the GPLv2
10 use 5.008;
11 use strict;
12 use warnings;
13 # handle ACL in file access tests
14 use filetest 'access';
15 use CGI qw(:standard :escapeHTML -nosticky);
16 use CGI::Util qw(unescape);
17 use CGI::Carp qw(fatalsToBrowser set_message);
18 use Encode;
19 use Fcntl ':mode';
20 use File::Find qw();
21 use File::Basename qw(basename);
22 use Time::HiRes qw(gettimeofday tv_interval);
23 binmode STDOUT, ':utf8';
25 if (!defined($CGI::VERSION) || $CGI::VERSION < 4.08) {
26 eval 'sub CGI::multi_param { CGI::param(@_) }'
29 our $t0 = [ gettimeofday() ];
30 our $number_of_git_cmds = 0;
32 BEGIN {
33 CGI->compile() if $ENV{'MOD_PERL'};
36 our $version = "++GIT_VERSION++";
38 our ($my_url, $my_uri, $base_url, $path_info, $home_link);
39 sub evaluate_uri {
40 our $cgi;
42 our $my_url = $cgi->url();
43 our $my_uri = $cgi->url(-absolute => 1);
45 # Base URL for relative URLs in gitweb ($logo, $favicon, ...),
46 # needed and used only for URLs with nonempty PATH_INFO
47 our $base_url = $my_url;
49 # When the script is used as DirectoryIndex, the URL does not contain the name
50 # of the script file itself, and $cgi->url() fails to strip PATH_INFO, so we
51 # have to do it ourselves. We make $path_info global because it's also used
52 # later on.
54 # Another issue with the script being the DirectoryIndex is that the resulting
55 # $my_url data is not the full script URL: this is good, because we want
56 # generated links to keep implying the script name if it wasn't explicitly
57 # indicated in the URL we're handling, but it means that $my_url cannot be used
58 # as base URL.
59 # Therefore, if we needed to strip PATH_INFO, then we know that we have
60 # to build the base URL ourselves:
61 our $path_info = decode_utf8($ENV{"PATH_INFO"});
62 if ($path_info) {
63 # $path_info has already been URL-decoded by the web server, but
64 # $my_url and $my_uri have not. URL-decode them so we can properly
65 # strip $path_info.
66 $my_url = unescape($my_url);
67 $my_uri = unescape($my_uri);
68 if ($my_url =~ s,\Q$path_info\E$,, &&
69 $my_uri =~ s,\Q$path_info\E$,, &&
70 defined $ENV{'SCRIPT_NAME'}) {
71 $base_url = $cgi->url(-base => 1) . $ENV{'SCRIPT_NAME'};
75 # target of the home link on top of all pages
76 our $home_link = $my_uri || "/";
79 # core git executable to use
80 # this can just be "git" if your webserver has a sensible PATH
81 our $GIT = "++GIT_BINDIR++/git";
83 # absolute fs-path which will be prepended to the project path
84 #our $projectroot = "/pub/scm";
85 our $projectroot = "++GITWEB_PROJECTROOT++";
87 # fs traversing limit for getting project list
88 # the number is relative to the projectroot
89 our $project_maxdepth = "++GITWEB_PROJECT_MAXDEPTH++";
91 # string of the home link on top of all pages
92 our $home_link_str = "++GITWEB_HOME_LINK_STR++";
94 # extra breadcrumbs preceding the home link
95 our @extra_breadcrumbs = ();
97 # name of your site or organization to appear in page titles
98 # replace this with something more descriptive for clearer bookmarks
99 our $site_name = "++GITWEB_SITENAME++"
100 || ($ENV{'SERVER_NAME'} || "Untitled") . " Git";
102 # html snippet to include in the <head> section of each page
103 our $site_html_head_string = "++GITWEB_SITE_HTML_HEAD_STRING++";
104 # filename of html text to include at top of each page
105 our $site_header = "++GITWEB_SITE_HEADER++";
106 # html text to include at home page
107 our $home_text = "++GITWEB_HOMETEXT++";
108 # filename of html text to include at bottom of each page
109 our $site_footer = "++GITWEB_SITE_FOOTER++";
111 # URI of stylesheets
112 our @stylesheets = ("++GITWEB_CSS++");
113 # URI of a single stylesheet, which can be overridden in GITWEB_CONFIG.
114 our $stylesheet = undef;
115 # URI of GIT logo (72x27 size)
116 our $logo = "++GITWEB_LOGO++";
117 # URI of GIT favicon, assumed to be image/png type
118 our $favicon = "++GITWEB_FAVICON++";
119 # URI of gitweb.js (JavaScript code for gitweb)
120 our $javascript = "++GITWEB_JS++";
122 # URI and label (title) of GIT logo link
123 #our $logo_url = "http://www.kernel.org/pub/software/scm/git/docs/";
124 #our $logo_label = "git documentation";
125 our $logo_url = "http://git-scm.com/";
126 our $logo_label = "git homepage";
128 # source of projects list
129 our $projects_list = "++GITWEB_LIST++";
131 # the width (in characters) of the projects list "Description" column
132 our $projects_list_description_width = 25;
134 # group projects by category on the projects list
135 # (enabled if this variable evaluates to true)
136 our $projects_list_group_categories = 0;
138 # default category if none specified
139 # (leave the empty string for no category)
140 our $project_list_default_category = "";
142 # default order of projects list
143 # valid values are none, project, descr, owner, and age
144 our $default_projects_order = "project";
146 # show repository only if this file exists
147 # (only effective if this variable evaluates to true)
148 our $export_ok = "++GITWEB_EXPORT_OK++";
150 # don't generate age column on the projects list page
151 our $omit_age_column = 0;
153 # don't generate information about owners of repositories
154 our $omit_owner=0;
156 # show repository only if this subroutine returns true
157 # when given the path to the project, for example:
158 # sub { return -e "$_[0]/git-daemon-export-ok"; }
159 our $export_auth_hook = undef;
161 # only allow viewing of repositories also shown on the overview page
162 our $strict_export = "++GITWEB_STRICT_EXPORT++";
164 # list of git base URLs used for URL to where fetch project from,
165 # i.e. full URL is "$git_base_url/$project"
166 our @git_base_url_list = grep { $_ ne '' } ("++GITWEB_BASE_URL++");
168 # default blob_plain mimetype and default charset for text/plain blob
169 our $default_blob_plain_mimetype = 'text/plain';
170 our $default_text_plain_charset = undef;
172 # file to use for guessing MIME types before trying /etc/mime.types
173 # (relative to the current git repository)
174 our $mimetypes_file = undef;
176 # assume this charset if line contains non-UTF-8 characters;
177 # it should be valid encoding (see Encoding::Supported(3pm) for list),
178 # for which encoding all byte sequences are valid, for example
179 # 'iso-8859-1' aka 'latin1' (it is decoded without checking, so it
180 # could be even 'utf-8' for the old behavior)
181 our $fallback_encoding = 'latin1';
183 # rename detection options for git-diff and git-diff-tree
184 # - default is '-M', with the cost proportional to
185 # (number of removed files) * (number of new files).
186 # - more costly is '-C' (which implies '-M'), with the cost proportional to
187 # (number of changed files + number of removed files) * (number of new files)
188 # - even more costly is '-C', '--find-copies-harder' with cost
189 # (number of files in the original tree) * (number of new files)
190 # - one might want to include '-B' option, e.g. '-B', '-M'
191 our @diff_opts = ('-M'); # taken from git_commit
193 # Disables features that would allow repository owners to inject script into
194 # the gitweb domain.
195 our $prevent_xss = 0;
197 # Path to the highlight executable to use (must be the one from
198 # http://www.andre-simon.de due to assumptions about parameters and output).
199 # Useful if highlight is not installed on your webserver's PATH.
200 # [Default: highlight]
201 our $highlight_bin = "++HIGHLIGHT_BIN++";
203 # information about snapshot formats that gitweb is capable of serving
204 our %known_snapshot_formats = (
205 # name => {
206 # 'display' => display name,
207 # 'type' => mime type,
208 # 'suffix' => filename suffix,
209 # 'format' => --format for git-archive,
210 # 'compressor' => [compressor command and arguments]
211 # (array reference, optional)
212 # 'disabled' => boolean (optional)}
214 'tgz' => {
215 'display' => 'tar.gz',
216 'type' => 'application/x-gzip',
217 'suffix' => '.tar.gz',
218 'format' => 'tar',
219 'compressor' => ['gzip', '-n']},
221 'tbz2' => {
222 'display' => 'tar.bz2',
223 'type' => 'application/x-bzip2',
224 'suffix' => '.tar.bz2',
225 'format' => 'tar',
226 'compressor' => ['bzip2']},
228 'txz' => {
229 'display' => 'tar.xz',
230 'type' => 'application/x-xz',
231 'suffix' => '.tar.xz',
232 'format' => 'tar',
233 'compressor' => ['xz'],
234 'disabled' => 1},
236 'zip' => {
237 'display' => 'zip',
238 'type' => 'application/x-zip',
239 'suffix' => '.zip',
240 'format' => 'zip'},
243 # Aliases so we understand old gitweb.snapshot values in repository
244 # configuration.
245 our %known_snapshot_format_aliases = (
246 'gzip' => 'tgz',
247 'bzip2' => 'tbz2',
248 'xz' => 'txz',
250 # backward compatibility: legacy gitweb config support
251 'x-gzip' => undef, 'gz' => undef,
252 'x-bzip2' => undef, 'bz2' => undef,
253 'x-zip' => undef, '' => undef,
256 # Pixel sizes for icons and avatars. If the default font sizes or lineheights
257 # are changed, it may be appropriate to change these values too via
258 # $GITWEB_CONFIG.
259 our %avatar_size = (
260 'default' => 16,
261 'double' => 32
264 # Used to set the maximum load that we will still respond to gitweb queries.
265 # If server load exceed this value then return "503 server busy" error.
266 # If gitweb cannot determined server load, it is taken to be 0.
267 # Leave it undefined (or set to 'undef') to turn off load checking.
268 our $maxload = 300;
270 # configuration for 'highlight' (http://www.andre-simon.de/)
271 # match by basename
272 our %highlight_basename = (
273 #'Program' => 'py',
274 #'Library' => 'py',
275 'SConstruct' => 'py', # SCons equivalent of Makefile
276 'Makefile' => 'make',
278 # match by extension
279 our %highlight_ext = (
280 # main extensions, defining name of syntax;
281 # see files in /usr/share/highlight/langDefs/ directory
282 (map { $_ => $_ } qw(py rb java css js tex bib xml awk bat ini spec tcl sql)),
283 # alternate extensions, see /etc/highlight/filetypes.conf
284 (map { $_ => 'c' } qw(c h)),
285 (map { $_ => 'sh' } qw(sh bash zsh ksh)),
286 (map { $_ => 'cpp' } qw(cpp cxx c++ cc)),
287 (map { $_ => 'php' } qw(php php3 php4 php5 phps)),
288 (map { $_ => 'pl' } qw(pl perl pm)), # perhaps also 'cgi'
289 (map { $_ => 'make'} qw(make mak mk)),
290 (map { $_ => 'xml' } qw(xml xhtml html htm)),
293 # You define site-wide feature defaults here; override them with
294 # $GITWEB_CONFIG as necessary.
295 our %feature = (
296 # feature => {
297 # 'sub' => feature-sub (subroutine),
298 # 'override' => allow-override (boolean),
299 # 'default' => [ default options...] (array reference)}
301 # if feature is overridable (it means that allow-override has true value),
302 # then feature-sub will be called with default options as parameters;
303 # return value of feature-sub indicates if to enable specified feature
305 # if there is no 'sub' key (no feature-sub), then feature cannot be
306 # overridden
308 # use gitweb_get_feature(<feature>) to retrieve the <feature> value
309 # (an array) or gitweb_check_feature(<feature>) to check if <feature>
310 # is enabled
312 # Enable the 'blame' blob view, showing the last commit that modified
313 # each line in the file. This can be very CPU-intensive.
315 # To enable system wide have in $GITWEB_CONFIG
316 # $feature{'blame'}{'default'} = [1];
317 # To have project specific config enable override in $GITWEB_CONFIG
318 # $feature{'blame'}{'override'} = 1;
319 # and in project config gitweb.blame = 0|1;
320 'blame' => {
321 'sub' => sub { feature_bool('blame', @_) },
322 'override' => 0,
323 'default' => [0]},
325 # Enable the 'snapshot' link, providing a compressed archive of any
326 # tree. This can potentially generate high traffic if you have large
327 # project.
329 # Value is a list of formats defined in %known_snapshot_formats that
330 # you wish to offer.
331 # To disable system wide have in $GITWEB_CONFIG
332 # $feature{'snapshot'}{'default'} = [];
333 # To have project specific config enable override in $GITWEB_CONFIG
334 # $feature{'snapshot'}{'override'} = 1;
335 # and in project config, a comma-separated list of formats or "none"
336 # to disable. Example: gitweb.snapshot = tbz2,zip;
337 'snapshot' => {
338 'sub' => \&feature_snapshot,
339 'override' => 0,
340 'default' => ['tgz']},
342 # Enable text search, which will list the commits which match author,
343 # committer or commit text to a given string. Enabled by default.
344 # Project specific override is not supported.
346 # Note that this controls all search features, which means that if
347 # it is disabled, then 'grep' and 'pickaxe' search would also be
348 # disabled.
349 'search' => {
350 'override' => 0,
351 'default' => [1]},
353 # Enable grep search, which will list the files in currently selected
354 # tree containing the given string. Enabled by default. This can be
355 # potentially CPU-intensive, of course.
356 # Note that you need to have 'search' feature enabled too.
358 # To enable system wide have in $GITWEB_CONFIG
359 # $feature{'grep'}{'default'} = [1];
360 # To have project specific config enable override in $GITWEB_CONFIG
361 # $feature{'grep'}{'override'} = 1;
362 # and in project config gitweb.grep = 0|1;
363 'grep' => {
364 'sub' => sub { feature_bool('grep', @_) },
365 'override' => 0,
366 'default' => [1]},
368 # Enable the pickaxe search, which will list the commits that modified
369 # a given string in a file. This can be practical and quite faster
370 # alternative to 'blame', but still potentially CPU-intensive.
371 # Note that you need to have 'search' feature enabled too.
373 # To enable system wide have in $GITWEB_CONFIG
374 # $feature{'pickaxe'}{'default'} = [1];
375 # To have project specific config enable override in $GITWEB_CONFIG
376 # $feature{'pickaxe'}{'override'} = 1;
377 # and in project config gitweb.pickaxe = 0|1;
378 'pickaxe' => {
379 'sub' => sub { feature_bool('pickaxe', @_) },
380 'override' => 0,
381 'default' => [1]},
383 # Enable showing size of blobs in a 'tree' view, in a separate
384 # column, similar to what 'ls -l' does. This cost a bit of IO.
386 # To disable system wide have in $GITWEB_CONFIG
387 # $feature{'show-sizes'}{'default'} = [0];
388 # To have project specific config enable override in $GITWEB_CONFIG
389 # $feature{'show-sizes'}{'override'} = 1;
390 # and in project config gitweb.showsizes = 0|1;
391 'show-sizes' => {
392 'sub' => sub { feature_bool('showsizes', @_) },
393 'override' => 0,
394 'default' => [1]},
396 # Make gitweb use an alternative format of the URLs which can be
397 # more readable and natural-looking: project name is embedded
398 # directly in the path and the query string contains other
399 # auxiliary information. All gitweb installations recognize
400 # URL in either format; this configures in which formats gitweb
401 # generates links.
403 # To enable system wide have in $GITWEB_CONFIG
404 # $feature{'pathinfo'}{'default'} = [1];
405 # Project specific override is not supported.
407 # Note that you will need to change the default location of CSS,
408 # favicon, logo and possibly other files to an absolute URL. Also,
409 # if gitweb.cgi serves as your indexfile, you will need to force
410 # $my_uri to contain the script name in your $GITWEB_CONFIG.
411 'pathinfo' => {
412 'override' => 0,
413 'default' => [0]},
415 # Make gitweb consider projects in project root subdirectories
416 # to be forks of existing projects. Given project $projname.git,
417 # projects matching $projname/*.git will not be shown in the main
418 # projects list, instead a '+' mark will be added to $projname
419 # there and a 'forks' view will be enabled for the project, listing
420 # all the forks. If project list is taken from a file, forks have
421 # to be listed after the main project.
423 # To enable system wide have in $GITWEB_CONFIG
424 # $feature{'forks'}{'default'} = [1];
425 # Project specific override is not supported.
426 'forks' => {
427 'override' => 0,
428 'default' => [0]},
430 # Insert custom links to the action bar of all project pages.
431 # This enables you mainly to link to third-party scripts integrating
432 # into gitweb; e.g. git-browser for graphical history representation
433 # or custom web-based repository administration interface.
435 # The 'default' value consists of a list of triplets in the form
436 # (label, link, position) where position is the label after which
437 # to insert the link and link is a format string where %n expands
438 # to the project name, %f to the project path within the filesystem,
439 # %h to the current hash (h gitweb parameter) and %b to the current
440 # hash base (hb gitweb parameter); %% expands to %.
442 # To enable system wide have in $GITWEB_CONFIG e.g.
443 # $feature{'actions'}{'default'} = [('graphiclog',
444 # '/git-browser/by-commit.html?r=%n', 'summary')];
445 # Project specific override is not supported.
446 'actions' => {
447 'override' => 0,
448 'default' => []},
450 # Allow gitweb scan project content tags of project repository,
451 # and display the popular Web 2.0-ish "tag cloud" near the projects
452 # list. Note that this is something COMPLETELY different from the
453 # normal Git tags.
455 # gitweb by itself can show existing tags, but it does not handle
456 # tagging itself; you need to do it externally, outside gitweb.
457 # The format is described in git_get_project_ctags() subroutine.
458 # You may want to install the HTML::TagCloud Perl module to get
459 # a pretty tag cloud instead of just a list of tags.
461 # To enable system wide have in $GITWEB_CONFIG
462 # $feature{'ctags'}{'default'} = [1];
463 # Project specific override is not supported.
465 # In the future whether ctags editing is enabled might depend
466 # on the value, but using 1 should always mean no editing of ctags.
467 'ctags' => {
468 'override' => 0,
469 'default' => [0]},
471 # The maximum number of patches in a patchset generated in patch
472 # view. Set this to 0 or undef to disable patch view, or to a
473 # negative number to remove any limit.
475 # To disable system wide have in $GITWEB_CONFIG
476 # $feature{'patches'}{'default'} = [0];
477 # To have project specific config enable override in $GITWEB_CONFIG
478 # $feature{'patches'}{'override'} = 1;
479 # and in project config gitweb.patches = 0|n;
480 # where n is the maximum number of patches allowed in a patchset.
481 'patches' => {
482 'sub' => \&feature_patches,
483 'override' => 0,
484 'default' => [16]},
486 # Avatar support. When this feature is enabled, views such as
487 # shortlog or commit will display an avatar associated with
488 # the email of the committer(s) and/or author(s).
490 # Currently available providers are gravatar and picon.
491 # If an unknown provider is specified, the feature is disabled.
493 # Gravatar depends on Digest::MD5.
494 # Picon currently relies on the indiana.edu database.
496 # To enable system wide have in $GITWEB_CONFIG
497 # $feature{'avatar'}{'default'} = ['<provider>'];
498 # where <provider> is either gravatar or picon.
499 # To have project specific config enable override in $GITWEB_CONFIG
500 # $feature{'avatar'}{'override'} = 1;
501 # and in project config gitweb.avatar = <provider>;
502 'avatar' => {
503 'sub' => \&feature_avatar,
504 'override' => 0,
505 'default' => ['']},
507 # Enable displaying how much time and how many git commands
508 # it took to generate and display page. Disabled by default.
509 # Project specific override is not supported.
510 'timed' => {
511 'override' => 0,
512 'default' => [0]},
514 # Enable turning some links into links to actions which require
515 # JavaScript to run (like 'blame_incremental'). Not enabled by
516 # default. Project specific override is currently not supported.
517 'javascript-actions' => {
518 'override' => 0,
519 'default' => [0]},
521 # Enable and configure ability to change common timezone for dates
522 # in gitweb output via JavaScript. Enabled by default.
523 # Project specific override is not supported.
524 'javascript-timezone' => {
525 'override' => 0,
526 'default' => [
527 'local', # default timezone: 'utc', 'local', or '(-|+)HHMM' format,
528 # or undef to turn off this feature
529 'gitweb_tz', # name of cookie where to store selected timezone
530 'datetime', # CSS class used to mark up dates for manipulation
533 # Syntax highlighting support. This is based on Daniel Svensson's
534 # and Sham Chukoury's work in gitweb-xmms2.git.
535 # It requires the 'highlight' program present in $PATH,
536 # and therefore is disabled by default.
538 # To enable system wide have in $GITWEB_CONFIG
539 # $feature{'highlight'}{'default'} = [1];
541 'highlight' => {
542 'sub' => sub { feature_bool('highlight', @_) },
543 'override' => 0,
544 'default' => [0]},
546 # Enable displaying of remote heads in the heads list
548 # To enable system wide have in $GITWEB_CONFIG
549 # $feature{'remote_heads'}{'default'} = [1];
550 # To have project specific config enable override in $GITWEB_CONFIG
551 # $feature{'remote_heads'}{'override'} = 1;
552 # and in project config gitweb.remoteheads = 0|1;
553 'remote_heads' => {
554 'sub' => sub { feature_bool('remote_heads', @_) },
555 'override' => 0,
556 'default' => [0]},
558 # Enable showing branches under other refs in addition to heads
560 # To set system wide extra branch refs have in $GITWEB_CONFIG
561 # $feature{'extra-branch-refs'}{'default'} = ['dirs', 'of', 'choice'];
562 # To have project specific config enable override in $GITWEB_CONFIG
563 # $feature{'extra-branch-refs'}{'override'} = 1;
564 # and in project config gitweb.extrabranchrefs = dirs of choice
565 # Every directory is separated with whitespace.
567 'extra-branch-refs' => {
568 'sub' => \&feature_extra_branch_refs,
569 'override' => 0,
570 'default' => []},
573 sub gitweb_get_feature {
574 my ($name) = @_;
575 return unless exists $feature{$name};
576 my ($sub, $override, @defaults) = (
577 $feature{$name}{'sub'},
578 $feature{$name}{'override'},
579 @{$feature{$name}{'default'}});
580 # project specific override is possible only if we have project
581 our $git_dir; # global variable, declared later
582 if (!$override || !defined $git_dir) {
583 return @defaults;
585 if (!defined $sub) {
586 warn "feature $name is not overridable";
587 return @defaults;
589 return $sub->(@defaults);
592 # A wrapper to check if a given feature is enabled.
593 # With this, you can say
595 # my $bool_feat = gitweb_check_feature('bool_feat');
596 # gitweb_check_feature('bool_feat') or somecode;
598 # instead of
600 # my ($bool_feat) = gitweb_get_feature('bool_feat');
601 # (gitweb_get_feature('bool_feat'))[0] or somecode;
603 sub gitweb_check_feature {
604 return (gitweb_get_feature(@_))[0];
608 sub feature_bool {
609 my $key = shift;
610 my ($val) = git_get_project_config($key, '--bool');
612 if (!defined $val) {
613 return ($_[0]);
614 } elsif ($val eq 'true') {
615 return (1);
616 } elsif ($val eq 'false') {
617 return (0);
621 sub feature_snapshot {
622 my (@fmts) = @_;
624 my ($val) = git_get_project_config('snapshot');
626 if ($val) {
627 @fmts = ($val eq 'none' ? () : split /\s*[,\s]\s*/, $val);
630 return @fmts;
633 sub feature_patches {
634 my @val = (git_get_project_config('patches', '--int'));
636 if (@val) {
637 return @val;
640 return ($_[0]);
643 sub feature_avatar {
644 my @val = (git_get_project_config('avatar'));
646 return @val ? @val : @_;
649 sub feature_extra_branch_refs {
650 my (@branch_refs) = @_;
651 my $values = git_get_project_config('extrabranchrefs');
653 if ($values) {
654 $values = config_to_multi ($values);
655 @branch_refs = ();
656 foreach my $value (@{$values}) {
657 push @branch_refs, split /\s+/, $value;
661 return @branch_refs;
664 # checking HEAD file with -e is fragile if the repository was
665 # initialized long time ago (i.e. symlink HEAD) and was pack-ref'ed
666 # and then pruned.
667 sub check_head_link {
668 my ($dir) = @_;
669 my $headfile = "$dir/HEAD";
670 return ((-e $headfile) ||
671 (-l $headfile && readlink($headfile) =~ /^refs\/heads\//));
674 sub check_export_ok {
675 my ($dir) = @_;
676 return (check_head_link($dir) &&
677 (!$export_ok || -e "$dir/$export_ok") &&
678 (!$export_auth_hook || $export_auth_hook->($dir)));
681 # process alternate names for backward compatibility
682 # filter out unsupported (unknown) snapshot formats
683 sub filter_snapshot_fmts {
684 my @fmts = @_;
686 @fmts = map {
687 exists $known_snapshot_format_aliases{$_} ?
688 $known_snapshot_format_aliases{$_} : $_} @fmts;
689 @fmts = grep {
690 exists $known_snapshot_formats{$_} &&
691 !$known_snapshot_formats{$_}{'disabled'}} @fmts;
694 sub filter_and_validate_refs {
695 my @refs = @_;
696 my %unique_refs = ();
698 foreach my $ref (@refs) {
699 die_error(500, "Invalid ref '$ref' in 'extra-branch-refs' feature") unless (is_valid_ref_format($ref));
700 # 'heads' are added implicitly in get_branch_refs().
701 $unique_refs{$ref} = 1 if ($ref ne 'heads');
703 return sort keys %unique_refs;
706 # If it is set to code reference, it is code that it is to be run once per
707 # request, allowing updating configurations that change with each request,
708 # while running other code in config file only once.
710 # Otherwise, if it is false then gitweb would process config file only once;
711 # if it is true then gitweb config would be run for each request.
712 our $per_request_config = 1;
714 # read and parse gitweb config file given by its parameter.
715 # returns true on success, false on recoverable error, allowing
716 # to chain this subroutine, using first file that exists.
717 # dies on errors during parsing config file, as it is unrecoverable.
718 sub read_config_file {
719 my $filename = shift;
720 return unless defined $filename;
721 # die if there are errors parsing config file
722 if (-e $filename) {
723 do $filename;
724 die $@ if $@;
725 return 1;
727 return;
730 our ($GITWEB_CONFIG, $GITWEB_CONFIG_SYSTEM, $GITWEB_CONFIG_COMMON);
731 sub evaluate_gitweb_config {
732 our $GITWEB_CONFIG = $ENV{'GITWEB_CONFIG'} || "++GITWEB_CONFIG++";
733 our $GITWEB_CONFIG_SYSTEM = $ENV{'GITWEB_CONFIG_SYSTEM'} || "++GITWEB_CONFIG_SYSTEM++";
734 our $GITWEB_CONFIG_COMMON = $ENV{'GITWEB_CONFIG_COMMON'} || "++GITWEB_CONFIG_COMMON++";
736 # Protect against duplications of file names, to not read config twice.
737 # Only one of $GITWEB_CONFIG and $GITWEB_CONFIG_SYSTEM is used, so
738 # there possibility of duplication of filename there doesn't matter.
739 $GITWEB_CONFIG = "" if ($GITWEB_CONFIG eq $GITWEB_CONFIG_COMMON);
740 $GITWEB_CONFIG_SYSTEM = "" if ($GITWEB_CONFIG_SYSTEM eq $GITWEB_CONFIG_COMMON);
742 # Common system-wide settings for convenience.
743 # Those settings can be ovverriden by GITWEB_CONFIG or GITWEB_CONFIG_SYSTEM.
744 read_config_file($GITWEB_CONFIG_COMMON);
746 # Use first config file that exists. This means use the per-instance
747 # GITWEB_CONFIG if exists, otherwise use GITWEB_SYSTEM_CONFIG.
748 read_config_file($GITWEB_CONFIG) and return;
749 read_config_file($GITWEB_CONFIG_SYSTEM);
752 # Get loadavg of system, to compare against $maxload.
753 # Currently it requires '/proc/loadavg' present to get loadavg;
754 # if it is not present it returns 0, which means no load checking.
755 sub get_loadavg {
756 if( -e '/proc/loadavg' ){
757 open my $fd, '<', '/proc/loadavg'
758 or return 0;
759 my @load = split(/\s+/, scalar <$fd>);
760 close $fd;
762 # The first three columns measure CPU and IO utilization of the last one,
763 # five, and 10 minute periods. The fourth column shows the number of
764 # currently running processes and the total number of processes in the m/n
765 # format. The last column displays the last process ID used.
766 return $load[0] || 0;
768 # additional checks for load average should go here for things that don't export
769 # /proc/loadavg
771 return 0;
774 # version of the core git binary
775 our $git_version;
776 sub evaluate_git_version {
777 our $git_version = qx("$GIT" --version) =~ m/git version (.*)$/ ? $1 : "unknown";
778 $number_of_git_cmds++;
781 sub check_loadavg {
782 if (defined $maxload && get_loadavg() > $maxload) {
783 die_error(503, "The load average on the server is too high");
787 # ======================================================================
788 # input validation and dispatch
790 # input parameters can be collected from a variety of sources (presently, CGI
791 # and PATH_INFO), so we define an %input_params hash that collects them all
792 # together during validation: this allows subsequent uses (e.g. href()) to be
793 # agnostic of the parameter origin
795 our %input_params = ();
797 # input parameters are stored with the long parameter name as key. This will
798 # also be used in the href subroutine to convert parameters to their CGI
799 # equivalent, and since the href() usage is the most frequent one, we store
800 # the name -> CGI key mapping here, instead of the reverse.
802 # XXX: Warning: If you touch this, check the search form for updating,
803 # too.
805 our @cgi_param_mapping = (
806 project => "p",
807 action => "a",
808 file_name => "f",
809 file_parent => "fp",
810 hash => "h",
811 hash_parent => "hp",
812 hash_base => "hb",
813 hash_parent_base => "hpb",
814 page => "pg",
815 order => "o",
816 searchtext => "s",
817 searchtype => "st",
818 snapshot_format => "sf",
819 extra_options => "opt",
820 search_use_regexp => "sr",
821 ctag => "by_tag",
822 diff_style => "ds",
823 project_filter => "pf",
824 # this must be last entry (for manipulation from JavaScript)
825 javascript => "js"
827 our %cgi_param_mapping = @cgi_param_mapping;
829 # we will also need to know the possible actions, for validation
830 our %actions = (
831 "blame" => \&git_blame,
832 "blame_incremental" => \&git_blame_incremental,
833 "blame_data" => \&git_blame_data,
834 "blobdiff" => \&git_blobdiff,
835 "blobdiff_plain" => \&git_blobdiff_plain,
836 "blob" => \&git_blob,
837 "blob_plain" => \&git_blob_plain,
838 "commitdiff" => \&git_commitdiff,
839 "commitdiff_plain" => \&git_commitdiff_plain,
840 "commit" => \&git_commit,
841 "forks" => \&git_forks,
842 "heads" => \&git_heads,
843 "history" => \&git_history,
844 "log" => \&git_log,
845 "patch" => \&git_patch,
846 "patches" => \&git_patches,
847 "remotes" => \&git_remotes,
848 "rss" => \&git_rss,
849 "atom" => \&git_atom,
850 "search" => \&git_search,
851 "search_help" => \&git_search_help,
852 "shortlog" => \&git_shortlog,
853 "summary" => \&git_summary,
854 "tag" => \&git_tag,
855 "tags" => \&git_tags,
856 "tree" => \&git_tree,
857 "snapshot" => \&git_snapshot,
858 "object" => \&git_object,
859 # those below don't need $project
860 "opml" => \&git_opml,
861 "project_list" => \&git_project_list,
862 "project_index" => \&git_project_index,
865 # finally, we have the hash of allowed extra_options for the commands that
866 # allow them
867 our %allowed_options = (
868 "--no-merges" => [ qw(rss atom log shortlog history) ],
871 # fill %input_params with the CGI parameters. All values except for 'opt'
872 # should be single values, but opt can be an array. We should probably
873 # build an array of parameters that can be multi-valued, but since for the time
874 # being it's only this one, we just single it out
875 sub evaluate_query_params {
876 our $cgi;
878 while (my ($name, $symbol) = each %cgi_param_mapping) {
879 if ($symbol eq 'opt') {
880 $input_params{$name} = [ map { decode_utf8($_) } $cgi->multi_param($symbol) ];
881 } else {
882 $input_params{$name} = decode_utf8($cgi->param($symbol));
887 # now read PATH_INFO and update the parameter list for missing parameters
888 sub evaluate_path_info {
889 return if defined $input_params{'project'};
890 return if !$path_info;
891 $path_info =~ s,^/+,,;
892 return if !$path_info;
894 # find which part of PATH_INFO is project
895 my $project = $path_info;
896 $project =~ s,/+$,,;
897 while ($project && !check_head_link("$projectroot/$project")) {
898 $project =~ s,/*[^/]*$,,;
900 return unless $project;
901 $input_params{'project'} = $project;
903 # do not change any parameters if an action is given using the query string
904 return if $input_params{'action'};
905 $path_info =~ s,^\Q$project\E/*,,;
907 # next, check if we have an action
908 my $action = $path_info;
909 $action =~ s,/.*$,,;
910 if (exists $actions{$action}) {
911 $path_info =~ s,^$action/*,,;
912 $input_params{'action'} = $action;
915 # list of actions that want hash_base instead of hash, but can have no
916 # pathname (f) parameter
917 my @wants_base = (
918 'tree',
919 'history',
922 # we want to catch, among others
923 # [$hash_parent_base[:$file_parent]..]$hash_parent[:$file_name]
924 my ($parentrefname, $parentpathname, $refname, $pathname) =
925 ($path_info =~ /^(?:(.+?)(?::(.+))?\.\.)?([^:]+?)?(?::(.+))?$/);
927 # first, analyze the 'current' part
928 if (defined $pathname) {
929 # we got "branch:filename" or "branch:dir/"
930 # we could use git_get_type(branch:pathname), but:
931 # - it needs $git_dir
932 # - it does a git() call
933 # - the convention of terminating directories with a slash
934 # makes it superfluous
935 # - embedding the action in the PATH_INFO would make it even
936 # more superfluous
937 $pathname =~ s,^/+,,;
938 if (!$pathname || substr($pathname, -1) eq "/") {
939 $input_params{'action'} ||= "tree";
940 $pathname =~ s,/$,,;
941 } else {
942 # the default action depends on whether we had parent info
943 # or not
944 if ($parentrefname) {
945 $input_params{'action'} ||= "blobdiff_plain";
946 } else {
947 $input_params{'action'} ||= "blob_plain";
950 $input_params{'hash_base'} ||= $refname;
951 $input_params{'file_name'} ||= $pathname;
952 } elsif (defined $refname) {
953 # we got "branch". In this case we have to choose if we have to
954 # set hash or hash_base.
956 # Most of the actions without a pathname only want hash to be
957 # set, except for the ones specified in @wants_base that want
958 # hash_base instead. It should also be noted that hand-crafted
959 # links having 'history' as an action and no pathname or hash
960 # set will fail, but that happens regardless of PATH_INFO.
961 if (defined $parentrefname) {
962 # if there is parent let the default be 'shortlog' action
963 # (for http://git.example.com/repo.git/A..B links); if there
964 # is no parent, dispatch will detect type of object and set
965 # action appropriately if required (if action is not set)
966 $input_params{'action'} ||= "shortlog";
968 if ($input_params{'action'} &&
969 grep { $_ eq $input_params{'action'} } @wants_base) {
970 $input_params{'hash_base'} ||= $refname;
971 } else {
972 $input_params{'hash'} ||= $refname;
976 # next, handle the 'parent' part, if present
977 if (defined $parentrefname) {
978 # a missing pathspec defaults to the 'current' filename, allowing e.g.
979 # someproject/blobdiff/oldrev..newrev:/filename
980 if ($parentpathname) {
981 $parentpathname =~ s,^/+,,;
982 $parentpathname =~ s,/$,,;
983 $input_params{'file_parent'} ||= $parentpathname;
984 } else {
985 $input_params{'file_parent'} ||= $input_params{'file_name'};
987 # we assume that hash_parent_base is wanted if a path was specified,
988 # or if the action wants hash_base instead of hash
989 if (defined $input_params{'file_parent'} ||
990 grep { $_ eq $input_params{'action'} } @wants_base) {
991 $input_params{'hash_parent_base'} ||= $parentrefname;
992 } else {
993 $input_params{'hash_parent'} ||= $parentrefname;
997 # for the snapshot action, we allow URLs in the form
998 # $project/snapshot/$hash.ext
999 # where .ext determines the snapshot and gets removed from the
1000 # passed $refname to provide the $hash.
1002 # To be able to tell that $refname includes the format extension, we
1003 # require the following two conditions to be satisfied:
1004 # - the hash input parameter MUST have been set from the $refname part
1005 # of the URL (i.e. they must be equal)
1006 # - the snapshot format MUST NOT have been defined already (e.g. from
1007 # CGI parameter sf)
1008 # It's also useless to try any matching unless $refname has a dot,
1009 # so we check for that too
1010 if (defined $input_params{'action'} &&
1011 $input_params{'action'} eq 'snapshot' &&
1012 defined $refname && index($refname, '.') != -1 &&
1013 $refname eq $input_params{'hash'} &&
1014 !defined $input_params{'snapshot_format'}) {
1015 # We loop over the known snapshot formats, checking for
1016 # extensions. Allowed extensions are both the defined suffix
1017 # (which includes the initial dot already) and the snapshot
1018 # format key itself, with a prepended dot
1019 while (my ($fmt, $opt) = each %known_snapshot_formats) {
1020 my $hash = $refname;
1021 unless ($hash =~ s/(\Q$opt->{'suffix'}\E|\Q.$fmt\E)$//) {
1022 next;
1024 my $sfx = $1;
1025 # a valid suffix was found, so set the snapshot format
1026 # and reset the hash parameter
1027 $input_params{'snapshot_format'} = $fmt;
1028 $input_params{'hash'} = $hash;
1029 # we also set the format suffix to the one requested
1030 # in the URL: this way a request for e.g. .tgz returns
1031 # a .tgz instead of a .tar.gz
1032 $known_snapshot_formats{$fmt}{'suffix'} = $sfx;
1033 last;
1038 our ($action, $project, $file_name, $file_parent, $hash, $hash_parent, $hash_base,
1039 $hash_parent_base, @extra_options, $page, $searchtype, $search_use_regexp,
1040 $searchtext, $search_regexp, $project_filter);
1041 sub evaluate_and_validate_params {
1042 our $action = $input_params{'action'};
1043 if (defined $action) {
1044 if (!is_valid_action($action)) {
1045 die_error(400, "Invalid action parameter");
1049 # parameters which are pathnames
1050 our $project = $input_params{'project'};
1051 if (defined $project) {
1052 if (!is_valid_project($project)) {
1053 undef $project;
1054 die_error(404, "No such project");
1058 our $project_filter = $input_params{'project_filter'};
1059 if (defined $project_filter) {
1060 if (!is_valid_pathname($project_filter)) {
1061 die_error(404, "Invalid project_filter parameter");
1065 our $file_name = $input_params{'file_name'};
1066 if (defined $file_name) {
1067 if (!is_valid_pathname($file_name)) {
1068 die_error(400, "Invalid file parameter");
1072 our $file_parent = $input_params{'file_parent'};
1073 if (defined $file_parent) {
1074 if (!is_valid_pathname($file_parent)) {
1075 die_error(400, "Invalid file parent parameter");
1079 # parameters which are refnames
1080 our $hash = $input_params{'hash'};
1081 if (defined $hash) {
1082 if (!is_valid_refname($hash)) {
1083 die_error(400, "Invalid hash parameter");
1087 our $hash_parent = $input_params{'hash_parent'};
1088 if (defined $hash_parent) {
1089 if (!is_valid_refname($hash_parent)) {
1090 die_error(400, "Invalid hash parent parameter");
1094 our $hash_base = $input_params{'hash_base'};
1095 if (defined $hash_base) {
1096 if (!is_valid_refname($hash_base)) {
1097 die_error(400, "Invalid hash base parameter");
1101 our @extra_options = @{$input_params{'extra_options'}};
1102 # @extra_options is always defined, since it can only be (currently) set from
1103 # CGI, and $cgi->param() returns the empty array in array context if the param
1104 # is not set
1105 foreach my $opt (@extra_options) {
1106 if (not exists $allowed_options{$opt}) {
1107 die_error(400, "Invalid option parameter");
1109 if (not grep(/^$action$/, @{$allowed_options{$opt}})) {
1110 die_error(400, "Invalid option parameter for this action");
1114 our $hash_parent_base = $input_params{'hash_parent_base'};
1115 if (defined $hash_parent_base) {
1116 if (!is_valid_refname($hash_parent_base)) {
1117 die_error(400, "Invalid hash parent base parameter");
1121 # other parameters
1122 our $page = $input_params{'page'};
1123 if (defined $page) {
1124 if ($page =~ m/[^0-9]/) {
1125 die_error(400, "Invalid page parameter");
1129 our $searchtype = $input_params{'searchtype'};
1130 if (defined $searchtype) {
1131 if ($searchtype =~ m/[^a-z]/) {
1132 die_error(400, "Invalid searchtype parameter");
1136 our $search_use_regexp = $input_params{'search_use_regexp'};
1138 our $searchtext = $input_params{'searchtext'};
1139 our $search_regexp = undef;
1140 if (defined $searchtext) {
1141 if (length($searchtext) < 2) {
1142 die_error(403, "At least two characters are required for search parameter");
1144 if ($search_use_regexp) {
1145 $search_regexp = $searchtext;
1146 if (!eval { qr/$search_regexp/; 1; }) {
1147 (my $error = $@) =~ s/ at \S+ line \d+.*\n?//;
1148 die_error(400, "Invalid search regexp '$search_regexp'",
1149 esc_html($error));
1151 } else {
1152 $search_regexp = quotemeta $searchtext;
1157 # path to the current git repository
1158 our $git_dir;
1159 sub evaluate_git_dir {
1160 our $git_dir = "$projectroot/$project" if $project;
1163 our (@snapshot_fmts, $git_avatar, @extra_branch_refs);
1164 sub configure_gitweb_features {
1165 # list of supported snapshot formats
1166 our @snapshot_fmts = gitweb_get_feature('snapshot');
1167 @snapshot_fmts = filter_snapshot_fmts(@snapshot_fmts);
1169 # check that the avatar feature is set to a known provider name,
1170 # and for each provider check if the dependencies are satisfied.
1171 # if the provider name is invalid or the dependencies are not met,
1172 # reset $git_avatar to the empty string.
1173 our ($git_avatar) = gitweb_get_feature('avatar');
1174 if ($git_avatar eq 'gravatar') {
1175 $git_avatar = '' unless (eval { require Digest::MD5; 1; });
1176 } elsif ($git_avatar eq 'picon') {
1177 # no dependencies
1178 } else {
1179 $git_avatar = '';
1182 our @extra_branch_refs = gitweb_get_feature('extra-branch-refs');
1183 @extra_branch_refs = filter_and_validate_refs (@extra_branch_refs);
1186 sub get_branch_refs {
1187 return ('heads', @extra_branch_refs);
1190 # custom error handler: 'die <message>' is Internal Server Error
1191 sub handle_errors_html {
1192 my $msg = shift; # it is already HTML escaped
1194 # to avoid infinite loop where error occurs in die_error,
1195 # change handler to default handler, disabling handle_errors_html
1196 set_message("Error occurred when inside die_error:\n$msg");
1198 # you cannot jump out of die_error when called as error handler;
1199 # the subroutine set via CGI::Carp::set_message is called _after_
1200 # HTTP headers are already written, so it cannot write them itself
1201 die_error(undef, undef, $msg, -error_handler => 1, -no_http_header => 1);
1203 set_message(\&handle_errors_html);
1205 # dispatch
1206 sub dispatch {
1207 if (!defined $action) {
1208 if (defined $hash) {
1209 $action = git_get_type($hash);
1210 $action or die_error(404, "Object does not exist");
1211 } elsif (defined $hash_base && defined $file_name) {
1212 $action = git_get_type("$hash_base:$file_name");
1213 $action or die_error(404, "File or directory does not exist");
1214 } elsif (defined $project) {
1215 $action = 'summary';
1216 } else {
1217 $action = 'project_list';
1220 if (!defined($actions{$action})) {
1221 die_error(400, "Unknown action");
1223 if ($action !~ m/^(?:opml|project_list|project_index)$/ &&
1224 !$project) {
1225 die_error(400, "Project needed");
1227 $actions{$action}->();
1230 sub reset_timer {
1231 our $t0 = [ gettimeofday() ]
1232 if defined $t0;
1233 our $number_of_git_cmds = 0;
1236 our $first_request = 1;
1237 sub run_request {
1238 reset_timer();
1240 evaluate_uri();
1241 if ($first_request) {
1242 evaluate_gitweb_config();
1243 evaluate_git_version();
1245 if ($per_request_config) {
1246 if (ref($per_request_config) eq 'CODE') {
1247 $per_request_config->();
1248 } elsif (!$first_request) {
1249 evaluate_gitweb_config();
1252 check_loadavg();
1254 # $projectroot and $projects_list might be set in gitweb config file
1255 $projects_list ||= $projectroot;
1257 evaluate_query_params();
1258 evaluate_path_info();
1259 evaluate_and_validate_params();
1260 evaluate_git_dir();
1262 configure_gitweb_features();
1264 dispatch();
1267 our $is_last_request = sub { 1 };
1268 our ($pre_dispatch_hook, $post_dispatch_hook, $pre_listen_hook);
1269 our $CGI = 'CGI';
1270 our $cgi;
1271 sub configure_as_fcgi {
1272 require CGI::Fast;
1273 our $CGI = 'CGI::Fast';
1275 my $request_number = 0;
1276 # let each child service 100 requests
1277 our $is_last_request = sub { ++$request_number > 100 };
1279 sub evaluate_argv {
1280 my $script_name = $ENV{'SCRIPT_NAME'} || $ENV{'SCRIPT_FILENAME'} || __FILE__;
1281 configure_as_fcgi()
1282 if $script_name =~ /\.fcgi$/;
1284 return unless (@ARGV);
1286 require Getopt::Long;
1287 Getopt::Long::GetOptions(
1288 'fastcgi|fcgi|f' => \&configure_as_fcgi,
1289 'nproc|n=i' => sub {
1290 my ($arg, $val) = @_;
1291 return unless eval { require FCGI::ProcManager; 1; };
1292 my $proc_manager = FCGI::ProcManager->new({
1293 n_processes => $val,
1295 our $pre_listen_hook = sub { $proc_manager->pm_manage() };
1296 our $pre_dispatch_hook = sub { $proc_manager->pm_pre_dispatch() };
1297 our $post_dispatch_hook = sub { $proc_manager->pm_post_dispatch() };
1302 sub run {
1303 evaluate_argv();
1305 $first_request = 1;
1306 $pre_listen_hook->()
1307 if $pre_listen_hook;
1309 REQUEST:
1310 while ($cgi = $CGI->new()) {
1311 $pre_dispatch_hook->()
1312 if $pre_dispatch_hook;
1314 run_request();
1316 $post_dispatch_hook->()
1317 if $post_dispatch_hook;
1318 $first_request = 0;
1320 last REQUEST if ($is_last_request->());
1323 DONE_GITWEB:
1327 run();
1329 if (defined caller) {
1330 # wrapped in a subroutine processing requests,
1331 # e.g. mod_perl with ModPerl::Registry, or PSGI with Plack::App::WrapCGI
1332 return;
1333 } else {
1334 # pure CGI script, serving single request
1335 exit;
1338 ## ======================================================================
1339 ## action links
1341 # possible values of extra options
1342 # -full => 0|1 - use absolute/full URL ($my_uri/$my_url as base)
1343 # -replay => 1 - start from a current view (replay with modifications)
1344 # -path_info => 0|1 - don't use/use path_info URL (if possible)
1345 # -anchor => ANCHOR - add #ANCHOR to end of URL, implies -replay if used alone
1346 sub href {
1347 my %params = @_;
1348 # default is to use -absolute url() i.e. $my_uri
1349 my $href = $params{-full} ? $my_url : $my_uri;
1351 # implicit -replay, must be first of implicit params
1352 $params{-replay} = 1 if (keys %params == 1 && $params{-anchor});
1354 $params{'project'} = $project unless exists $params{'project'};
1356 if ($params{-replay}) {
1357 while (my ($name, $symbol) = each %cgi_param_mapping) {
1358 if (!exists $params{$name}) {
1359 $params{$name} = $input_params{$name};
1364 my $use_pathinfo = gitweb_check_feature('pathinfo');
1365 if (defined $params{'project'} &&
1366 (exists $params{-path_info} ? $params{-path_info} : $use_pathinfo)) {
1367 # try to put as many parameters as possible in PATH_INFO:
1368 # - project name
1369 # - action
1370 # - hash_parent or hash_parent_base:/file_parent
1371 # - hash or hash_base:/filename
1372 # - the snapshot_format as an appropriate suffix
1374 # When the script is the root DirectoryIndex for the domain,
1375 # $href here would be something like http://gitweb.example.com/
1376 # Thus, we strip any trailing / from $href, to spare us double
1377 # slashes in the final URL
1378 $href =~ s,/$,,;
1380 # Then add the project name, if present
1381 $href .= "/".esc_path_info($params{'project'});
1382 delete $params{'project'};
1384 # since we destructively absorb parameters, we keep this
1385 # boolean that remembers if we're handling a snapshot
1386 my $is_snapshot = $params{'action'} eq 'snapshot';
1388 # Summary just uses the project path URL, any other action is
1389 # added to the URL
1390 if (defined $params{'action'}) {
1391 $href .= "/".esc_path_info($params{'action'})
1392 unless $params{'action'} eq 'summary';
1393 delete $params{'action'};
1396 # Next, we put hash_parent_base:/file_parent..hash_base:/file_name,
1397 # stripping nonexistent or useless pieces
1398 $href .= "/" if ($params{'hash_base'} || $params{'hash_parent_base'}
1399 || $params{'hash_parent'} || $params{'hash'});
1400 if (defined $params{'hash_base'}) {
1401 if (defined $params{'hash_parent_base'}) {
1402 $href .= esc_path_info($params{'hash_parent_base'});
1403 # skip the file_parent if it's the same as the file_name
1404 if (defined $params{'file_parent'}) {
1405 if (defined $params{'file_name'} && $params{'file_parent'} eq $params{'file_name'}) {
1406 delete $params{'file_parent'};
1407 } elsif ($params{'file_parent'} !~ /\.\./) {
1408 $href .= ":/".esc_path_info($params{'file_parent'});
1409 delete $params{'file_parent'};
1412 $href .= "..";
1413 delete $params{'hash_parent'};
1414 delete $params{'hash_parent_base'};
1415 } elsif (defined $params{'hash_parent'}) {
1416 $href .= esc_path_info($params{'hash_parent'}). "..";
1417 delete $params{'hash_parent'};
1420 $href .= esc_path_info($params{'hash_base'});
1421 if (defined $params{'file_name'} && $params{'file_name'} !~ /\.\./) {
1422 $href .= ":/".esc_path_info($params{'file_name'});
1423 delete $params{'file_name'};
1425 delete $params{'hash'};
1426 delete $params{'hash_base'};
1427 } elsif (defined $params{'hash'}) {
1428 $href .= esc_path_info($params{'hash'});
1429 delete $params{'hash'};
1432 # If the action was a snapshot, we can absorb the
1433 # snapshot_format parameter too
1434 if ($is_snapshot) {
1435 my $fmt = $params{'snapshot_format'};
1436 # snapshot_format should always be defined when href()
1437 # is called, but just in case some code forgets, we
1438 # fall back to the default
1439 $fmt ||= $snapshot_fmts[0];
1440 $href .= $known_snapshot_formats{$fmt}{'suffix'};
1441 delete $params{'snapshot_format'};
1445 # now encode the parameters explicitly
1446 my @result = ();
1447 for (my $i = 0; $i < @cgi_param_mapping; $i += 2) {
1448 my ($name, $symbol) = ($cgi_param_mapping[$i], $cgi_param_mapping[$i+1]);
1449 if (defined $params{$name}) {
1450 if (ref($params{$name}) eq "ARRAY") {
1451 foreach my $par (@{$params{$name}}) {
1452 push @result, $symbol . "=" . esc_param($par);
1454 } else {
1455 push @result, $symbol . "=" . esc_param($params{$name});
1459 $href .= "?" . join(';', @result) if scalar @result;
1461 # final transformation: trailing spaces must be escaped (URI-encoded)
1462 $href =~ s/(\s+)$/CGI::escape($1)/e;
1464 if ($params{-anchor}) {
1465 $href .= "#".esc_param($params{-anchor});
1468 return $href;
1472 ## ======================================================================
1473 ## validation, quoting/unquoting and escaping
1475 sub is_valid_action {
1476 my $input = shift;
1477 return undef unless exists $actions{$input};
1478 return 1;
1481 sub is_valid_project {
1482 my $input = shift;
1484 return unless defined $input;
1485 if (!is_valid_pathname($input) ||
1486 !(-d "$projectroot/$input") ||
1487 !check_export_ok("$projectroot/$input") ||
1488 ($strict_export && !project_in_list($input))) {
1489 return undef;
1490 } else {
1491 return 1;
1495 sub is_valid_pathname {
1496 my $input = shift;
1498 return undef unless defined $input;
1499 # no '.' or '..' as elements of path, i.e. no '.' or '..'
1500 # at the beginning, at the end, and between slashes.
1501 # also this catches doubled slashes
1502 if ($input =~ m!(^|/)(|\.|\.\.)(/|$)!) {
1503 return undef;
1505 # no null characters
1506 if ($input =~ m!\0!) {
1507 return undef;
1509 return 1;
1512 sub is_valid_ref_format {
1513 my $input = shift;
1515 return undef unless defined $input;
1516 # restrictions on ref name according to git-check-ref-format
1517 if ($input =~ m!(/\.|\.\.|[\000-\040\177 ~^:?*\[]|/$)!) {
1518 return undef;
1520 return 1;
1523 sub is_valid_refname {
1524 my $input = shift;
1526 return undef unless defined $input;
1527 # textual hashes are O.K.
1528 if ($input =~ m/^[0-9a-fA-F]{40}$/) {
1529 return 1;
1531 # it must be correct pathname
1532 is_valid_pathname($input) or return undef;
1533 # check git-check-ref-format restrictions
1534 is_valid_ref_format($input) or return undef;
1535 return 1;
1538 # decode sequences of octets in utf8 into Perl's internal form,
1539 # which is utf-8 with utf8 flag set if needed. gitweb writes out
1540 # in utf-8 thanks to "binmode STDOUT, ':utf8'" at beginning
1541 sub to_utf8 {
1542 my $str = shift;
1543 return undef unless defined $str;
1545 if (utf8::is_utf8($str) || utf8::decode($str)) {
1546 return $str;
1547 } else {
1548 return decode($fallback_encoding, $str, Encode::FB_DEFAULT);
1552 # quote unsafe chars, but keep the slash, even when it's not
1553 # correct, but quoted slashes look too horrible in bookmarks
1554 sub esc_param {
1555 my $str = shift;
1556 return undef unless defined $str;
1557 $str =~ s/([^A-Za-z0-9\-_.~()\/:@ ]+)/CGI::escape($1)/eg;
1558 $str =~ s/ /\+/g;
1559 return $str;
1562 # the quoting rules for path_info fragment are slightly different
1563 sub esc_path_info {
1564 my $str = shift;
1565 return undef unless defined $str;
1567 # path_info doesn't treat '+' as space (specially), but '?' must be escaped
1568 $str =~ s/([^A-Za-z0-9\-_.~();\/;:@&= +]+)/CGI::escape($1)/eg;
1570 return $str;
1573 # quote unsafe chars in whole URL, so some characters cannot be quoted
1574 sub esc_url {
1575 my $str = shift;
1576 return undef unless defined $str;
1577 $str =~ s/([^A-Za-z0-9\-_.~();\/;?:@&= ]+)/CGI::escape($1)/eg;
1578 $str =~ s/ /\+/g;
1579 return $str;
1582 # quote unsafe characters in HTML attributes
1583 sub esc_attr {
1585 # for XHTML conformance escaping '"' to '&quot;' is not enough
1586 return esc_html(@_);
1589 # replace invalid utf8 character with SUBSTITUTION sequence
1590 sub esc_html {
1591 my $str = shift;
1592 my %opts = @_;
1594 return undef unless defined $str;
1596 $str = to_utf8($str);
1597 $str = $cgi->escapeHTML($str);
1598 if ($opts{'-nbsp'}) {
1599 $str =~ s/ /&nbsp;/g;
1601 $str =~ s|([[:cntrl:]])|(($1 ne "\t") ? quot_cec($1) : $1)|eg;
1602 return $str;
1605 # quote control characters and escape filename to HTML
1606 sub esc_path {
1607 my $str = shift;
1608 my %opts = @_;
1610 return undef unless defined $str;
1612 $str = to_utf8($str);
1613 $str = $cgi->escapeHTML($str);
1614 if ($opts{'-nbsp'}) {
1615 $str =~ s/ /&nbsp;/g;
1617 $str =~ s|([[:cntrl:]])|quot_cec($1)|eg;
1618 return $str;
1621 # Sanitize for use in XHTML + application/xml+xhtml (valid XML 1.0)
1622 sub sanitize {
1623 my $str = shift;
1625 return undef unless defined $str;
1627 $str = to_utf8($str);
1628 $str =~ s|([[:cntrl:]])|(index("\t\n\r", $1) != -1 ? $1 : quot_cec($1))|eg;
1629 return $str;
1632 # Make control characters "printable", using character escape codes (CEC)
1633 sub quot_cec {
1634 my $cntrl = shift;
1635 my %opts = @_;
1636 my %es = ( # character escape codes, aka escape sequences
1637 "\t" => '\t', # tab (HT)
1638 "\n" => '\n', # line feed (LF)
1639 "\r" => '\r', # carrige return (CR)
1640 "\f" => '\f', # form feed (FF)
1641 "\b" => '\b', # backspace (BS)
1642 "\a" => '\a', # alarm (bell) (BEL)
1643 "\e" => '\e', # escape (ESC)
1644 "\013" => '\v', # vertical tab (VT)
1645 "\000" => '\0', # nul character (NUL)
1647 my $chr = ( (exists $es{$cntrl})
1648 ? $es{$cntrl}
1649 : sprintf('\%2x', ord($cntrl)) );
1650 if ($opts{-nohtml}) {
1651 return $chr;
1652 } else {
1653 return "<span class=\"cntrl\">$chr</span>";
1657 # Alternatively use unicode control pictures codepoints,
1658 # Unicode "printable representation" (PR)
1659 sub quot_upr {
1660 my $cntrl = shift;
1661 my %opts = @_;
1663 my $chr = sprintf('&#%04d;', 0x2400+ord($cntrl));
1664 if ($opts{-nohtml}) {
1665 return $chr;
1666 } else {
1667 return "<span class=\"cntrl\">$chr</span>";
1671 # git may return quoted and escaped filenames
1672 sub unquote {
1673 my $str = shift;
1675 sub unq {
1676 my $seq = shift;
1677 my %es = ( # character escape codes, aka escape sequences
1678 't' => "\t", # tab (HT, TAB)
1679 'n' => "\n", # newline (NL)
1680 'r' => "\r", # return (CR)
1681 'f' => "\f", # form feed (FF)
1682 'b' => "\b", # backspace (BS)
1683 'a' => "\a", # alarm (bell) (BEL)
1684 'e' => "\e", # escape (ESC)
1685 'v' => "\013", # vertical tab (VT)
1688 if ($seq =~ m/^[0-7]{1,3}$/) {
1689 # octal char sequence
1690 return chr(oct($seq));
1691 } elsif (exists $es{$seq}) {
1692 # C escape sequence, aka character escape code
1693 return $es{$seq};
1695 # quoted ordinary character
1696 return $seq;
1699 if ($str =~ m/^"(.*)"$/) {
1700 # needs unquoting
1701 $str = $1;
1702 $str =~ s/\\([^0-7]|[0-7]{1,3})/unq($1)/eg;
1704 return $str;
1707 # escape tabs (convert tabs to spaces)
1708 sub untabify {
1709 my $line = shift;
1711 while ((my $pos = index($line, "\t")) != -1) {
1712 if (my $count = (8 - ($pos % 8))) {
1713 my $spaces = ' ' x $count;
1714 $line =~ s/\t/$spaces/;
1718 return $line;
1721 sub project_in_list {
1722 my $project = shift;
1723 my @list = git_get_projects_list();
1724 return @list && scalar(grep { $_->{'path'} eq $project } @list);
1727 ## ----------------------------------------------------------------------
1728 ## HTML aware string manipulation
1730 # Try to chop given string on a word boundary between position
1731 # $len and $len+$add_len. If there is no word boundary there,
1732 # chop at $len+$add_len. Do not chop if chopped part plus ellipsis
1733 # (marking chopped part) would be longer than given string.
1734 sub chop_str {
1735 my $str = shift;
1736 my $len = shift;
1737 my $add_len = shift || 10;
1738 my $where = shift || 'right'; # 'left' | 'center' | 'right'
1740 # Make sure perl knows it is utf8 encoded so we don't
1741 # cut in the middle of a utf8 multibyte char.
1742 $str = to_utf8($str);
1744 # allow only $len chars, but don't cut a word if it would fit in $add_len
1745 # if it doesn't fit, cut it if it's still longer than the dots we would add
1746 # remove chopped character entities entirely
1748 # when chopping in the middle, distribute $len into left and right part
1749 # return early if chopping wouldn't make string shorter
1750 if ($where eq 'center') {
1751 return $str if ($len + 5 >= length($str)); # filler is length 5
1752 $len = int($len/2);
1753 } else {
1754 return $str if ($len + 4 >= length($str)); # filler is length 4
1757 # regexps: ending and beginning with word part up to $add_len
1758 my $endre = qr/.{$len}\w{0,$add_len}/;
1759 my $begre = qr/\w{0,$add_len}.{$len}/;
1761 if ($where eq 'left') {
1762 $str =~ m/^(.*?)($begre)$/;
1763 my ($lead, $body) = ($1, $2);
1764 if (length($lead) > 4) {
1765 $lead = " ...";
1767 return "$lead$body";
1769 } elsif ($where eq 'center') {
1770 $str =~ m/^($endre)(.*)$/;
1771 my ($left, $str) = ($1, $2);
1772 $str =~ m/^(.*?)($begre)$/;
1773 my ($mid, $right) = ($1, $2);
1774 if (length($mid) > 5) {
1775 $mid = " ... ";
1777 return "$left$mid$right";
1779 } else {
1780 $str =~ m/^($endre)(.*)$/;
1781 my $body = $1;
1782 my $tail = $2;
1783 if (length($tail) > 4) {
1784 $tail = "... ";
1786 return "$body$tail";
1790 # takes the same arguments as chop_str, but also wraps a <span> around the
1791 # result with a title attribute if it does get chopped. Additionally, the
1792 # string is HTML-escaped.
1793 sub chop_and_escape_str {
1794 my ($str) = @_;
1796 my $chopped = chop_str(@_);
1797 $str = to_utf8($str);
1798 if ($chopped eq $str) {
1799 return esc_html($chopped);
1800 } else {
1801 $str =~ s/[[:cntrl:]]/?/g;
1802 return $cgi->span({-title=>$str}, esc_html($chopped));
1806 # Highlight selected fragments of string, using given CSS class,
1807 # and escape HTML. It is assumed that fragments do not overlap.
1808 # Regions are passed as list of pairs (array references).
1810 # Example: esc_html_hl_regions("foobar", "mark", [ 0, 3 ]) returns
1811 # '<span class="mark">foo</span>bar'
1812 sub esc_html_hl_regions {
1813 my ($str, $css_class, @sel) = @_;
1814 my %opts = grep { ref($_) ne 'ARRAY' } @sel;
1815 @sel = grep { ref($_) eq 'ARRAY' } @sel;
1816 return esc_html($str, %opts) unless @sel;
1818 my $out = '';
1819 my $pos = 0;
1821 for my $s (@sel) {
1822 my ($begin, $end) = @$s;
1824 # Don't create empty <span> elements.
1825 next if $end <= $begin;
1827 my $escaped = esc_html(substr($str, $begin, $end - $begin),
1828 %opts);
1830 $out .= esc_html(substr($str, $pos, $begin - $pos), %opts)
1831 if ($begin - $pos > 0);
1832 $out .= $cgi->span({-class => $css_class}, $escaped);
1834 $pos = $end;
1836 $out .= esc_html(substr($str, $pos), %opts)
1837 if ($pos < length($str));
1839 return $out;
1842 # return positions of beginning and end of each match
1843 sub matchpos_list {
1844 my ($str, $regexp) = @_;
1845 return unless (defined $str && defined $regexp);
1847 my @matches;
1848 while ($str =~ /$regexp/g) {
1849 push @matches, [$-[0], $+[0]];
1851 return @matches;
1854 # highlight match (if any), and escape HTML
1855 sub esc_html_match_hl {
1856 my ($str, $regexp) = @_;
1857 return esc_html($str) unless defined $regexp;
1859 my @matches = matchpos_list($str, $regexp);
1860 return esc_html($str) unless @matches;
1862 return esc_html_hl_regions($str, 'match', @matches);
1866 # highlight match (if any) of shortened string, and escape HTML
1867 sub esc_html_match_hl_chopped {
1868 my ($str, $chopped, $regexp) = @_;
1869 return esc_html_match_hl($str, $regexp) unless defined $chopped;
1871 my @matches = matchpos_list($str, $regexp);
1872 return esc_html($chopped) unless @matches;
1874 # filter matches so that we mark chopped string
1875 my $tail = "... "; # see chop_str
1876 unless ($chopped =~ s/\Q$tail\E$//) {
1877 $tail = '';
1879 my $chop_len = length($chopped);
1880 my $tail_len = length($tail);
1881 my @filtered;
1883 for my $m (@matches) {
1884 if ($m->[0] > $chop_len) {
1885 push @filtered, [ $chop_len, $chop_len + $tail_len ] if ($tail_len > 0);
1886 last;
1887 } elsif ($m->[1] > $chop_len) {
1888 push @filtered, [ $m->[0], $chop_len + $tail_len ];
1889 last;
1891 push @filtered, $m;
1894 return esc_html_hl_regions($chopped . $tail, 'match', @filtered);
1897 ## ----------------------------------------------------------------------
1898 ## functions returning short strings
1900 # CSS class for given age value (in seconds)
1901 sub age_class {
1902 my $age = shift;
1904 if (!defined $age) {
1905 return "noage";
1906 } elsif ($age < 60*60*2) {
1907 return "age0";
1908 } elsif ($age < 60*60*24*2) {
1909 return "age1";
1910 } else {
1911 return "age2";
1915 # convert age in seconds to "nn units ago" string
1916 sub age_string {
1917 my $age = shift;
1918 my $age_str;
1920 if ($age > 60*60*24*365*2) {
1921 $age_str = (int $age/60/60/24/365);
1922 $age_str .= " years ago";
1923 } elsif ($age > 60*60*24*(365/12)*2) {
1924 $age_str = int $age/60/60/24/(365/12);
1925 $age_str .= " months ago";
1926 } elsif ($age > 60*60*24*7*2) {
1927 $age_str = int $age/60/60/24/7;
1928 $age_str .= " weeks ago";
1929 } elsif ($age > 60*60*24*2) {
1930 $age_str = int $age/60/60/24;
1931 $age_str .= " days ago";
1932 } elsif ($age > 60*60*2) {
1933 $age_str = int $age/60/60;
1934 $age_str .= " hours ago";
1935 } elsif ($age > 60*2) {
1936 $age_str = int $age/60;
1937 $age_str .= " min ago";
1938 } elsif ($age > 2) {
1939 $age_str = int $age;
1940 $age_str .= " sec ago";
1941 } else {
1942 $age_str .= " right now";
1944 return $age_str;
1947 use constant {
1948 S_IFINVALID => 0030000,
1949 S_IFGITLINK => 0160000,
1952 # submodule/subproject, a commit object reference
1953 sub S_ISGITLINK {
1954 my $mode = shift;
1956 return (($mode & S_IFMT) == S_IFGITLINK)
1959 # convert file mode in octal to symbolic file mode string
1960 sub mode_str {
1961 my $mode = oct shift;
1963 if (S_ISGITLINK($mode)) {
1964 return 'm---------';
1965 } elsif (S_ISDIR($mode & S_IFMT)) {
1966 return 'drwxr-xr-x';
1967 } elsif (S_ISLNK($mode)) {
1968 return 'lrwxrwxrwx';
1969 } elsif (S_ISREG($mode)) {
1970 # git cares only about the executable bit
1971 if ($mode & S_IXUSR) {
1972 return '-rwxr-xr-x';
1973 } else {
1974 return '-rw-r--r--';
1976 } else {
1977 return '----------';
1981 # convert file mode in octal to file type string
1982 sub file_type {
1983 my $mode = shift;
1985 if ($mode !~ m/^[0-7]+$/) {
1986 return $mode;
1987 } else {
1988 $mode = oct $mode;
1991 if (S_ISGITLINK($mode)) {
1992 return "submodule";
1993 } elsif (S_ISDIR($mode & S_IFMT)) {
1994 return "directory";
1995 } elsif (S_ISLNK($mode)) {
1996 return "symlink";
1997 } elsif (S_ISREG($mode)) {
1998 return "file";
1999 } else {
2000 return "unknown";
2004 # convert file mode in octal to file type description string
2005 sub file_type_long {
2006 my $mode = shift;
2008 if ($mode !~ m/^[0-7]+$/) {
2009 return $mode;
2010 } else {
2011 $mode = oct $mode;
2014 if (S_ISGITLINK($mode)) {
2015 return "submodule";
2016 } elsif (S_ISDIR($mode & S_IFMT)) {
2017 return "directory";
2018 } elsif (S_ISLNK($mode)) {
2019 return "symlink";
2020 } elsif (S_ISREG($mode)) {
2021 if ($mode & S_IXUSR) {
2022 return "executable";
2023 } else {
2024 return "file";
2026 } else {
2027 return "unknown";
2032 ## ----------------------------------------------------------------------
2033 ## functions returning short HTML fragments, or transforming HTML fragments
2034 ## which don't belong to other sections
2036 # format line of commit message.
2037 sub format_log_line_html {
2038 my $line = shift;
2040 $line = esc_html($line, -nbsp=>1);
2041 $line =~ s{
2044 # The output of "git describe", e.g. v2.10.0-297-gf6727b0
2045 # or hadoop-20160921-113441-20-g094fb7d
2046 (?<!-) # see strbuf_check_tag_ref(). Tags can't start with -
2047 [A-Za-z0-9.-]+
2048 (?!\.) # refs can't end with ".", see check_refname_format()
2049 -g[0-9a-fA-F]{7,40}
2051 # Just a normal looking Git SHA1
2052 [0-9a-fA-F]{7,40}
2056 $cgi->a({-href => href(action=>"object", hash=>$1),
2057 -class => "text"}, $1);
2058 }egx;
2060 return $line;
2063 # format marker of refs pointing to given object
2065 # the destination action is chosen based on object type and current context:
2066 # - for annotated tags, we choose the tag view unless it's the current view
2067 # already, in which case we go to shortlog view
2068 # - for other refs, we keep the current view if we're in history, shortlog or
2069 # log view, and select shortlog otherwise
2070 sub format_ref_marker {
2071 my ($refs, $id) = @_;
2072 my $markers = '';
2074 if (defined $refs->{$id}) {
2075 foreach my $ref (@{$refs->{$id}}) {
2076 # this code exploits the fact that non-lightweight tags are the
2077 # only indirect objects, and that they are the only objects for which
2078 # we want to use tag instead of shortlog as action
2079 my ($type, $name) = qw();
2080 my $indirect = ($ref =~ s/\^\{\}$//);
2081 # e.g. tags/v2.6.11 or heads/next
2082 if ($ref =~ m!^(.*?)s?/(.*)$!) {
2083 $type = $1;
2084 $name = $2;
2085 } else {
2086 $type = "ref";
2087 $name = $ref;
2090 my $class = $type;
2091 $class .= " indirect" if $indirect;
2093 my $dest_action = "shortlog";
2095 if ($indirect) {
2096 $dest_action = "tag" unless $action eq "tag";
2097 } elsif ($action =~ /^(history|(short)?log)$/) {
2098 $dest_action = $action;
2101 my $dest = "";
2102 $dest .= "refs/" unless $ref =~ m!^refs/!;
2103 $dest .= $ref;
2105 my $link = $cgi->a({
2106 -href => href(
2107 action=>$dest_action,
2108 hash=>$dest
2109 )}, esc_html($name));
2111 $markers .= " <span class=\"".esc_attr($class)."\" title=\"".esc_attr($ref)."\">" .
2112 $link . "</span>";
2116 if ($markers) {
2117 return ' <span class="refs">'. $markers . '</span>';
2118 } else {
2119 return "";
2123 # format, perhaps shortened and with markers, title line
2124 sub format_subject_html {
2125 my ($long, $short, $href, $extra) = @_;
2126 $extra = '' unless defined($extra);
2128 if (length($short) < length($long)) {
2129 $long =~ s/[[:cntrl:]]/?/g;
2130 return $cgi->a({-href => $href, -class => "list subject",
2131 -title => to_utf8($long)},
2132 esc_html($short)) . $extra;
2133 } else {
2134 return $cgi->a({-href => $href, -class => "list subject"},
2135 esc_html($long)) . $extra;
2139 # Rather than recomputing the url for an email multiple times, we cache it
2140 # after the first hit. This gives a visible benefit in views where the avatar
2141 # for the same email is used repeatedly (e.g. shortlog).
2142 # The cache is shared by all avatar engines (currently gravatar only), which
2143 # are free to use it as preferred. Since only one avatar engine is used for any
2144 # given page, there's no risk for cache conflicts.
2145 our %avatar_cache = ();
2147 # Compute the picon url for a given email, by using the picon search service over at
2148 # http://www.cs.indiana.edu/picons/search.html
2149 sub picon_url {
2150 my $email = lc shift;
2151 if (!$avatar_cache{$email}) {
2152 my ($user, $domain) = split('@', $email);
2153 $avatar_cache{$email} =
2154 "//www.cs.indiana.edu/cgi-pub/kinzler/piconsearch.cgi/" .
2155 "$domain/$user/" .
2156 "users+domains+unknown/up/single";
2158 return $avatar_cache{$email};
2161 # Compute the gravatar url for a given email, if it's not in the cache already.
2162 # Gravatar stores only the part of the URL before the size, since that's the
2163 # one computationally more expensive. This also allows reuse of the cache for
2164 # different sizes (for this particular engine).
2165 sub gravatar_url {
2166 my $email = lc shift;
2167 my $size = shift;
2168 $avatar_cache{$email} ||=
2169 "//www.gravatar.com/avatar/" .
2170 Digest::MD5::md5_hex($email) . "?s=";
2171 return $avatar_cache{$email} . $size;
2174 # Insert an avatar for the given $email at the given $size if the feature
2175 # is enabled.
2176 sub git_get_avatar {
2177 my ($email, %opts) = @_;
2178 my $pre_white = ($opts{-pad_before} ? "&nbsp;" : "");
2179 my $post_white = ($opts{-pad_after} ? "&nbsp;" : "");
2180 $opts{-size} ||= 'default';
2181 my $size = $avatar_size{$opts{-size}} || $avatar_size{'default'};
2182 my $url = "";
2183 if ($git_avatar eq 'gravatar') {
2184 $url = gravatar_url($email, $size);
2185 } elsif ($git_avatar eq 'picon') {
2186 $url = picon_url($email);
2188 # Other providers can be added by extending the if chain, defining $url
2189 # as needed. If no variant puts something in $url, we assume avatars
2190 # are completely disabled/unavailable.
2191 if ($url) {
2192 return $pre_white .
2193 "<img width=\"$size\" " .
2194 "class=\"avatar\" " .
2195 "src=\"".esc_url($url)."\" " .
2196 "alt=\"\" " .
2197 "/>" . $post_white;
2198 } else {
2199 return "";
2203 sub format_search_author {
2204 my ($author, $searchtype, $displaytext) = @_;
2205 my $have_search = gitweb_check_feature('search');
2207 if ($have_search) {
2208 my $performed = "";
2209 if ($searchtype eq 'author') {
2210 $performed = "authored";
2211 } elsif ($searchtype eq 'committer') {
2212 $performed = "committed";
2215 return $cgi->a({-href => href(action=>"search", hash=>$hash,
2216 searchtext=>$author,
2217 searchtype=>$searchtype), class=>"list",
2218 title=>"Search for commits $performed by $author"},
2219 $displaytext);
2221 } else {
2222 return $displaytext;
2226 # format the author name of the given commit with the given tag
2227 # the author name is chopped and escaped according to the other
2228 # optional parameters (see chop_str).
2229 sub format_author_html {
2230 my $tag = shift;
2231 my $co = shift;
2232 my $author = chop_and_escape_str($co->{'author_name'}, @_);
2233 return "<$tag class=\"author\">" .
2234 format_search_author($co->{'author_name'}, "author",
2235 git_get_avatar($co->{'author_email'}, -pad_after => 1) .
2236 $author) .
2237 "</$tag>";
2240 # format git diff header line, i.e. "diff --(git|combined|cc) ..."
2241 sub format_git_diff_header_line {
2242 my $line = shift;
2243 my $diffinfo = shift;
2244 my ($from, $to) = @_;
2246 if ($diffinfo->{'nparents'}) {
2247 # combined diff
2248 $line =~ s!^(diff (.*?) )"?.*$!$1!;
2249 if ($to->{'href'}) {
2250 $line .= $cgi->a({-href => $to->{'href'}, -class => "path"},
2251 esc_path($to->{'file'}));
2252 } else { # file was deleted (no href)
2253 $line .= esc_path($to->{'file'});
2255 } else {
2256 # "ordinary" diff
2257 $line =~ s!^(diff (.*?) )"?a/.*$!$1!;
2258 if ($from->{'href'}) {
2259 $line .= $cgi->a({-href => $from->{'href'}, -class => "path"},
2260 'a/' . esc_path($from->{'file'}));
2261 } else { # file was added (no href)
2262 $line .= 'a/' . esc_path($from->{'file'});
2264 $line .= ' ';
2265 if ($to->{'href'}) {
2266 $line .= $cgi->a({-href => $to->{'href'}, -class => "path"},
2267 'b/' . esc_path($to->{'file'}));
2268 } else { # file was deleted
2269 $line .= 'b/' . esc_path($to->{'file'});
2273 return "<div class=\"diff header\">$line</div>\n";
2276 # format extended diff header line, before patch itself
2277 sub format_extended_diff_header_line {
2278 my $line = shift;
2279 my $diffinfo = shift;
2280 my ($from, $to) = @_;
2282 # match <path>
2283 if ($line =~ s!^((copy|rename) from ).*$!$1! && $from->{'href'}) {
2284 $line .= $cgi->a({-href=>$from->{'href'}, -class=>"path"},
2285 esc_path($from->{'file'}));
2287 if ($line =~ s!^((copy|rename) to ).*$!$1! && $to->{'href'}) {
2288 $line .= $cgi->a({-href=>$to->{'href'}, -class=>"path"},
2289 esc_path($to->{'file'}));
2291 # match single <mode>
2292 if ($line =~ m/\s(\d{6})$/) {
2293 $line .= '<span class="info"> (' .
2294 file_type_long($1) .
2295 ')</span>';
2297 # match <hash>
2298 if ($line =~ m/^index [0-9a-fA-F]{40},[0-9a-fA-F]{40}/) {
2299 # can match only for combined diff
2300 $line = 'index ';
2301 for (my $i = 0; $i < $diffinfo->{'nparents'}; $i++) {
2302 if ($from->{'href'}[$i]) {
2303 $line .= $cgi->a({-href=>$from->{'href'}[$i],
2304 -class=>"hash"},
2305 substr($diffinfo->{'from_id'}[$i],0,7));
2306 } else {
2307 $line .= '0' x 7;
2309 # separator
2310 $line .= ',' if ($i < $diffinfo->{'nparents'} - 1);
2312 $line .= '..';
2313 if ($to->{'href'}) {
2314 $line .= $cgi->a({-href=>$to->{'href'}, -class=>"hash"},
2315 substr($diffinfo->{'to_id'},0,7));
2316 } else {
2317 $line .= '0' x 7;
2320 } elsif ($line =~ m/^index [0-9a-fA-F]{40}..[0-9a-fA-F]{40}/) {
2321 # can match only for ordinary diff
2322 my ($from_link, $to_link);
2323 if ($from->{'href'}) {
2324 $from_link = $cgi->a({-href=>$from->{'href'}, -class=>"hash"},
2325 substr($diffinfo->{'from_id'},0,7));
2326 } else {
2327 $from_link = '0' x 7;
2329 if ($to->{'href'}) {
2330 $to_link = $cgi->a({-href=>$to->{'href'}, -class=>"hash"},
2331 substr($diffinfo->{'to_id'},0,7));
2332 } else {
2333 $to_link = '0' x 7;
2335 my ($from_id, $to_id) = ($diffinfo->{'from_id'}, $diffinfo->{'to_id'});
2336 $line =~ s!$from_id\.\.$to_id!$from_link..$to_link!;
2339 return $line . "<br/>\n";
2342 # format from-file/to-file diff header
2343 sub format_diff_from_to_header {
2344 my ($from_line, $to_line, $diffinfo, $from, $to, @parents) = @_;
2345 my $line;
2346 my $result = '';
2348 $line = $from_line;
2349 #assert($line =~ m/^---/) if DEBUG;
2350 # no extra formatting for "^--- /dev/null"
2351 if (! $diffinfo->{'nparents'}) {
2352 # ordinary (single parent) diff
2353 if ($line =~ m!^--- "?a/!) {
2354 if ($from->{'href'}) {
2355 $line = '--- a/' .
2356 $cgi->a({-href=>$from->{'href'}, -class=>"path"},
2357 esc_path($from->{'file'}));
2358 } else {
2359 $line = '--- a/' .
2360 esc_path($from->{'file'});
2363 $result .= qq!<div class="diff from_file">$line</div>\n!;
2365 } else {
2366 # combined diff (merge commit)
2367 for (my $i = 0; $i < $diffinfo->{'nparents'}; $i++) {
2368 if ($from->{'href'}[$i]) {
2369 $line = '--- ' .
2370 $cgi->a({-href=>href(action=>"blobdiff",
2371 hash_parent=>$diffinfo->{'from_id'}[$i],
2372 hash_parent_base=>$parents[$i],
2373 file_parent=>$from->{'file'}[$i],
2374 hash=>$diffinfo->{'to_id'},
2375 hash_base=>$hash,
2376 file_name=>$to->{'file'}),
2377 -class=>"path",
2378 -title=>"diff" . ($i+1)},
2379 $i+1) .
2380 '/' .
2381 $cgi->a({-href=>$from->{'href'}[$i], -class=>"path"},
2382 esc_path($from->{'file'}[$i]));
2383 } else {
2384 $line = '--- /dev/null';
2386 $result .= qq!<div class="diff from_file">$line</div>\n!;
2390 $line = $to_line;
2391 #assert($line =~ m/^\+\+\+/) if DEBUG;
2392 # no extra formatting for "^+++ /dev/null"
2393 if ($line =~ m!^\+\+\+ "?b/!) {
2394 if ($to->{'href'}) {
2395 $line = '+++ b/' .
2396 $cgi->a({-href=>$to->{'href'}, -class=>"path"},
2397 esc_path($to->{'file'}));
2398 } else {
2399 $line = '+++ b/' .
2400 esc_path($to->{'file'});
2403 $result .= qq!<div class="diff to_file">$line</div>\n!;
2405 return $result;
2408 # create note for patch simplified by combined diff
2409 sub format_diff_cc_simplified {
2410 my ($diffinfo, @parents) = @_;
2411 my $result = '';
2413 $result .= "<div class=\"diff header\">" .
2414 "diff --cc ";
2415 if (!is_deleted($diffinfo)) {
2416 $result .= $cgi->a({-href => href(action=>"blob",
2417 hash_base=>$hash,
2418 hash=>$diffinfo->{'to_id'},
2419 file_name=>$diffinfo->{'to_file'}),
2420 -class => "path"},
2421 esc_path($diffinfo->{'to_file'}));
2422 } else {
2423 $result .= esc_path($diffinfo->{'to_file'});
2425 $result .= "</div>\n" . # class="diff header"
2426 "<div class=\"diff nodifferences\">" .
2427 "Simple merge" .
2428 "</div>\n"; # class="diff nodifferences"
2430 return $result;
2433 sub diff_line_class {
2434 my ($line, $from, $to) = @_;
2436 # ordinary diff
2437 my $num_sign = 1;
2438 # combined diff
2439 if ($from && $to && ref($from->{'href'}) eq "ARRAY") {
2440 $num_sign = scalar @{$from->{'href'}};
2443 my @diff_line_classifier = (
2444 { regexp => qr/^\@\@{$num_sign} /, class => "chunk_header"},
2445 { regexp => qr/^\\/, class => "incomplete" },
2446 { regexp => qr/^ {$num_sign}/, class => "ctx" },
2447 # classifier for context must come before classifier add/rem,
2448 # or we would have to use more complicated regexp, for example
2449 # qr/(?= {0,$m}\+)[+ ]{$num_sign}/, where $m = $num_sign - 1;
2450 { regexp => qr/^[+ ]{$num_sign}/, class => "add" },
2451 { regexp => qr/^[- ]{$num_sign}/, class => "rem" },
2453 for my $clsfy (@diff_line_classifier) {
2454 return $clsfy->{'class'}
2455 if ($line =~ $clsfy->{'regexp'});
2458 # fallback
2459 return "";
2462 # assumes that $from and $to are defined and correctly filled,
2463 # and that $line holds a line of chunk header for unified diff
2464 sub format_unidiff_chunk_header {
2465 my ($line, $from, $to) = @_;
2467 my ($from_text, $from_start, $from_lines, $to_text, $to_start, $to_lines, $section) =
2468 $line =~ m/^\@{2} (-(\d+)(?:,(\d+))?) (\+(\d+)(?:,(\d+))?) \@{2}(.*)$/;
2470 $from_lines = 0 unless defined $from_lines;
2471 $to_lines = 0 unless defined $to_lines;
2473 if ($from->{'href'}) {
2474 $from_text = $cgi->a({-href=>"$from->{'href'}#l$from_start",
2475 -class=>"list"}, $from_text);
2477 if ($to->{'href'}) {
2478 $to_text = $cgi->a({-href=>"$to->{'href'}#l$to_start",
2479 -class=>"list"}, $to_text);
2481 $line = "<span class=\"chunk_info\">@@ $from_text $to_text @@</span>" .
2482 "<span class=\"section\">" . esc_html($section, -nbsp=>1) . "</span>";
2483 return $line;
2486 # assumes that $from and $to are defined and correctly filled,
2487 # and that $line holds a line of chunk header for combined diff
2488 sub format_cc_diff_chunk_header {
2489 my ($line, $from, $to) = @_;
2491 my ($prefix, $ranges, $section) = $line =~ m/^(\@+) (.*?) \@+(.*)$/;
2492 my (@from_text, @from_start, @from_nlines, $to_text, $to_start, $to_nlines);
2494 @from_text = split(' ', $ranges);
2495 for (my $i = 0; $i < @from_text; ++$i) {
2496 ($from_start[$i], $from_nlines[$i]) =
2497 (split(',', substr($from_text[$i], 1)), 0);
2500 $to_text = pop @from_text;
2501 $to_start = pop @from_start;
2502 $to_nlines = pop @from_nlines;
2504 $line = "<span class=\"chunk_info\">$prefix ";
2505 for (my $i = 0; $i < @from_text; ++$i) {
2506 if ($from->{'href'}[$i]) {
2507 $line .= $cgi->a({-href=>"$from->{'href'}[$i]#l$from_start[$i]",
2508 -class=>"list"}, $from_text[$i]);
2509 } else {
2510 $line .= $from_text[$i];
2512 $line .= " ";
2514 if ($to->{'href'}) {
2515 $line .= $cgi->a({-href=>"$to->{'href'}#l$to_start",
2516 -class=>"list"}, $to_text);
2517 } else {
2518 $line .= $to_text;
2520 $line .= " $prefix</span>" .
2521 "<span class=\"section\">" . esc_html($section, -nbsp=>1) . "</span>";
2522 return $line;
2525 # process patch (diff) line (not to be used for diff headers),
2526 # returning HTML-formatted (but not wrapped) line.
2527 # If the line is passed as a reference, it is treated as HTML and not
2528 # esc_html()'ed.
2529 sub format_diff_line {
2530 my ($line, $diff_class, $from, $to) = @_;
2532 if (ref($line)) {
2533 $line = $$line;
2534 } else {
2535 chomp $line;
2536 $line = untabify($line);
2538 if ($from && $to && $line =~ m/^\@{2} /) {
2539 $line = format_unidiff_chunk_header($line, $from, $to);
2540 } elsif ($from && $to && $line =~ m/^\@{3}/) {
2541 $line = format_cc_diff_chunk_header($line, $from, $to);
2542 } else {
2543 $line = esc_html($line, -nbsp=>1);
2547 my $diff_classes = "diff";
2548 $diff_classes .= " $diff_class" if ($diff_class);
2549 $line = "<div class=\"$diff_classes\">$line</div>\n";
2551 return $line;
2554 # Generates undef or something like "_snapshot_" or "snapshot (_tbz2_ _zip_)",
2555 # linked. Pass the hash of the tree/commit to snapshot.
2556 sub format_snapshot_links {
2557 my ($hash) = @_;
2558 my $num_fmts = @snapshot_fmts;
2559 if ($num_fmts > 1) {
2560 # A parenthesized list of links bearing format names.
2561 # e.g. "snapshot (_tar.gz_ _zip_)"
2562 return "snapshot (" . join(' ', map
2563 $cgi->a({
2564 -href => href(
2565 action=>"snapshot",
2566 hash=>$hash,
2567 snapshot_format=>$_
2569 }, $known_snapshot_formats{$_}{'display'})
2570 , @snapshot_fmts) . ")";
2571 } elsif ($num_fmts == 1) {
2572 # A single "snapshot" link whose tooltip bears the format name.
2573 # i.e. "_snapshot_"
2574 my ($fmt) = @snapshot_fmts;
2575 return
2576 $cgi->a({
2577 -href => href(
2578 action=>"snapshot",
2579 hash=>$hash,
2580 snapshot_format=>$fmt
2582 -title => "in format: $known_snapshot_formats{$fmt}{'display'}"
2583 }, "snapshot");
2584 } else { # $num_fmts == 0
2585 return undef;
2589 ## ......................................................................
2590 ## functions returning values to be passed, perhaps after some
2591 ## transformation, to other functions; e.g. returning arguments to href()
2593 # returns hash to be passed to href to generate gitweb URL
2594 # in -title key it returns description of link
2595 sub get_feed_info {
2596 my $format = shift || 'Atom';
2597 my %res = (action => lc($format));
2598 my $matched_ref = 0;
2600 # feed links are possible only for project views
2601 return unless (defined $project);
2602 # some views should link to OPML, or to generic project feed,
2603 # or don't have specific feed yet (so they should use generic)
2604 return if (!$action || $action =~ /^(?:tags|heads|forks|tag|search)$/x);
2606 my $branch = undef;
2607 # branches refs uses 'refs/' + $get_branch_refs()[x] + '/' prefix
2608 # (fullname) to differentiate from tag links; this also makes
2609 # possible to detect branch links
2610 for my $ref (get_branch_refs()) {
2611 if ((defined $hash_base && $hash_base =~ m!^refs/\Q$ref\E/(.*)$!) ||
2612 (defined $hash && $hash =~ m!^refs/\Q$ref\E/(.*)$!)) {
2613 $branch = $1;
2614 $matched_ref = $ref;
2615 last;
2618 # find log type for feed description (title)
2619 my $type = 'log';
2620 if (defined $file_name) {
2621 $type = "history of $file_name";
2622 $type .= "/" if ($action eq 'tree');
2623 $type .= " on '$branch'" if (defined $branch);
2624 } else {
2625 $type = "log of $branch" if (defined $branch);
2628 $res{-title} = $type;
2629 $res{'hash'} = (defined $branch ? "refs/$matched_ref/$branch" : undef);
2630 $res{'file_name'} = $file_name;
2632 return %res;
2635 ## ----------------------------------------------------------------------
2636 ## git utility subroutines, invoking git commands
2638 # returns path to the core git executable and the --git-dir parameter as list
2639 sub git_cmd {
2640 $number_of_git_cmds++;
2641 return $GIT, '--git-dir='.$git_dir;
2644 # quote the given arguments for passing them to the shell
2645 # quote_command("command", "arg 1", "arg with ' and ! characters")
2646 # => "'command' 'arg 1' 'arg with '\'' and '\!' characters'"
2647 # Try to avoid using this function wherever possible.
2648 sub quote_command {
2649 return join(' ',
2650 map { my $a = $_; $a =~ s/(['!])/'\\$1'/g; "'$a'" } @_ );
2653 # get HEAD ref of given project as hash
2654 sub git_get_head_hash {
2655 return git_get_full_hash(shift, 'HEAD');
2658 sub git_get_full_hash {
2659 return git_get_hash(@_);
2662 sub git_get_short_hash {
2663 return git_get_hash(@_, '--short=7');
2666 sub git_get_hash {
2667 my ($project, $hash, @options) = @_;
2668 my $o_git_dir = $git_dir;
2669 my $retval = undef;
2670 $git_dir = "$projectroot/$project";
2671 if (open my $fd, '-|', git_cmd(), 'rev-parse',
2672 '--verify', '-q', @options, $hash) {
2673 $retval = <$fd>;
2674 chomp $retval if defined $retval;
2675 close $fd;
2677 if (defined $o_git_dir) {
2678 $git_dir = $o_git_dir;
2680 return $retval;
2683 # get type of given object
2684 sub git_get_type {
2685 my $hash = shift;
2687 open my $fd, "-|", git_cmd(), "cat-file", '-t', $hash or return;
2688 my $type = <$fd>;
2689 close $fd or return;
2690 chomp $type;
2691 return $type;
2694 # repository configuration
2695 our $config_file = '';
2696 our %config;
2698 # store multiple values for single key as anonymous array reference
2699 # single values stored directly in the hash, not as [ <value> ]
2700 sub hash_set_multi {
2701 my ($hash, $key, $value) = @_;
2703 if (!exists $hash->{$key}) {
2704 $hash->{$key} = $value;
2705 } elsif (!ref $hash->{$key}) {
2706 $hash->{$key} = [ $hash->{$key}, $value ];
2707 } else {
2708 push @{$hash->{$key}}, $value;
2712 # return hash of git project configuration
2713 # optionally limited to some section, e.g. 'gitweb'
2714 sub git_parse_project_config {
2715 my $section_regexp = shift;
2716 my %config;
2718 local $/ = "\0";
2720 open my $fh, "-|", git_cmd(), "config", '-z', '-l',
2721 or return;
2723 while (my $keyval = <$fh>) {
2724 chomp $keyval;
2725 my ($key, $value) = split(/\n/, $keyval, 2);
2727 hash_set_multi(\%config, $key, $value)
2728 if (!defined $section_regexp || $key =~ /^(?:$section_regexp)\./o);
2730 close $fh;
2732 return %config;
2735 # convert config value to boolean: 'true' or 'false'
2736 # no value, number > 0, 'true' and 'yes' values are true
2737 # rest of values are treated as false (never as error)
2738 sub config_to_bool {
2739 my $val = shift;
2741 return 1 if !defined $val; # section.key
2743 # strip leading and trailing whitespace
2744 $val =~ s/^\s+//;
2745 $val =~ s/\s+$//;
2747 return (($val =~ /^\d+$/ && $val) || # section.key = 1
2748 ($val =~ /^(?:true|yes)$/i)); # section.key = true
2751 # convert config value to simple decimal number
2752 # an optional value suffix of 'k', 'm', or 'g' will cause the value
2753 # to be multiplied by 1024, 1048576, or 1073741824
2754 sub config_to_int {
2755 my $val = shift;
2757 # strip leading and trailing whitespace
2758 $val =~ s/^\s+//;
2759 $val =~ s/\s+$//;
2761 if (my ($num, $unit) = ($val =~ /^([0-9]*)([kmg])$/i)) {
2762 $unit = lc($unit);
2763 # unknown unit is treated as 1
2764 return $num * ($unit eq 'g' ? 1073741824 :
2765 $unit eq 'm' ? 1048576 :
2766 $unit eq 'k' ? 1024 : 1);
2768 return $val;
2771 # convert config value to array reference, if needed
2772 sub config_to_multi {
2773 my $val = shift;
2775 return ref($val) ? $val : (defined($val) ? [ $val ] : []);
2778 sub git_get_project_config {
2779 my ($key, $type) = @_;
2781 return unless defined $git_dir;
2783 # key sanity check
2784 return unless ($key);
2785 # only subsection, if exists, is case sensitive,
2786 # and not lowercased by 'git config -z -l'
2787 if (my ($hi, $mi, $lo) = ($key =~ /^([^.]*)\.(.*)\.([^.]*)$/)) {
2788 $lo =~ s/_//g;
2789 $key = join(".", lc($hi), $mi, lc($lo));
2790 return if ($lo =~ /\W/ || $hi =~ /\W/);
2791 } else {
2792 $key = lc($key);
2793 $key =~ s/_//g;
2794 return if ($key =~ /\W/);
2796 $key =~ s/^gitweb\.//;
2798 # type sanity check
2799 if (defined $type) {
2800 $type =~ s/^--//;
2801 $type = undef
2802 unless ($type eq 'bool' || $type eq 'int');
2805 # get config
2806 if (!defined $config_file ||
2807 $config_file ne "$git_dir/config") {
2808 %config = git_parse_project_config('gitweb');
2809 $config_file = "$git_dir/config";
2812 # check if config variable (key) exists
2813 return unless exists $config{"gitweb.$key"};
2815 # ensure given type
2816 if (!defined $type) {
2817 return $config{"gitweb.$key"};
2818 } elsif ($type eq 'bool') {
2819 # backward compatibility: 'git config --bool' returns true/false
2820 return config_to_bool($config{"gitweb.$key"}) ? 'true' : 'false';
2821 } elsif ($type eq 'int') {
2822 return config_to_int($config{"gitweb.$key"});
2824 return $config{"gitweb.$key"};
2827 # get hash of given path at given ref
2828 sub git_get_hash_by_path {
2829 my $base = shift;
2830 my $path = shift || return undef;
2831 my $type = shift;
2833 $path =~ s,/+$,,;
2835 open my $fd, "-|", git_cmd(), "ls-tree", $base, "--", $path
2836 or die_error(500, "Open git-ls-tree failed");
2837 my $line = <$fd>;
2838 close $fd or return undef;
2840 if (!defined $line) {
2841 # there is no tree or hash given by $path at $base
2842 return undef;
2845 #'100644 blob 0fa3f3a66fb6a137f6ec2c19351ed4d807070ffa panic.c'
2846 $line =~ m/^([0-9]+) (.+) ([0-9a-fA-F]{40})\t/;
2847 if (defined $type && $type ne $2) {
2848 # type doesn't match
2849 return undef;
2851 return $3;
2854 # get path of entry with given hash at given tree-ish (ref)
2855 # used to get 'from' filename for combined diff (merge commit) for renames
2856 sub git_get_path_by_hash {
2857 my $base = shift || return;
2858 my $hash = shift || return;
2860 local $/ = "\0";
2862 open my $fd, "-|", git_cmd(), "ls-tree", '-r', '-t', '-z', $base
2863 or return undef;
2864 while (my $line = <$fd>) {
2865 chomp $line;
2867 #'040000 tree 595596a6a9117ddba9fe379b6b012b558bac8423 gitweb'
2868 #'100644 blob e02e90f0429be0d2a69b76571101f20b8f75530f gitweb/README'
2869 if ($line =~ m/(?:[0-9]+) (?:.+) $hash\t(.+)$/) {
2870 close $fd;
2871 return $1;
2874 close $fd;
2875 return undef;
2878 ## ......................................................................
2879 ## git utility functions, directly accessing git repository
2881 # get the value of config variable either from file named as the variable
2882 # itself in the repository ($GIT_DIR/$name file), or from gitweb.$name
2883 # configuration variable in the repository config file.
2884 sub git_get_file_or_project_config {
2885 my ($path, $name) = @_;
2887 $git_dir = "$projectroot/$path";
2888 open my $fd, '<', "$git_dir/$name"
2889 or return git_get_project_config($name);
2890 my $conf = <$fd>;
2891 close $fd;
2892 if (defined $conf) {
2893 chomp $conf;
2895 return $conf;
2898 sub git_get_project_description {
2899 my $path = shift;
2900 return git_get_file_or_project_config($path, 'description');
2903 sub git_get_project_category {
2904 my $path = shift;
2905 return git_get_file_or_project_config($path, 'category');
2909 # supported formats:
2910 # * $GIT_DIR/ctags/<tagname> file (in 'ctags' subdirectory)
2911 # - if its contents is a number, use it as tag weight,
2912 # - otherwise add a tag with weight 1
2913 # * $GIT_DIR/ctags file, each line is a tag (with weight 1)
2914 # the same value multiple times increases tag weight
2915 # * `gitweb.ctag' multi-valued repo config variable
2916 sub git_get_project_ctags {
2917 my $project = shift;
2918 my $ctags = {};
2920 $git_dir = "$projectroot/$project";
2921 if (opendir my $dh, "$git_dir/ctags") {
2922 my @files = grep { -f $_ } map { "$git_dir/ctags/$_" } readdir($dh);
2923 foreach my $tagfile (@files) {
2924 open my $ct, '<', $tagfile
2925 or next;
2926 my $val = <$ct>;
2927 chomp $val if $val;
2928 close $ct;
2930 (my $ctag = $tagfile) =~ s#.*/##;
2931 if ($val =~ /^\d+$/) {
2932 $ctags->{$ctag} = $val;
2933 } else {
2934 $ctags->{$ctag} = 1;
2937 closedir $dh;
2939 } elsif (open my $fh, '<', "$git_dir/ctags") {
2940 while (my $line = <$fh>) {
2941 chomp $line;
2942 $ctags->{$line}++ if $line;
2944 close $fh;
2946 } else {
2947 my $taglist = config_to_multi(git_get_project_config('ctag'));
2948 foreach my $tag (@$taglist) {
2949 $ctags->{$tag}++;
2953 return $ctags;
2956 # return hash, where keys are content tags ('ctags'),
2957 # and values are sum of weights of given tag in every project
2958 sub git_gather_all_ctags {
2959 my $projects = shift;
2960 my $ctags = {};
2962 foreach my $p (@$projects) {
2963 foreach my $ct (keys %{$p->{'ctags'}}) {
2964 $ctags->{$ct} += $p->{'ctags'}->{$ct};
2968 return $ctags;
2971 sub git_populate_project_tagcloud {
2972 my $ctags = shift;
2974 # First, merge different-cased tags; tags vote on casing
2975 my %ctags_lc;
2976 foreach (keys %$ctags) {
2977 $ctags_lc{lc $_}->{count} += $ctags->{$_};
2978 if (not $ctags_lc{lc $_}->{topcount}
2979 or $ctags_lc{lc $_}->{topcount} < $ctags->{$_}) {
2980 $ctags_lc{lc $_}->{topcount} = $ctags->{$_};
2981 $ctags_lc{lc $_}->{topname} = $_;
2985 my $cloud;
2986 my $matched = $input_params{'ctag'};
2987 if (eval { require HTML::TagCloud; 1; }) {
2988 $cloud = HTML::TagCloud->new;
2989 foreach my $ctag (sort keys %ctags_lc) {
2990 # Pad the title with spaces so that the cloud looks
2991 # less crammed.
2992 my $title = esc_html($ctags_lc{$ctag}->{topname});
2993 $title =~ s/ /&nbsp;/g;
2994 $title =~ s/^/&nbsp;/g;
2995 $title =~ s/$/&nbsp;/g;
2996 if (defined $matched && $matched eq $ctag) {
2997 $title = qq(<span class="match">$title</span>);
2999 $cloud->add($title, href(project=>undef, ctag=>$ctag),
3000 $ctags_lc{$ctag}->{count});
3002 } else {
3003 $cloud = {};
3004 foreach my $ctag (keys %ctags_lc) {
3005 my $title = esc_html($ctags_lc{$ctag}->{topname}, -nbsp=>1);
3006 if (defined $matched && $matched eq $ctag) {
3007 $title = qq(<span class="match">$title</span>);
3009 $cloud->{$ctag}{count} = $ctags_lc{$ctag}->{count};
3010 $cloud->{$ctag}{ctag} =
3011 $cgi->a({-href=>href(project=>undef, ctag=>$ctag)}, $title);
3014 return $cloud;
3017 sub git_show_project_tagcloud {
3018 my ($cloud, $count) = @_;
3019 if (ref $cloud eq 'HTML::TagCloud') {
3020 return $cloud->html_and_css($count);
3021 } else {
3022 my @tags = sort { $cloud->{$a}->{'count'} <=> $cloud->{$b}->{'count'} } keys %$cloud;
3023 return
3024 '<div id="htmltagcloud"'.($project ? '' : ' align="center"').'>' .
3025 join (', ', map {
3026 $cloud->{$_}->{'ctag'}
3027 } splice(@tags, 0, $count)) .
3028 '</div>';
3032 sub git_get_project_url_list {
3033 my $path = shift;
3035 $git_dir = "$projectroot/$path";
3036 open my $fd, '<', "$git_dir/cloneurl"
3037 or return wantarray ?
3038 @{ config_to_multi(git_get_project_config('url')) } :
3039 config_to_multi(git_get_project_config('url'));
3040 my @git_project_url_list = map { chomp; $_ } <$fd>;
3041 close $fd;
3043 return wantarray ? @git_project_url_list : \@git_project_url_list;
3046 sub git_get_projects_list {
3047 my $filter = shift || '';
3048 my $paranoid = shift;
3049 my @list;
3051 if (-d $projects_list) {
3052 # search in directory
3053 my $dir = $projects_list;
3054 # remove the trailing "/"
3055 $dir =~ s!/+$!!;
3056 my $pfxlen = length("$dir");
3057 my $pfxdepth = ($dir =~ tr!/!!);
3058 # when filtering, search only given subdirectory
3059 if ($filter && !$paranoid) {
3060 $dir .= "/$filter";
3061 $dir =~ s!/+$!!;
3064 File::Find::find({
3065 follow_fast => 1, # follow symbolic links
3066 follow_skip => 2, # ignore duplicates
3067 dangling_symlinks => 0, # ignore dangling symlinks, silently
3068 wanted => sub {
3069 # global variables
3070 our $project_maxdepth;
3071 our $projectroot;
3072 # skip project-list toplevel, if we get it.
3073 return if (m!^[/.]$!);
3074 # only directories can be git repositories
3075 return unless (-d $_);
3076 # need search permission
3077 return unless (-x $_);
3078 # don't traverse too deep (Find is super slow on os x)
3079 # $project_maxdepth excludes depth of $projectroot
3080 if (($File::Find::name =~ tr!/!!) - $pfxdepth > $project_maxdepth) {
3081 $File::Find::prune = 1;
3082 return;
3085 my $path = substr($File::Find::name, $pfxlen + 1);
3086 # paranoidly only filter here
3087 if ($paranoid && $filter && $path !~ m!^\Q$filter\E/!) {
3088 next;
3090 # we check related file in $projectroot
3091 if (check_export_ok("$projectroot/$path")) {
3092 push @list, { path => $path };
3093 $File::Find::prune = 1;
3096 }, "$dir");
3098 } elsif (-f $projects_list) {
3099 # read from file(url-encoded):
3100 # 'git%2Fgit.git Linus+Torvalds'
3101 # 'libs%2Fklibc%2Fklibc.git H.+Peter+Anvin'
3102 # 'linux%2Fhotplug%2Fudev.git Greg+Kroah-Hartman'
3103 open my $fd, '<', $projects_list or return;
3104 PROJECT:
3105 while (my $line = <$fd>) {
3106 chomp $line;
3107 my ($path, $owner) = split ' ', $line;
3108 $path = unescape($path);
3109 $owner = unescape($owner);
3110 if (!defined $path) {
3111 next;
3113 # if $filter is rpovided, check if $path begins with $filter
3114 if ($filter && $path !~ m!^\Q$filter\E/!) {
3115 next;
3117 if (check_export_ok("$projectroot/$path")) {
3118 my $pr = {
3119 path => $path
3121 if ($owner) {
3122 $pr->{'owner'} = to_utf8($owner);
3124 push @list, $pr;
3127 close $fd;
3129 return @list;
3132 # written with help of Tree::Trie module (Perl Artistic License, GPL compatible)
3133 # as side effects it sets 'forks' field to list of forks for forked projects
3134 sub filter_forks_from_projects_list {
3135 my $projects = shift;
3137 my %trie; # prefix tree of directories (path components)
3138 # generate trie out of those directories that might contain forks
3139 foreach my $pr (@$projects) {
3140 my $path = $pr->{'path'};
3141 $path =~ s/\.git$//; # forks of 'repo.git' are in 'repo/' directory
3142 next if ($path =~ m!/$!); # skip non-bare repositories, e.g. 'repo/.git'
3143 next unless ($path); # skip '.git' repository: tests, git-instaweb
3144 next unless (-d "$projectroot/$path"); # containing directory exists
3145 $pr->{'forks'} = []; # there can be 0 or more forks of project
3147 # add to trie
3148 my @dirs = split('/', $path);
3149 # walk the trie, until either runs out of components or out of trie
3150 my $ref = \%trie;
3151 while (scalar @dirs &&
3152 exists($ref->{$dirs[0]})) {
3153 $ref = $ref->{shift @dirs};
3155 # create rest of trie structure from rest of components
3156 foreach my $dir (@dirs) {
3157 $ref = $ref->{$dir} = {};
3159 # create end marker, store $pr as a data
3160 $ref->{''} = $pr if (!exists $ref->{''});
3163 # filter out forks, by finding shortest prefix match for paths
3164 my @filtered;
3165 PROJECT:
3166 foreach my $pr (@$projects) {
3167 # trie lookup
3168 my $ref = \%trie;
3169 DIR:
3170 foreach my $dir (split('/', $pr->{'path'})) {
3171 if (exists $ref->{''}) {
3172 # found [shortest] prefix, is a fork - skip it
3173 push @{$ref->{''}{'forks'}}, $pr;
3174 next PROJECT;
3176 if (!exists $ref->{$dir}) {
3177 # not in trie, cannot have prefix, not a fork
3178 push @filtered, $pr;
3179 next PROJECT;
3181 # If the dir is there, we just walk one step down the trie.
3182 $ref = $ref->{$dir};
3184 # we ran out of trie
3185 # (shouldn't happen: it's either no match, or end marker)
3186 push @filtered, $pr;
3189 return @filtered;
3192 # note: fill_project_list_info must be run first,
3193 # for 'descr_long' and 'ctags' to be filled
3194 sub search_projects_list {
3195 my ($projlist, %opts) = @_;
3196 my $tagfilter = $opts{'tagfilter'};
3197 my $search_re = $opts{'search_regexp'};
3199 return @$projlist
3200 unless ($tagfilter || $search_re);
3202 # searching projects require filling to be run before it;
3203 fill_project_list_info($projlist,
3204 $tagfilter ? 'ctags' : (),
3205 $search_re ? ('path', 'descr') : ());
3206 my @projects;
3207 PROJECT:
3208 foreach my $pr (@$projlist) {
3210 if ($tagfilter) {
3211 next unless ref($pr->{'ctags'}) eq 'HASH';
3212 next unless
3213 grep { lc($_) eq lc($tagfilter) } keys %{$pr->{'ctags'}};
3216 if ($search_re) {
3217 next unless
3218 $pr->{'path'} =~ /$search_re/ ||
3219 $pr->{'descr_long'} =~ /$search_re/;
3222 push @projects, $pr;
3225 return @projects;
3228 our $gitweb_project_owner = undef;
3229 sub git_get_project_list_from_file {
3231 return if (defined $gitweb_project_owner);
3233 $gitweb_project_owner = {};
3234 # read from file (url-encoded):
3235 # 'git%2Fgit.git Linus+Torvalds'
3236 # 'libs%2Fklibc%2Fklibc.git H.+Peter+Anvin'
3237 # 'linux%2Fhotplug%2Fudev.git Greg+Kroah-Hartman'
3238 if (-f $projects_list) {
3239 open(my $fd, '<', $projects_list);
3240 while (my $line = <$fd>) {
3241 chomp $line;
3242 my ($pr, $ow) = split ' ', $line;
3243 $pr = unescape($pr);
3244 $ow = unescape($ow);
3245 $gitweb_project_owner->{$pr} = to_utf8($ow);
3247 close $fd;
3251 sub git_get_project_owner {
3252 my $project = shift;
3253 my $owner;
3255 return undef unless $project;
3256 $git_dir = "$projectroot/$project";
3258 if (!defined $gitweb_project_owner) {
3259 git_get_project_list_from_file();
3262 if (exists $gitweb_project_owner->{$project}) {
3263 $owner = $gitweb_project_owner->{$project};
3265 if (!defined $owner){
3266 $owner = git_get_project_config('owner');
3268 if (!defined $owner) {
3269 $owner = get_file_owner("$git_dir");
3272 return $owner;
3275 sub git_get_last_activity {
3276 my ($path) = @_;
3277 my $fd;
3279 $git_dir = "$projectroot/$path";
3280 open($fd, "-|", git_cmd(), 'for-each-ref',
3281 '--format=%(committer)',
3282 '--sort=-committerdate',
3283 '--count=1',
3284 map { "refs/$_" } get_branch_refs ()) or return;
3285 my $most_recent = <$fd>;
3286 close $fd or return;
3287 if (defined $most_recent &&
3288 $most_recent =~ / (\d+) [-+][01]\d\d\d$/) {
3289 my $timestamp = $1;
3290 my $age = time - $timestamp;
3291 return ($age, age_string($age));
3293 return (undef, undef);
3296 # Implementation note: when a single remote is wanted, we cannot use 'git
3297 # remote show -n' because that command always work (assuming it's a remote URL
3298 # if it's not defined), and we cannot use 'git remote show' because that would
3299 # try to make a network roundtrip. So the only way to find if that particular
3300 # remote is defined is to walk the list provided by 'git remote -v' and stop if
3301 # and when we find what we want.
3302 sub git_get_remotes_list {
3303 my $wanted = shift;
3304 my %remotes = ();
3306 open my $fd, '-|' , git_cmd(), 'remote', '-v';
3307 return unless $fd;
3308 while (my $remote = <$fd>) {
3309 chomp $remote;
3310 $remote =~ s!\t(.*?)\s+\((\w+)\)$!!;
3311 next if $wanted and not $remote eq $wanted;
3312 my ($url, $key) = ($1, $2);
3314 $remotes{$remote} ||= { 'heads' => () };
3315 $remotes{$remote}{$key} = $url;
3317 close $fd or return;
3318 return wantarray ? %remotes : \%remotes;
3321 # Takes a hash of remotes as first parameter and fills it by adding the
3322 # available remote heads for each of the indicated remotes.
3323 sub fill_remote_heads {
3324 my $remotes = shift;
3325 my @heads = map { "remotes/$_" } keys %$remotes;
3326 my @remoteheads = git_get_heads_list(undef, @heads);
3327 foreach my $remote (keys %$remotes) {
3328 $remotes->{$remote}{'heads'} = [ grep {
3329 $_->{'name'} =~ s!^$remote/!!
3330 } @remoteheads ];
3334 sub git_get_references {
3335 my $type = shift || "";
3336 my %refs;
3337 # 5dc01c595e6c6ec9ccda4f6f69c131c0dd945f8c refs/tags/v2.6.11
3338 # c39ae07f393806ccf406ef966e9a15afc43cc36a refs/tags/v2.6.11^{}
3339 open my $fd, "-|", git_cmd(), "show-ref", "--dereference",
3340 ($type ? ("--", "refs/$type") : ()) # use -- <pattern> if $type
3341 or return;
3343 while (my $line = <$fd>) {
3344 chomp $line;
3345 if ($line =~ m!^([0-9a-fA-F]{40})\srefs/($type.*)$!) {
3346 if (defined $refs{$1}) {
3347 push @{$refs{$1}}, $2;
3348 } else {
3349 $refs{$1} = [ $2 ];
3353 close $fd or return;
3354 return \%refs;
3357 sub git_get_rev_name_tags {
3358 my $hash = shift || return undef;
3360 open my $fd, "-|", git_cmd(), "name-rev", "--tags", $hash
3361 or return;
3362 my $name_rev = <$fd>;
3363 close $fd;
3365 if ($name_rev =~ m|^$hash tags/(.*)$|) {
3366 return $1;
3367 } else {
3368 # catches also '$hash undefined' output
3369 return undef;
3373 ## ----------------------------------------------------------------------
3374 ## parse to hash functions
3376 sub parse_date {
3377 my $epoch = shift;
3378 my $tz = shift || "-0000";
3380 my %date;
3381 my @months = ("Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov", "Dec");
3382 my @days = ("Sun", "Mon", "Tue", "Wed", "Thu", "Fri", "Sat");
3383 my ($sec, $min, $hour, $mday, $mon, $year, $wday, $yday) = gmtime($epoch);
3384 $date{'hour'} = $hour;
3385 $date{'minute'} = $min;
3386 $date{'mday'} = $mday;
3387 $date{'day'} = $days[$wday];
3388 $date{'month'} = $months[$mon];
3389 $date{'rfc2822'} = sprintf "%s, %d %s %4d %02d:%02d:%02d +0000",
3390 $days[$wday], $mday, $months[$mon], 1900+$year, $hour ,$min, $sec;
3391 $date{'mday-time'} = sprintf "%d %s %02d:%02d",
3392 $mday, $months[$mon], $hour ,$min;
3393 $date{'iso-8601'} = sprintf "%04d-%02d-%02dT%02d:%02d:%02dZ",
3394 1900+$year, 1+$mon, $mday, $hour ,$min, $sec;
3396 my ($tz_sign, $tz_hour, $tz_min) =
3397 ($tz =~ m/^([-+])(\d\d)(\d\d)$/);
3398 $tz_sign = ($tz_sign eq '-' ? -1 : +1);
3399 my $local = $epoch + $tz_sign*((($tz_hour*60) + $tz_min)*60);
3400 ($sec, $min, $hour, $mday, $mon, $year, $wday, $yday) = gmtime($local);
3401 $date{'hour_local'} = $hour;
3402 $date{'minute_local'} = $min;
3403 $date{'tz_local'} = $tz;
3404 $date{'iso-tz'} = sprintf("%04d-%02d-%02d %02d:%02d:%02d %s",
3405 1900+$year, $mon+1, $mday,
3406 $hour, $min, $sec, $tz);
3407 return %date;
3410 sub parse_tag {
3411 my $tag_id = shift;
3412 my %tag;
3413 my @comment;
3415 open my $fd, "-|", git_cmd(), "cat-file", "tag", $tag_id or return;
3416 $tag{'id'} = $tag_id;
3417 while (my $line = <$fd>) {
3418 chomp $line;
3419 if ($line =~ m/^object ([0-9a-fA-F]{40})$/) {
3420 $tag{'object'} = $1;
3421 } elsif ($line =~ m/^type (.+)$/) {
3422 $tag{'type'} = $1;
3423 } elsif ($line =~ m/^tag (.+)$/) {
3424 $tag{'name'} = $1;
3425 } elsif ($line =~ m/^tagger (.*) ([0-9]+) (.*)$/) {
3426 $tag{'author'} = $1;
3427 $tag{'author_epoch'} = $2;
3428 $tag{'author_tz'} = $3;
3429 if ($tag{'author'} =~ m/^([^<]+) <([^>]*)>/) {
3430 $tag{'author_name'} = $1;
3431 $tag{'author_email'} = $2;
3432 } else {
3433 $tag{'author_name'} = $tag{'author'};
3435 } elsif ($line =~ m/--BEGIN/) {
3436 push @comment, $line;
3437 last;
3438 } elsif ($line eq "") {
3439 last;
3442 push @comment, <$fd>;
3443 $tag{'comment'} = \@comment;
3444 close $fd or return;
3445 if (!defined $tag{'name'}) {
3446 return
3448 return %tag
3451 sub parse_commit_text {
3452 my ($commit_text, $withparents) = @_;
3453 my @commit_lines = split '\n', $commit_text;
3454 my %co;
3456 pop @commit_lines; # Remove '\0'
3458 if (! @commit_lines) {
3459 return;
3462 my $header = shift @commit_lines;
3463 if ($header !~ m/^[0-9a-fA-F]{40}/) {
3464 return;
3466 ($co{'id'}, my @parents) = split ' ', $header;
3467 while (my $line = shift @commit_lines) {
3468 last if $line eq "\n";
3469 if ($line =~ m/^tree ([0-9a-fA-F]{40})$/) {
3470 $co{'tree'} = $1;
3471 } elsif ((!defined $withparents) && ($line =~ m/^parent ([0-9a-fA-F]{40})$/)) {
3472 push @parents, $1;
3473 } elsif ($line =~ m/^author (.*) ([0-9]+) (.*)$/) {
3474 $co{'author'} = to_utf8($1);
3475 $co{'author_epoch'} = $2;
3476 $co{'author_tz'} = $3;
3477 if ($co{'author'} =~ m/^([^<]+) <([^>]*)>/) {
3478 $co{'author_name'} = $1;
3479 $co{'author_email'} = $2;
3480 } else {
3481 $co{'author_name'} = $co{'author'};
3483 } elsif ($line =~ m/^committer (.*) ([0-9]+) (.*)$/) {
3484 $co{'committer'} = to_utf8($1);
3485 $co{'committer_epoch'} = $2;
3486 $co{'committer_tz'} = $3;
3487 if ($co{'committer'} =~ m/^([^<]+) <([^>]*)>/) {
3488 $co{'committer_name'} = $1;
3489 $co{'committer_email'} = $2;
3490 } else {
3491 $co{'committer_name'} = $co{'committer'};
3495 if (!defined $co{'tree'}) {
3496 return;
3498 $co{'parents'} = \@parents;
3499 $co{'parent'} = $parents[0];
3501 foreach my $title (@commit_lines) {
3502 $title =~ s/^ //;
3503 if ($title ne "") {
3504 $co{'title'} = chop_str($title, 80, 5);
3505 # remove leading stuff of merges to make the interesting part visible
3506 if (length($title) > 50) {
3507 $title =~ s/^Automatic //;
3508 $title =~ s/^merge (of|with) /Merge ... /i;
3509 if (length($title) > 50) {
3510 $title =~ s/(http|rsync):\/\///;
3512 if (length($title) > 50) {
3513 $title =~ s/(master|www|rsync)\.//;
3515 if (length($title) > 50) {
3516 $title =~ s/kernel.org:?//;
3518 if (length($title) > 50) {
3519 $title =~ s/\/pub\/scm//;
3522 $co{'title_short'} = chop_str($title, 50, 5);
3523 last;
3526 if (! defined $co{'title'} || $co{'title'} eq "") {
3527 $co{'title'} = $co{'title_short'} = '(no commit message)';
3529 # remove added spaces
3530 foreach my $line (@commit_lines) {
3531 $line =~ s/^ //;
3533 $co{'comment'} = \@commit_lines;
3535 my $age = time - $co{'committer_epoch'};
3536 $co{'age'} = $age;
3537 $co{'age_string'} = age_string($age);
3538 my ($sec, $min, $hour, $mday, $mon, $year, $wday, $yday) = gmtime($co{'committer_epoch'});
3539 if ($age > 60*60*24*7*2) {
3540 $co{'age_string_date'} = sprintf "%4i-%02u-%02i", 1900 + $year, $mon+1, $mday;
3541 $co{'age_string_age'} = $co{'age_string'};
3542 } else {
3543 $co{'age_string_date'} = $co{'age_string'};
3544 $co{'age_string_age'} = sprintf "%4i-%02u-%02i", 1900 + $year, $mon+1, $mday;
3546 return %co;
3549 sub parse_commit {
3550 my ($commit_id) = @_;
3551 my %co;
3553 local $/ = "\0";
3555 open my $fd, "-|", git_cmd(), "rev-list",
3556 "--parents",
3557 "--header",
3558 "--max-count=1",
3559 $commit_id,
3560 "--",
3561 or die_error(500, "Open git-rev-list failed");
3562 %co = parse_commit_text(<$fd>, 1);
3563 close $fd;
3565 return %co;
3568 sub parse_commits {
3569 my ($commit_id, $maxcount, $skip, $filename, @args) = @_;
3570 my @cos;
3572 $maxcount ||= 1;
3573 $skip ||= 0;
3575 local $/ = "\0";
3577 open my $fd, "-|", git_cmd(), "rev-list",
3578 "--header",
3579 @args,
3580 ("--max-count=" . $maxcount),
3581 ("--skip=" . $skip),
3582 @extra_options,
3583 $commit_id,
3584 "--",
3585 ($filename ? ($filename) : ())
3586 or die_error(500, "Open git-rev-list failed");
3587 while (my $line = <$fd>) {
3588 my %co = parse_commit_text($line);
3589 push @cos, \%co;
3591 close $fd;
3593 return wantarray ? @cos : \@cos;
3596 # parse line of git-diff-tree "raw" output
3597 sub parse_difftree_raw_line {
3598 my $line = shift;
3599 my %res;
3601 # ':100644 100644 03b218260e99b78c6df0ed378e59ed9205ccc96d 3b93d5e7cc7f7dd4ebed13a5cc1a4ad976fc94d8 M ls-files.c'
3602 # ':100644 100644 7f9281985086971d3877aca27704f2aaf9c448ce bc190ebc71bbd923f2b728e505408f5e54bd073a M rev-tree.c'
3603 if ($line =~ m/^:([0-7]{6}) ([0-7]{6}) ([0-9a-fA-F]{40}) ([0-9a-fA-F]{40}) (.)([0-9]{0,3})\t(.*)$/) {
3604 $res{'from_mode'} = $1;
3605 $res{'to_mode'} = $2;
3606 $res{'from_id'} = $3;
3607 $res{'to_id'} = $4;
3608 $res{'status'} = $5;
3609 $res{'similarity'} = $6;
3610 if ($res{'status'} eq 'R' || $res{'status'} eq 'C') { # renamed or copied
3611 ($res{'from_file'}, $res{'to_file'}) = map { unquote($_) } split("\t", $7);
3612 } else {
3613 $res{'from_file'} = $res{'to_file'} = $res{'file'} = unquote($7);
3616 # '::100755 100755 100755 60e79ca1b01bc8b057abe17ddab484699a7f5fdb 94067cc5f73388f33722d52ae02f44692bc07490 94067cc5f73388f33722d52ae02f44692bc07490 MR git-gui/git-gui.sh'
3617 # combined diff (for merge commit)
3618 elsif ($line =~ s/^(::+)((?:[0-7]{6} )+)((?:[0-9a-fA-F]{40} )+)([a-zA-Z]+)\t(.*)$//) {
3619 $res{'nparents'} = length($1);
3620 $res{'from_mode'} = [ split(' ', $2) ];
3621 $res{'to_mode'} = pop @{$res{'from_mode'}};
3622 $res{'from_id'} = [ split(' ', $3) ];
3623 $res{'to_id'} = pop @{$res{'from_id'}};
3624 $res{'status'} = [ split('', $4) ];
3625 $res{'to_file'} = unquote($5);
3627 # 'c512b523472485aef4fff9e57b229d9d243c967f'
3628 elsif ($line =~ m/^([0-9a-fA-F]{40})$/) {
3629 $res{'commit'} = $1;
3632 return wantarray ? %res : \%res;
3635 # wrapper: return parsed line of git-diff-tree "raw" output
3636 # (the argument might be raw line, or parsed info)
3637 sub parsed_difftree_line {
3638 my $line_or_ref = shift;
3640 if (ref($line_or_ref) eq "HASH") {
3641 # pre-parsed (or generated by hand)
3642 return $line_or_ref;
3643 } else {
3644 return parse_difftree_raw_line($line_or_ref);
3648 # parse line of git-ls-tree output
3649 sub parse_ls_tree_line {
3650 my $line = shift;
3651 my %opts = @_;
3652 my %res;
3654 if ($opts{'-l'}) {
3655 #'100644 blob 0fa3f3a66fb6a137f6ec2c19351ed4d807070ffa 16717 panic.c'
3656 $line =~ m/^([0-9]+) (.+) ([0-9a-fA-F]{40}) +(-|[0-9]+)\t(.+)$/s;
3658 $res{'mode'} = $1;
3659 $res{'type'} = $2;
3660 $res{'hash'} = $3;
3661 $res{'size'} = $4;
3662 if ($opts{'-z'}) {
3663 $res{'name'} = $5;
3664 } else {
3665 $res{'name'} = unquote($5);
3667 } else {
3668 #'100644 blob 0fa3f3a66fb6a137f6ec2c19351ed4d807070ffa panic.c'
3669 $line =~ m/^([0-9]+) (.+) ([0-9a-fA-F]{40})\t(.+)$/s;
3671 $res{'mode'} = $1;
3672 $res{'type'} = $2;
3673 $res{'hash'} = $3;
3674 if ($opts{'-z'}) {
3675 $res{'name'} = $4;
3676 } else {
3677 $res{'name'} = unquote($4);
3681 return wantarray ? %res : \%res;
3684 # generates _two_ hashes, references to which are passed as 2 and 3 argument
3685 sub parse_from_to_diffinfo {
3686 my ($diffinfo, $from, $to, @parents) = @_;
3688 if ($diffinfo->{'nparents'}) {
3689 # combined diff
3690 $from->{'file'} = [];
3691 $from->{'href'} = [];
3692 fill_from_file_info($diffinfo, @parents)
3693 unless exists $diffinfo->{'from_file'};
3694 for (my $i = 0; $i < $diffinfo->{'nparents'}; $i++) {
3695 $from->{'file'}[$i] =
3696 defined $diffinfo->{'from_file'}[$i] ?
3697 $diffinfo->{'from_file'}[$i] :
3698 $diffinfo->{'to_file'};
3699 if ($diffinfo->{'status'}[$i] ne "A") { # not new (added) file
3700 $from->{'href'}[$i] = href(action=>"blob",
3701 hash_base=>$parents[$i],
3702 hash=>$diffinfo->{'from_id'}[$i],
3703 file_name=>$from->{'file'}[$i]);
3704 } else {
3705 $from->{'href'}[$i] = undef;
3708 } else {
3709 # ordinary (not combined) diff
3710 $from->{'file'} = $diffinfo->{'from_file'};
3711 if ($diffinfo->{'status'} ne "A") { # not new (added) file
3712 $from->{'href'} = href(action=>"blob", hash_base=>$hash_parent,
3713 hash=>$diffinfo->{'from_id'},
3714 file_name=>$from->{'file'});
3715 } else {
3716 delete $from->{'href'};
3720 $to->{'file'} = $diffinfo->{'to_file'};
3721 if (!is_deleted($diffinfo)) { # file exists in result
3722 $to->{'href'} = href(action=>"blob", hash_base=>$hash,
3723 hash=>$diffinfo->{'to_id'},
3724 file_name=>$to->{'file'});
3725 } else {
3726 delete $to->{'href'};
3730 ## ......................................................................
3731 ## parse to array of hashes functions
3733 sub git_get_heads_list {
3734 my ($limit, @classes) = @_;
3735 @classes = get_branch_refs() unless @classes;
3736 my @patterns = map { "refs/$_" } @classes;
3737 my @headslist;
3739 open my $fd, '-|', git_cmd(), 'for-each-ref',
3740 ($limit ? '--count='.($limit+1) : ()), '--sort=-committerdate',
3741 '--format=%(objectname) %(refname) %(subject)%00%(committer)',
3742 @patterns
3743 or return;
3744 while (my $line = <$fd>) {
3745 my %ref_item;
3747 chomp $line;
3748 my ($refinfo, $committerinfo) = split(/\0/, $line);
3749 my ($hash, $name, $title) = split(' ', $refinfo, 3);
3750 my ($committer, $epoch, $tz) =
3751 ($committerinfo =~ /^(.*) ([0-9]+) (.*)$/);
3752 $ref_item{'fullname'} = $name;
3753 my $strip_refs = join '|', map { quotemeta } get_branch_refs();
3754 $name =~ s!^refs/($strip_refs|remotes)/!!;
3755 $ref_item{'name'} = $name;
3756 # for refs neither in 'heads' nor 'remotes' we want to
3757 # show their ref dir
3758 my $ref_dir = (defined $1) ? $1 : '';
3759 if ($ref_dir ne '' and $ref_dir ne 'heads' and $ref_dir ne 'remotes') {
3760 $ref_item{'name'} .= ' (' . $ref_dir . ')';
3763 $ref_item{'id'} = $hash;
3764 $ref_item{'title'} = $title || '(no commit message)';
3765 $ref_item{'epoch'} = $epoch;
3766 if ($epoch) {
3767 $ref_item{'age'} = age_string(time - $ref_item{'epoch'});
3768 } else {
3769 $ref_item{'age'} = "unknown";
3772 push @headslist, \%ref_item;
3774 close $fd;
3776 return wantarray ? @headslist : \@headslist;
3779 sub git_get_tags_list {
3780 my $limit = shift;
3781 my @tagslist;
3783 open my $fd, '-|', git_cmd(), 'for-each-ref',
3784 ($limit ? '--count='.($limit+1) : ()), '--sort=-creatordate',
3785 '--format=%(objectname) %(objecttype) %(refname) '.
3786 '%(*objectname) %(*objecttype) %(subject)%00%(creator)',
3787 'refs/tags'
3788 or return;
3789 while (my $line = <$fd>) {
3790 my %ref_item;
3792 chomp $line;
3793 my ($refinfo, $creatorinfo) = split(/\0/, $line);
3794 my ($id, $type, $name, $refid, $reftype, $title) = split(' ', $refinfo, 6);
3795 my ($creator, $epoch, $tz) =
3796 ($creatorinfo =~ /^(.*) ([0-9]+) (.*)$/);
3797 $ref_item{'fullname'} = $name;
3798 $name =~ s!^refs/tags/!!;
3800 $ref_item{'type'} = $type;
3801 $ref_item{'id'} = $id;
3802 $ref_item{'name'} = $name;
3803 if ($type eq "tag") {
3804 $ref_item{'subject'} = $title;
3805 $ref_item{'reftype'} = $reftype;
3806 $ref_item{'refid'} = $refid;
3807 } else {
3808 $ref_item{'reftype'} = $type;
3809 $ref_item{'refid'} = $id;
3812 if ($type eq "tag" || $type eq "commit") {
3813 $ref_item{'epoch'} = $epoch;
3814 if ($epoch) {
3815 $ref_item{'age'} = age_string(time - $ref_item{'epoch'});
3816 } else {
3817 $ref_item{'age'} = "unknown";
3821 push @tagslist, \%ref_item;
3823 close $fd;
3825 return wantarray ? @tagslist : \@tagslist;
3828 ## ----------------------------------------------------------------------
3829 ## filesystem-related functions
3831 sub get_file_owner {
3832 my $path = shift;
3834 my ($dev, $ino, $mode, $nlink, $st_uid, $st_gid, $rdev, $size) = stat($path);
3835 my ($name, $passwd, $uid, $gid, $quota, $comment, $gcos, $dir, $shell) = getpwuid($st_uid);
3836 if (!defined $gcos) {
3837 return undef;
3839 my $owner = $gcos;
3840 $owner =~ s/[,;].*$//;
3841 return to_utf8($owner);
3844 # assume that file exists
3845 sub insert_file {
3846 my $filename = shift;
3848 open my $fd, '<', $filename;
3849 print map { to_utf8($_) } <$fd>;
3850 close $fd;
3853 ## ......................................................................
3854 ## mimetype related functions
3856 sub mimetype_guess_file {
3857 my $filename = shift;
3858 my $mimemap = shift;
3859 -r $mimemap or return undef;
3861 my %mimemap;
3862 open(my $mh, '<', $mimemap) or return undef;
3863 while (<$mh>) {
3864 next if m/^#/; # skip comments
3865 my ($mimetype, @exts) = split(/\s+/);
3866 foreach my $ext (@exts) {
3867 $mimemap{$ext} = $mimetype;
3870 close($mh);
3872 $filename =~ /\.([^.]*)$/;
3873 return $mimemap{$1};
3876 sub mimetype_guess {
3877 my $filename = shift;
3878 my $mime;
3879 $filename =~ /\./ or return undef;
3881 if ($mimetypes_file) {
3882 my $file = $mimetypes_file;
3883 if ($file !~ m!^/!) { # if it is relative path
3884 # it is relative to project
3885 $file = "$projectroot/$project/$file";
3887 $mime = mimetype_guess_file($filename, $file);
3889 $mime ||= mimetype_guess_file($filename, '/etc/mime.types');
3890 return $mime;
3893 sub blob_mimetype {
3894 my $fd = shift;
3895 my $filename = shift;
3897 if ($filename) {
3898 my $mime = mimetype_guess($filename);
3899 $mime and return $mime;
3902 # just in case
3903 return $default_blob_plain_mimetype unless $fd;
3905 if (-T $fd) {
3906 return 'text/plain';
3907 } elsif (! $filename) {
3908 return 'application/octet-stream';
3909 } elsif ($filename =~ m/\.png$/i) {
3910 return 'image/png';
3911 } elsif ($filename =~ m/\.gif$/i) {
3912 return 'image/gif';
3913 } elsif ($filename =~ m/\.jpe?g$/i) {
3914 return 'image/jpeg';
3915 } else {
3916 return 'application/octet-stream';
3920 sub blob_contenttype {
3921 my ($fd, $file_name, $type) = @_;
3923 $type ||= blob_mimetype($fd, $file_name);
3924 if ($type eq 'text/plain' && defined $default_text_plain_charset) {
3925 $type .= "; charset=$default_text_plain_charset";
3928 return $type;
3931 # guess file syntax for syntax highlighting; return undef if no highlighting
3932 # the name of syntax can (in the future) depend on syntax highlighter used
3933 sub guess_file_syntax {
3934 my ($highlight, $file_name) = @_;
3935 return undef unless ($highlight && defined $file_name);
3936 my $basename = basename($file_name, '.in');
3937 return $highlight_basename{$basename}
3938 if exists $highlight_basename{$basename};
3940 $basename =~ /\.([^.]*)$/;
3941 my $ext = $1 or return undef;
3942 return $highlight_ext{$ext}
3943 if exists $highlight_ext{$ext};
3945 return undef;
3948 # run highlighter and return FD of its output,
3949 # or return original FD if no highlighting
3950 sub run_highlighter {
3951 my ($fd, $highlight, $syntax) = @_;
3952 return $fd unless ($highlight);
3954 close $fd;
3955 my $syntax_arg = (defined $syntax) ? "--syntax $syntax" : "--force";
3956 open $fd, quote_command(git_cmd(), "cat-file", "blob", $hash)." | ".
3957 quote_command($^X, '-CO', '-MEncode=decode,FB_DEFAULT', '-pse',
3958 '$_ = decode($fe, $_, FB_DEFAULT) if !utf8::decode($_);',
3959 '--', "-fe=$fallback_encoding")." | ".
3960 quote_command($highlight_bin).
3961 " --replace-tabs=8 --fragment $syntax_arg |"
3962 or die_error(500, "Couldn't open file or run syntax highlighter");
3963 return $fd;
3966 ## ======================================================================
3967 ## functions printing HTML: header, footer, error page
3969 sub get_page_title {
3970 my $title = to_utf8($site_name);
3972 unless (defined $project) {
3973 if (defined $project_filter) {
3974 $title .= " - projects in '" . esc_path($project_filter) . "'";
3976 return $title;
3978 $title .= " - " . to_utf8($project);
3980 return $title unless (defined $action);
3981 $title .= "/$action"; # $action is US-ASCII (7bit ASCII)
3983 return $title unless (defined $file_name);
3984 $title .= " - " . esc_path($file_name);
3985 if ($action eq "tree" && $file_name !~ m|/$|) {
3986 $title .= "/";
3989 return $title;
3992 sub get_content_type_html {
3993 # require explicit support from the UA if we are to send the page as
3994 # 'application/xhtml+xml', otherwise send it as plain old 'text/html'.
3995 # we have to do this because MSIE sometimes globs '*/*', pretending to
3996 # support xhtml+xml but choking when it gets what it asked for.
3997 if (defined $cgi->http('HTTP_ACCEPT') &&
3998 $cgi->http('HTTP_ACCEPT') =~ m/(,|;|\s|^)application\/xhtml\+xml(,|;|\s|$)/ &&
3999 $cgi->Accept('application/xhtml+xml') != 0) {
4000 return 'application/xhtml+xml';
4001 } else {
4002 return 'text/html';
4006 sub print_feed_meta {
4007 if (defined $project) {
4008 my %href_params = get_feed_info();
4009 if (!exists $href_params{'-title'}) {
4010 $href_params{'-title'} = 'log';
4013 foreach my $format (qw(RSS Atom)) {
4014 my $type = lc($format);
4015 my %link_attr = (
4016 '-rel' => 'alternate',
4017 '-title' => esc_attr("$project - $href_params{'-title'} - $format feed"),
4018 '-type' => "application/$type+xml"
4021 $href_params{'extra_options'} = undef;
4022 $href_params{'action'} = $type;
4023 $link_attr{'-href'} = href(%href_params);
4024 print "<link ".
4025 "rel=\"$link_attr{'-rel'}\" ".
4026 "title=\"$link_attr{'-title'}\" ".
4027 "href=\"$link_attr{'-href'}\" ".
4028 "type=\"$link_attr{'-type'}\" ".
4029 "/>\n";
4031 $href_params{'extra_options'} = '--no-merges';
4032 $link_attr{'-href'} = href(%href_params);
4033 $link_attr{'-title'} .= ' (no merges)';
4034 print "<link ".
4035 "rel=\"$link_attr{'-rel'}\" ".
4036 "title=\"$link_attr{'-title'}\" ".
4037 "href=\"$link_attr{'-href'}\" ".
4038 "type=\"$link_attr{'-type'}\" ".
4039 "/>\n";
4042 } else {
4043 printf('<link rel="alternate" title="%s projects list" '.
4044 'href="%s" type="text/plain; charset=utf-8" />'."\n",
4045 esc_attr($site_name), href(project=>undef, action=>"project_index"));
4046 printf('<link rel="alternate" title="%s projects feeds" '.
4047 'href="%s" type="text/x-opml" />'."\n",
4048 esc_attr($site_name), href(project=>undef, action=>"opml"));
4052 sub print_header_links {
4053 my $status = shift;
4055 # print out each stylesheet that exist, providing backwards capability
4056 # for those people who defined $stylesheet in a config file
4057 if (defined $stylesheet) {
4058 print '<link rel="stylesheet" type="text/css" href="'.esc_url($stylesheet).'"/>'."\n";
4059 } else {
4060 foreach my $stylesheet (@stylesheets) {
4061 next unless $stylesheet;
4062 print '<link rel="stylesheet" type="text/css" href="'.esc_url($stylesheet).'"/>'."\n";
4065 print_feed_meta()
4066 if ($status eq '200 OK');
4067 if (defined $favicon) {
4068 print qq(<link rel="shortcut icon" href=").esc_url($favicon).qq(" type="image/png" />\n);
4072 sub print_nav_breadcrumbs_path {
4073 my $dirprefix = undef;
4074 while (my $part = shift) {
4075 $dirprefix .= "/" if defined $dirprefix;
4076 $dirprefix .= $part;
4077 print $cgi->a({-href => href(project => undef,
4078 project_filter => $dirprefix,
4079 action => "project_list")},
4080 esc_html($part)) . " / ";
4084 sub print_nav_breadcrumbs {
4085 my %opts = @_;
4087 for my $crumb (@extra_breadcrumbs, [ $home_link_str => $home_link ]) {
4088 print $cgi->a({-href => esc_url($crumb->[1])}, $crumb->[0]) . " / ";
4090 if (defined $project) {
4091 my @dirname = split '/', $project;
4092 my $projectbasename = pop @dirname;
4093 print_nav_breadcrumbs_path(@dirname);
4094 print $cgi->a({-href => href(action=>"summary")}, esc_html($projectbasename));
4095 if (defined $action) {
4096 my $action_print = $action ;
4097 if (defined $opts{-action_extra}) {
4098 $action_print = $cgi->a({-href => href(action=>$action)},
4099 $action);
4101 print " / $action_print";
4103 if (defined $opts{-action_extra}) {
4104 print " / $opts{-action_extra}";
4106 print "\n";
4107 } elsif (defined $project_filter) {
4108 print_nav_breadcrumbs_path(split '/', $project_filter);
4112 sub print_search_form {
4113 if (!defined $searchtext) {
4114 $searchtext = "";
4116 my $search_hash;
4117 if (defined $hash_base) {
4118 $search_hash = $hash_base;
4119 } elsif (defined $hash) {
4120 $search_hash = $hash;
4121 } else {
4122 $search_hash = "HEAD";
4124 my $action = $my_uri;
4125 my $use_pathinfo = gitweb_check_feature('pathinfo');
4126 if ($use_pathinfo) {
4127 $action .= "/".esc_url($project);
4129 print $cgi->start_form(-method => "get", -action => $action) .
4130 "<div class=\"search\">\n" .
4131 (!$use_pathinfo &&
4132 $cgi->input({-name=>"p", -value=>$project, -type=>"hidden"}) . "\n") .
4133 $cgi->input({-name=>"a", -value=>"search", -type=>"hidden"}) . "\n" .
4134 $cgi->input({-name=>"h", -value=>$search_hash, -type=>"hidden"}) . "\n" .
4135 $cgi->popup_menu(-name => 'st', -default => 'commit',
4136 -values => ['commit', 'grep', 'author', 'committer', 'pickaxe']) .
4137 " " . $cgi->a({-href => href(action=>"search_help"),
4138 -title => "search help" }, "?") . " search:\n",
4139 $cgi->textfield(-name => "s", -value => $searchtext, -override => 1) . "\n" .
4140 "<span title=\"Extended regular expression\">" .
4141 $cgi->checkbox(-name => 'sr', -value => 1, -label => 're',
4142 -checked => $search_use_regexp) .
4143 "</span>" .
4144 "</div>" .
4145 $cgi->end_form() . "\n";
4148 sub git_header_html {
4149 my $status = shift || "200 OK";
4150 my $expires = shift;
4151 my %opts = @_;
4153 my $title = get_page_title();
4154 my $content_type = get_content_type_html();
4155 print $cgi->header(-type=>$content_type, -charset => 'utf-8',
4156 -status=> $status, -expires => $expires)
4157 unless ($opts{'-no_http_header'});
4158 my $mod_perl_version = $ENV{'MOD_PERL'} ? " $ENV{'MOD_PERL'}" : '';
4159 print <<EOF;
4160 <?xml version="1.0" encoding="utf-8"?>
4161 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
4162 <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en-US" lang="en-US">
4163 <!-- git web interface version $version, (C) 2005-2006, Kay Sievers <kay.sievers\@vrfy.org>, Christian Gierke -->
4164 <!-- git core binaries version $git_version -->
4165 <head>
4166 <meta http-equiv="content-type" content="$content_type; charset=utf-8"/>
4167 <meta name="generator" content="gitweb/$version git/$git_version$mod_perl_version"/>
4168 <meta name="robots" content="index, nofollow"/>
4169 <title>$title</title>
4171 # the stylesheet, favicon etc urls won't work correctly with path_info
4172 # unless we set the appropriate base URL
4173 if ($ENV{'PATH_INFO'}) {
4174 print "<base href=\"".esc_url($base_url)."\" />\n";
4176 print_header_links($status);
4178 if (defined $site_html_head_string) {
4179 print to_utf8($site_html_head_string);
4182 print "</head>\n" .
4183 "<body>\n";
4185 if (defined $site_header && -f $site_header) {
4186 insert_file($site_header);
4189 print "<div class=\"page_header\">\n";
4190 if (defined $logo) {
4191 print $cgi->a({-href => esc_url($logo_url),
4192 -title => $logo_label},
4193 $cgi->img({-src => esc_url($logo),
4194 -width => 72, -height => 27,
4195 -alt => "git",
4196 -class => "logo"}));
4198 print_nav_breadcrumbs(%opts);
4199 print "</div>\n";
4201 my $have_search = gitweb_check_feature('search');
4202 if (defined $project && $have_search) {
4203 print_search_form();
4207 sub git_footer_html {
4208 my $feed_class = 'rss_logo';
4210 print "<div class=\"page_footer\">\n";
4211 if (defined $project) {
4212 my $descr = git_get_project_description($project);
4213 if (defined $descr) {
4214 print "<div class=\"page_footer_text\">" . esc_html($descr) . "</div>\n";
4217 my %href_params = get_feed_info();
4218 if (!%href_params) {
4219 $feed_class .= ' generic';
4221 $href_params{'-title'} ||= 'log';
4223 foreach my $format (qw(RSS Atom)) {
4224 $href_params{'action'} = lc($format);
4225 print $cgi->a({-href => href(%href_params),
4226 -title => "$href_params{'-title'} $format feed",
4227 -class => $feed_class}, $format)."\n";
4230 } else {
4231 print $cgi->a({-href => href(project=>undef, action=>"opml",
4232 project_filter => $project_filter),
4233 -class => $feed_class}, "OPML") . " ";
4234 print $cgi->a({-href => href(project=>undef, action=>"project_index",
4235 project_filter => $project_filter),
4236 -class => $feed_class}, "TXT") . "\n";
4238 print "</div>\n"; # class="page_footer"
4240 if (defined $t0 && gitweb_check_feature('timed')) {
4241 print "<div id=\"generating_info\">\n";
4242 print 'This page took '.
4243 '<span id="generating_time" class="time_span">'.
4244 tv_interval($t0, [ gettimeofday() ]).
4245 ' seconds </span>'.
4246 ' and '.
4247 '<span id="generating_cmd">'.
4248 $number_of_git_cmds.
4249 '</span> git commands '.
4250 " to generate.\n";
4251 print "</div>\n"; # class="page_footer"
4254 if (defined $site_footer && -f $site_footer) {
4255 insert_file($site_footer);
4258 print qq!<script type="text/javascript" src="!.esc_url($javascript).qq!"></script>\n!;
4259 if (defined $action &&
4260 $action eq 'blame_incremental') {
4261 print qq!<script type="text/javascript">\n!.
4262 qq!startBlame("!. href(action=>"blame_data", -replay=>1) .qq!",\n!.
4263 qq! "!. href() .qq!");\n!.
4264 qq!</script>\n!;
4265 } else {
4266 my ($jstimezone, $tz_cookie, $datetime_class) =
4267 gitweb_get_feature('javascript-timezone');
4269 print qq!<script type="text/javascript">\n!.
4270 qq!window.onload = function () {\n!;
4271 if (gitweb_check_feature('javascript-actions')) {
4272 print qq! fixLinks();\n!;
4274 if ($jstimezone && $tz_cookie && $datetime_class) {
4275 print qq! var tz_cookie = { name: '$tz_cookie', expires: 14, path: '/' };\n!. # in days
4276 qq! onloadTZSetup('$jstimezone', tz_cookie, '$datetime_class');\n!;
4278 print qq!};\n!.
4279 qq!</script>\n!;
4282 print "</body>\n" .
4283 "</html>";
4286 # die_error(<http_status_code>, <error_message>[, <detailed_html_description>])
4287 # Example: die_error(404, 'Hash not found')
4288 # By convention, use the following status codes (as defined in RFC 2616):
4289 # 400: Invalid or missing CGI parameters, or
4290 # requested object exists but has wrong type.
4291 # 403: Requested feature (like "pickaxe" or "snapshot") not enabled on
4292 # this server or project.
4293 # 404: Requested object/revision/project doesn't exist.
4294 # 500: The server isn't configured properly, or
4295 # an internal error occurred (e.g. failed assertions caused by bugs), or
4296 # an unknown error occurred (e.g. the git binary died unexpectedly).
4297 # 503: The server is currently unavailable (because it is overloaded,
4298 # or down for maintenance). Generally, this is a temporary state.
4299 sub die_error {
4300 my $status = shift || 500;
4301 my $error = esc_html(shift) || "Internal Server Error";
4302 my $extra = shift;
4303 my %opts = @_;
4305 my %http_responses = (
4306 400 => '400 Bad Request',
4307 403 => '403 Forbidden',
4308 404 => '404 Not Found',
4309 500 => '500 Internal Server Error',
4310 503 => '503 Service Unavailable',
4312 git_header_html($http_responses{$status}, undef, %opts);
4313 print <<EOF;
4314 <div class="page_body">
4315 <br /><br />
4316 $status - $error
4317 <br />
4319 if (defined $extra) {
4320 print "<hr />\n" .
4321 "$extra\n";
4323 print "</div>\n";
4325 git_footer_html();
4326 goto DONE_GITWEB
4327 unless ($opts{'-error_handler'});
4330 ## ----------------------------------------------------------------------
4331 ## functions printing or outputting HTML: navigation
4333 sub git_print_page_nav {
4334 my ($current, $suppress, $head, $treehead, $treebase, $extra) = @_;
4335 $extra = '' if !defined $extra; # pager or formats
4337 my @navs = qw(summary shortlog log commit commitdiff tree);
4338 if ($suppress) {
4339 @navs = grep { $_ ne $suppress } @navs;
4342 my %arg = map { $_ => {action=>$_} } @navs;
4343 if (defined $head) {
4344 for (qw(commit commitdiff)) {
4345 $arg{$_}{'hash'} = $head;
4347 if ($current =~ m/^(tree | log | shortlog | commit | commitdiff | search)$/x) {
4348 for (qw(shortlog log)) {
4349 $arg{$_}{'hash'} = $head;
4354 $arg{'tree'}{'hash'} = $treehead if defined $treehead;
4355 $arg{'tree'}{'hash_base'} = $treebase if defined $treebase;
4357 my @actions = gitweb_get_feature('actions');
4358 my %repl = (
4359 '%' => '%',
4360 'n' => $project, # project name
4361 'f' => $git_dir, # project path within filesystem
4362 'h' => $treehead || '', # current hash ('h' parameter)
4363 'b' => $treebase || '', # hash base ('hb' parameter)
4365 while (@actions) {
4366 my ($label, $link, $pos) = splice(@actions,0,3);
4367 # insert
4368 @navs = map { $_ eq $pos ? ($_, $label) : $_ } @navs;
4369 # munch munch
4370 $link =~ s/%([%nfhb])/$repl{$1}/g;
4371 $arg{$label}{'_href'} = $link;
4374 print "<div class=\"page_nav\">\n" .
4375 (join " | ",
4376 map { $_ eq $current ?
4377 $_ : $cgi->a({-href => ($arg{$_}{_href} ? $arg{$_}{_href} : href(%{$arg{$_}}))}, "$_")
4378 } @navs);
4379 print "<br/>\n$extra<br/>\n" .
4380 "</div>\n";
4383 # returns a submenu for the navigation of the refs views (tags, heads,
4384 # remotes) with the current view disabled and the remotes view only
4385 # available if the feature is enabled
4386 sub format_ref_views {
4387 my ($current) = @_;
4388 my @ref_views = qw{tags heads};
4389 push @ref_views, 'remotes' if gitweb_check_feature('remote_heads');
4390 return join " | ", map {
4391 $_ eq $current ? $_ :
4392 $cgi->a({-href => href(action=>$_)}, $_)
4393 } @ref_views
4396 sub format_paging_nav {
4397 my ($action, $page, $has_next_link) = @_;
4398 my $paging_nav;
4401 if ($page > 0) {
4402 $paging_nav .=
4403 $cgi->a({-href => href(-replay=>1, page=>undef)}, "first") .
4404 " &sdot; " .
4405 $cgi->a({-href => href(-replay=>1, page=>$page-1),
4406 -accesskey => "p", -title => "Alt-p"}, "prev");
4407 } else {
4408 $paging_nav .= "first &sdot; prev";
4411 if ($has_next_link) {
4412 $paging_nav .= " &sdot; " .
4413 $cgi->a({-href => href(-replay=>1, page=>$page+1),
4414 -accesskey => "n", -title => "Alt-n"}, "next");
4415 } else {
4416 $paging_nav .= " &sdot; next";
4419 return $paging_nav;
4422 ## ......................................................................
4423 ## functions printing or outputting HTML: div
4425 sub git_print_header_div {
4426 my ($action, $title, $hash, $hash_base) = @_;
4427 my %args = ();
4429 $args{'action'} = $action;
4430 $args{'hash'} = $hash if $hash;
4431 $args{'hash_base'} = $hash_base if $hash_base;
4433 print "<div class=\"header\">\n" .
4434 $cgi->a({-href => href(%args), -class => "title"},
4435 $title ? $title : $action) .
4436 "\n</div>\n";
4439 sub format_repo_url {
4440 my ($name, $url) = @_;
4441 return "<tr class=\"metadata_url\"><td>$name</td><td>$url</td></tr>\n";
4444 # Group output by placing it in a DIV element and adding a header.
4445 # Options for start_div() can be provided by passing a hash reference as the
4446 # first parameter to the function.
4447 # Options to git_print_header_div() can be provided by passing an array
4448 # reference. This must follow the options to start_div if they are present.
4449 # The content can be a scalar, which is output as-is, a scalar reference, which
4450 # is output after html escaping, an IO handle passed either as *handle or
4451 # *handle{IO}, or a function reference. In the latter case all following
4452 # parameters will be taken as argument to the content function call.
4453 sub git_print_section {
4454 my ($div_args, $header_args, $content);
4455 my $arg = shift;
4456 if (ref($arg) eq 'HASH') {
4457 $div_args = $arg;
4458 $arg = shift;
4460 if (ref($arg) eq 'ARRAY') {
4461 $header_args = $arg;
4462 $arg = shift;
4464 $content = $arg;
4466 print $cgi->start_div($div_args);
4467 git_print_header_div(@$header_args);
4469 if (ref($content) eq 'CODE') {
4470 $content->(@_);
4471 } elsif (ref($content) eq 'SCALAR') {
4472 print esc_html($$content);
4473 } elsif (ref($content) eq 'GLOB' or ref($content) eq 'IO::Handle') {
4474 print <$content>;
4475 } elsif (!ref($content) && defined($content)) {
4476 print $content;
4479 print $cgi->end_div;
4482 sub format_timestamp_html {
4483 my $date = shift;
4484 my $strtime = $date->{'rfc2822'};
4486 my (undef, undef, $datetime_class) =
4487 gitweb_get_feature('javascript-timezone');
4488 if ($datetime_class) {
4489 $strtime = qq!<span class="$datetime_class">$strtime</span>!;
4492 my $localtime_format = '(%02d:%02d %s)';
4493 if ($date->{'hour_local'} < 6) {
4494 $localtime_format = '(<span class="atnight">%02d:%02d</span> %s)';
4496 $strtime .= ' ' .
4497 sprintf($localtime_format,
4498 $date->{'hour_local'}, $date->{'minute_local'}, $date->{'tz_local'});
4500 return $strtime;
4503 # Outputs the author name and date in long form
4504 sub git_print_authorship {
4505 my $co = shift;
4506 my %opts = @_;
4507 my $tag = $opts{-tag} || 'div';
4508 my $author = $co->{'author_name'};
4510 my %ad = parse_date($co->{'author_epoch'}, $co->{'author_tz'});
4511 print "<$tag class=\"author_date\">" .
4512 format_search_author($author, "author", esc_html($author)) .
4513 " [".format_timestamp_html(\%ad)."]".
4514 git_get_avatar($co->{'author_email'}, -pad_before => 1) .
4515 "</$tag>\n";
4518 # Outputs table rows containing the full author or committer information,
4519 # in the format expected for 'commit' view (& similar).
4520 # Parameters are a commit hash reference, followed by the list of people
4521 # to output information for. If the list is empty it defaults to both
4522 # author and committer.
4523 sub git_print_authorship_rows {
4524 my $co = shift;
4525 # too bad we can't use @people = @_ || ('author', 'committer')
4526 my @people = @_;
4527 @people = ('author', 'committer') unless @people;
4528 foreach my $who (@people) {
4529 my %wd = parse_date($co->{"${who}_epoch"}, $co->{"${who}_tz"});
4530 print "<tr><td>$who</td><td>" .
4531 format_search_author($co->{"${who}_name"}, $who,
4532 esc_html($co->{"${who}_name"})) . " " .
4533 format_search_author($co->{"${who}_email"}, $who,
4534 esc_html("<" . $co->{"${who}_email"} . ">")) .
4535 "</td><td rowspan=\"2\">" .
4536 git_get_avatar($co->{"${who}_email"}, -size => 'double') .
4537 "</td></tr>\n" .
4538 "<tr>" .
4539 "<td></td><td>" .
4540 format_timestamp_html(\%wd) .
4541 "</td>" .
4542 "</tr>\n";
4546 sub git_print_page_path {
4547 my $name = shift;
4548 my $type = shift;
4549 my $hb = shift;
4552 print "<div class=\"page_path\">";
4553 print $cgi->a({-href => href(action=>"tree", hash_base=>$hb),
4554 -title => 'tree root'}, to_utf8("[$project]"));
4555 print " / ";
4556 if (defined $name) {
4557 my @dirname = split '/', $name;
4558 my $basename = pop @dirname;
4559 my $fullname = '';
4561 foreach my $dir (@dirname) {
4562 $fullname .= ($fullname ? '/' : '') . $dir;
4563 print $cgi->a({-href => href(action=>"tree", file_name=>$fullname,
4564 hash_base=>$hb),
4565 -title => $fullname}, esc_path($dir));
4566 print " / ";
4568 if (defined $type && $type eq 'blob') {
4569 print $cgi->a({-href => href(action=>"blob_plain", file_name=>$file_name,
4570 hash_base=>$hb),
4571 -title => $name}, esc_path($basename));
4572 } elsif (defined $type && $type eq 'tree') {
4573 print $cgi->a({-href => href(action=>"tree", file_name=>$file_name,
4574 hash_base=>$hb),
4575 -title => $name}, esc_path($basename));
4576 print " / ";
4577 } else {
4578 print esc_path($basename);
4581 print "<br/></div>\n";
4584 sub git_print_log {
4585 my $log = shift;
4586 my %opts = @_;
4588 if ($opts{'-remove_title'}) {
4589 # remove title, i.e. first line of log
4590 shift @$log;
4592 # remove leading empty lines
4593 while (defined $log->[0] && $log->[0] eq "") {
4594 shift @$log;
4597 # print log
4598 my $skip_blank_line = 0;
4599 foreach my $line (@$log) {
4600 if ($line =~ m/^\s*([A-Z][-A-Za-z]*-[Bb]y|C[Cc]): /) {
4601 if (! $opts{'-remove_signoff'}) {
4602 print "<span class=\"signoff\">" . esc_html($line) . "</span><br/>\n";
4603 $skip_blank_line = 1;
4605 next;
4608 if ($line =~ m,\s*([a-z]*link): (https?://\S+),i) {
4609 if (! $opts{'-remove_signoff'}) {
4610 print "<span class=\"signoff\">" . esc_html($1) . ": " .
4611 "<a href=\"" . esc_html($2) . "\">" . esc_html($2) . "</a>" .
4612 "</span><br/>\n";
4613 $skip_blank_line = 1;
4615 next;
4618 # print only one empty line
4619 # do not print empty line after signoff
4620 if ($line eq "") {
4621 next if ($skip_blank_line);
4622 $skip_blank_line = 1;
4623 } else {
4624 $skip_blank_line = 0;
4627 print format_log_line_html($line) . "<br/>\n";
4630 if ($opts{'-final_empty_line'}) {
4631 # end with single empty line
4632 print "<br/>\n" unless $skip_blank_line;
4636 # return link target (what link points to)
4637 sub git_get_link_target {
4638 my $hash = shift;
4639 my $link_target;
4641 # read link
4642 open my $fd, "-|", git_cmd(), "cat-file", "blob", $hash
4643 or return;
4645 local $/ = undef;
4646 $link_target = <$fd>;
4648 close $fd
4649 or return;
4651 return $link_target;
4654 # given link target, and the directory (basedir) the link is in,
4655 # return target of link relative to top directory (top tree);
4656 # return undef if it is not possible (including absolute links).
4657 sub normalize_link_target {
4658 my ($link_target, $basedir) = @_;
4660 # absolute symlinks (beginning with '/') cannot be normalized
4661 return if (substr($link_target, 0, 1) eq '/');
4663 # normalize link target to path from top (root) tree (dir)
4664 my $path;
4665 if ($basedir) {
4666 $path = $basedir . '/' . $link_target;
4667 } else {
4668 # we are in top (root) tree (dir)
4669 $path = $link_target;
4672 # remove //, /./, and /../
4673 my @path_parts;
4674 foreach my $part (split('/', $path)) {
4675 # discard '.' and ''
4676 next if (!$part || $part eq '.');
4677 # handle '..'
4678 if ($part eq '..') {
4679 if (@path_parts) {
4680 pop @path_parts;
4681 } else {
4682 # link leads outside repository (outside top dir)
4683 return;
4685 } else {
4686 push @path_parts, $part;
4689 $path = join('/', @path_parts);
4691 return $path;
4694 # print tree entry (row of git_tree), but without encompassing <tr> element
4695 sub git_print_tree_entry {
4696 my ($t, $basedir, $hash_base, $have_blame) = @_;
4698 my %base_key = ();
4699 $base_key{'hash_base'} = $hash_base if defined $hash_base;
4701 # The format of a table row is: mode list link. Where mode is
4702 # the mode of the entry, list is the name of the entry, an href,
4703 # and link is the action links of the entry.
4705 print "<td class=\"mode\">" . mode_str($t->{'mode'}) . "</td>\n";
4706 if (exists $t->{'size'}) {
4707 print "<td class=\"size\">$t->{'size'}</td>\n";
4709 if ($t->{'type'} eq "blob") {
4710 print "<td class=\"list\">" .
4711 $cgi->a({-href => href(action=>"blob", hash=>$t->{'hash'},
4712 file_name=>"$basedir$t->{'name'}", %base_key),
4713 -class => "list"}, esc_path($t->{'name'}));
4714 if (S_ISLNK(oct $t->{'mode'})) {
4715 my $link_target = git_get_link_target($t->{'hash'});
4716 if ($link_target) {
4717 my $norm_target = normalize_link_target($link_target, $basedir);
4718 if (defined $norm_target) {
4719 print " -> " .
4720 $cgi->a({-href => href(action=>"object", hash_base=>$hash_base,
4721 file_name=>$norm_target),
4722 -title => $norm_target}, esc_path($link_target));
4723 } else {
4724 print " -> " . esc_path($link_target);
4728 print "</td>\n";
4729 print "<td class=\"link\">";
4730 print $cgi->a({-href => href(action=>"blob", hash=>$t->{'hash'},
4731 file_name=>"$basedir$t->{'name'}", %base_key)},
4732 "blob");
4733 if ($have_blame) {
4734 print " | " .
4735 $cgi->a({-href => href(action=>"blame", hash=>$t->{'hash'},
4736 file_name=>"$basedir$t->{'name'}", %base_key)},
4737 "blame");
4739 if (defined $hash_base) {
4740 print " | " .
4741 $cgi->a({-href => href(action=>"history", hash_base=>$hash_base,
4742 hash=>$t->{'hash'}, file_name=>"$basedir$t->{'name'}")},
4743 "history");
4745 print " | " .
4746 $cgi->a({-href => href(action=>"blob_plain", hash_base=>$hash_base,
4747 file_name=>"$basedir$t->{'name'}")},
4748 "raw");
4749 print "</td>\n";
4751 } elsif ($t->{'type'} eq "tree") {
4752 print "<td class=\"list\">";
4753 print $cgi->a({-href => href(action=>"tree", hash=>$t->{'hash'},
4754 file_name=>"$basedir$t->{'name'}",
4755 %base_key)},
4756 esc_path($t->{'name'}));
4757 print "</td>\n";
4758 print "<td class=\"link\">";
4759 print $cgi->a({-href => href(action=>"tree", hash=>$t->{'hash'},
4760 file_name=>"$basedir$t->{'name'}",
4761 %base_key)},
4762 "tree");
4763 if (defined $hash_base) {
4764 print " | " .
4765 $cgi->a({-href => href(action=>"history", hash_base=>$hash_base,
4766 file_name=>"$basedir$t->{'name'}")},
4767 "history");
4769 print "</td>\n";
4770 } else {
4771 # unknown object: we can only present history for it
4772 # (this includes 'commit' object, i.e. submodule support)
4773 print "<td class=\"list\">" .
4774 esc_path($t->{'name'}) .
4775 "</td>\n";
4776 print "<td class=\"link\">";
4777 if (defined $hash_base) {
4778 print $cgi->a({-href => href(action=>"history",
4779 hash_base=>$hash_base,
4780 file_name=>"$basedir$t->{'name'}")},
4781 "history");
4783 print "</td>\n";
4787 ## ......................................................................
4788 ## functions printing large fragments of HTML
4790 # get pre-image filenames for merge (combined) diff
4791 sub fill_from_file_info {
4792 my ($diff, @parents) = @_;
4794 $diff->{'from_file'} = [ ];
4795 $diff->{'from_file'}[$diff->{'nparents'} - 1] = undef;
4796 for (my $i = 0; $i < $diff->{'nparents'}; $i++) {
4797 if ($diff->{'status'}[$i] eq 'R' ||
4798 $diff->{'status'}[$i] eq 'C') {
4799 $diff->{'from_file'}[$i] =
4800 git_get_path_by_hash($parents[$i], $diff->{'from_id'}[$i]);
4804 return $diff;
4807 # is current raw difftree line of file deletion
4808 sub is_deleted {
4809 my $diffinfo = shift;
4811 return $diffinfo->{'to_id'} eq ('0' x 40);
4814 # does patch correspond to [previous] difftree raw line
4815 # $diffinfo - hashref of parsed raw diff format
4816 # $patchinfo - hashref of parsed patch diff format
4817 # (the same keys as in $diffinfo)
4818 sub is_patch_split {
4819 my ($diffinfo, $patchinfo) = @_;
4821 return defined $diffinfo && defined $patchinfo
4822 && $diffinfo->{'to_file'} eq $patchinfo->{'to_file'};
4826 sub git_difftree_body {
4827 my ($difftree, $hash, @parents) = @_;
4828 my ($parent) = $parents[0];
4829 my $have_blame = gitweb_check_feature('blame');
4830 print "<div class=\"list_head\">\n";
4831 if ($#{$difftree} > 10) {
4832 print(($#{$difftree} + 1) . " files changed:\n");
4834 print "</div>\n";
4836 print "<table class=\"" .
4837 (@parents > 1 ? "combined " : "") .
4838 "diff_tree\">\n";
4840 # header only for combined diff in 'commitdiff' view
4841 my $has_header = @$difftree && @parents > 1 && $action eq 'commitdiff';
4842 if ($has_header) {
4843 # table header
4844 print "<thead><tr>\n" .
4845 "<th></th><th></th>\n"; # filename, patchN link
4846 for (my $i = 0; $i < @parents; $i++) {
4847 my $par = $parents[$i];
4848 print "<th>" .
4849 $cgi->a({-href => href(action=>"commitdiff",
4850 hash=>$hash, hash_parent=>$par),
4851 -title => 'commitdiff to parent number ' .
4852 ($i+1) . ': ' . substr($par,0,7)},
4853 $i+1) .
4854 "&nbsp;</th>\n";
4856 print "</tr></thead>\n<tbody>\n";
4859 my $alternate = 1;
4860 my $patchno = 0;
4861 foreach my $line (@{$difftree}) {
4862 my $diff = parsed_difftree_line($line);
4864 if ($alternate) {
4865 print "<tr class=\"dark\">\n";
4866 } else {
4867 print "<tr class=\"light\">\n";
4869 $alternate ^= 1;
4871 if (exists $diff->{'nparents'}) { # combined diff
4873 fill_from_file_info($diff, @parents)
4874 unless exists $diff->{'from_file'};
4876 if (!is_deleted($diff)) {
4877 # file exists in the result (child) commit
4878 print "<td>" .
4879 $cgi->a({-href => href(action=>"blob", hash=>$diff->{'to_id'},
4880 file_name=>$diff->{'to_file'},
4881 hash_base=>$hash),
4882 -class => "list"}, esc_path($diff->{'to_file'})) .
4883 "</td>\n";
4884 } else {
4885 print "<td>" .
4886 esc_path($diff->{'to_file'}) .
4887 "</td>\n";
4890 if ($action eq 'commitdiff') {
4891 # link to patch
4892 $patchno++;
4893 print "<td class=\"link\">" .
4894 $cgi->a({-href => href(-anchor=>"patch$patchno")},
4895 "patch") .
4896 " | " .
4897 "</td>\n";
4900 my $has_history = 0;
4901 my $not_deleted = 0;
4902 for (my $i = 0; $i < $diff->{'nparents'}; $i++) {
4903 my $hash_parent = $parents[$i];
4904 my $from_hash = $diff->{'from_id'}[$i];
4905 my $from_path = $diff->{'from_file'}[$i];
4906 my $status = $diff->{'status'}[$i];
4908 $has_history ||= ($status ne 'A');
4909 $not_deleted ||= ($status ne 'D');
4911 if ($status eq 'A') {
4912 print "<td class=\"link\" align=\"right\"> | </td>\n";
4913 } elsif ($status eq 'D') {
4914 print "<td class=\"link\">" .
4915 $cgi->a({-href => href(action=>"blob",
4916 hash_base=>$hash,
4917 hash=>$from_hash,
4918 file_name=>$from_path)},
4919 "blob" . ($i+1)) .
4920 " | </td>\n";
4921 } else {
4922 if ($diff->{'to_id'} eq $from_hash) {
4923 print "<td class=\"link nochange\">";
4924 } else {
4925 print "<td class=\"link\">";
4927 print $cgi->a({-href => href(action=>"blobdiff",
4928 hash=>$diff->{'to_id'},
4929 hash_parent=>$from_hash,
4930 hash_base=>$hash,
4931 hash_parent_base=>$hash_parent,
4932 file_name=>$diff->{'to_file'},
4933 file_parent=>$from_path)},
4934 "diff" . ($i+1)) .
4935 " | </td>\n";
4939 print "<td class=\"link\">";
4940 if ($not_deleted) {
4941 print $cgi->a({-href => href(action=>"blob",
4942 hash=>$diff->{'to_id'},
4943 file_name=>$diff->{'to_file'},
4944 hash_base=>$hash)},
4945 "blob");
4946 print " | " if ($has_history);
4948 if ($has_history) {
4949 print $cgi->a({-href => href(action=>"history",
4950 file_name=>$diff->{'to_file'},
4951 hash_base=>$hash)},
4952 "history");
4954 print "</td>\n";
4956 print "</tr>\n";
4957 next; # instead of 'else' clause, to avoid extra indent
4959 # else ordinary diff
4961 my ($to_mode_oct, $to_mode_str, $to_file_type);
4962 my ($from_mode_oct, $from_mode_str, $from_file_type);
4963 if ($diff->{'to_mode'} ne ('0' x 6)) {
4964 $to_mode_oct = oct $diff->{'to_mode'};
4965 if (S_ISREG($to_mode_oct)) { # only for regular file
4966 $to_mode_str = sprintf("%04o", $to_mode_oct & 0777); # permission bits
4968 $to_file_type = file_type($diff->{'to_mode'});
4970 if ($diff->{'from_mode'} ne ('0' x 6)) {
4971 $from_mode_oct = oct $diff->{'from_mode'};
4972 if (S_ISREG($from_mode_oct)) { # only for regular file
4973 $from_mode_str = sprintf("%04o", $from_mode_oct & 0777); # permission bits
4975 $from_file_type = file_type($diff->{'from_mode'});
4978 if ($diff->{'status'} eq "A") { # created
4979 my $mode_chng = "<span class=\"file_status new\">[new $to_file_type";
4980 $mode_chng .= " with mode: $to_mode_str" if $to_mode_str;
4981 $mode_chng .= "]</span>";
4982 print "<td>";
4983 print $cgi->a({-href => href(action=>"blob", hash=>$diff->{'to_id'},
4984 hash_base=>$hash, file_name=>$diff->{'file'}),
4985 -class => "list"}, esc_path($diff->{'file'}));
4986 print "</td>\n";
4987 print "<td>$mode_chng</td>\n";
4988 print "<td class=\"link\">";
4989 if ($action eq 'commitdiff') {
4990 # link to patch
4991 $patchno++;
4992 print $cgi->a({-href => href(-anchor=>"patch$patchno")},
4993 "patch") .
4994 " | ";
4996 print $cgi->a({-href => href(action=>"blob", hash=>$diff->{'to_id'},
4997 hash_base=>$hash, file_name=>$diff->{'file'})},
4998 "blob");
4999 print "</td>\n";
5001 } elsif ($diff->{'status'} eq "D") { # deleted
5002 my $mode_chng = "<span class=\"file_status deleted\">[deleted $from_file_type]</span>";
5003 print "<td>";
5004 print $cgi->a({-href => href(action=>"blob", hash=>$diff->{'from_id'},
5005 hash_base=>$parent, file_name=>$diff->{'file'}),
5006 -class => "list"}, esc_path($diff->{'file'}));
5007 print "</td>\n";
5008 print "<td>$mode_chng</td>\n";
5009 print "<td class=\"link\">";
5010 if ($action eq 'commitdiff') {
5011 # link to patch
5012 $patchno++;
5013 print $cgi->a({-href => href(-anchor=>"patch$patchno")},
5014 "patch") .
5015 " | ";
5017 print $cgi->a({-href => href(action=>"blob", hash=>$diff->{'from_id'},
5018 hash_base=>$parent, file_name=>$diff->{'file'})},
5019 "blob") . " | ";
5020 if ($have_blame) {
5021 print $cgi->a({-href => href(action=>"blame", hash_base=>$parent,
5022 file_name=>$diff->{'file'})},
5023 "blame") . " | ";
5025 print $cgi->a({-href => href(action=>"history", hash_base=>$parent,
5026 file_name=>$diff->{'file'})},
5027 "history");
5028 print "</td>\n";
5030 } elsif ($diff->{'status'} eq "M" || $diff->{'status'} eq "T") { # modified, or type changed
5031 my $mode_chnge = "";
5032 if ($diff->{'from_mode'} != $diff->{'to_mode'}) {
5033 $mode_chnge = "<span class=\"file_status mode_chnge\">[changed";
5034 if ($from_file_type ne $to_file_type) {
5035 $mode_chnge .= " from $from_file_type to $to_file_type";
5037 if (($from_mode_oct & 0777) != ($to_mode_oct & 0777)) {
5038 if ($from_mode_str && $to_mode_str) {
5039 $mode_chnge .= " mode: $from_mode_str->$to_mode_str";
5040 } elsif ($to_mode_str) {
5041 $mode_chnge .= " mode: $to_mode_str";
5044 $mode_chnge .= "]</span>\n";
5046 print "<td>";
5047 print $cgi->a({-href => href(action=>"blob", hash=>$diff->{'to_id'},
5048 hash_base=>$hash, file_name=>$diff->{'file'}),
5049 -class => "list"}, esc_path($diff->{'file'}));
5050 print "</td>\n";
5051 print "<td>$mode_chnge</td>\n";
5052 print "<td class=\"link\">";
5053 if ($action eq 'commitdiff') {
5054 # link to patch
5055 $patchno++;
5056 print $cgi->a({-href => href(-anchor=>"patch$patchno")},
5057 "patch") .
5058 " | ";
5059 } elsif ($diff->{'to_id'} ne $diff->{'from_id'}) {
5060 # "commit" view and modified file (not onlu mode changed)
5061 print $cgi->a({-href => href(action=>"blobdiff",
5062 hash=>$diff->{'to_id'}, hash_parent=>$diff->{'from_id'},
5063 hash_base=>$hash, hash_parent_base=>$parent,
5064 file_name=>$diff->{'file'})},
5065 "diff") .
5066 " | ";
5068 print $cgi->a({-href => href(action=>"blob", hash=>$diff->{'to_id'},
5069 hash_base=>$hash, file_name=>$diff->{'file'})},
5070 "blob") . " | ";
5071 if ($have_blame) {
5072 print $cgi->a({-href => href(action=>"blame", hash_base=>$hash,
5073 file_name=>$diff->{'file'})},
5074 "blame") . " | ";
5076 print $cgi->a({-href => href(action=>"history", hash_base=>$hash,
5077 file_name=>$diff->{'file'})},
5078 "history");
5079 print "</td>\n";
5081 } elsif ($diff->{'status'} eq "R" || $diff->{'status'} eq "C") { # renamed or copied
5082 my %status_name = ('R' => 'moved', 'C' => 'copied');
5083 my $nstatus = $status_name{$diff->{'status'}};
5084 my $mode_chng = "";
5085 if ($diff->{'from_mode'} != $diff->{'to_mode'}) {
5086 # mode also for directories, so we cannot use $to_mode_str
5087 $mode_chng = sprintf(", mode: %04o", $to_mode_oct & 0777);
5089 print "<td>" .
5090 $cgi->a({-href => href(action=>"blob", hash_base=>$hash,
5091 hash=>$diff->{'to_id'}, file_name=>$diff->{'to_file'}),
5092 -class => "list"}, esc_path($diff->{'to_file'})) . "</td>\n" .
5093 "<td><span class=\"file_status $nstatus\">[$nstatus from " .
5094 $cgi->a({-href => href(action=>"blob", hash_base=>$parent,
5095 hash=>$diff->{'from_id'}, file_name=>$diff->{'from_file'}),
5096 -class => "list"}, esc_path($diff->{'from_file'})) .
5097 " with " . (int $diff->{'similarity'}) . "% similarity$mode_chng]</span></td>\n" .
5098 "<td class=\"link\">";
5099 if ($action eq 'commitdiff') {
5100 # link to patch
5101 $patchno++;
5102 print $cgi->a({-href => href(-anchor=>"patch$patchno")},
5103 "patch") .
5104 " | ";
5105 } elsif ($diff->{'to_id'} ne $diff->{'from_id'}) {
5106 # "commit" view and modified file (not only pure rename or copy)
5107 print $cgi->a({-href => href(action=>"blobdiff",
5108 hash=>$diff->{'to_id'}, hash_parent=>$diff->{'from_id'},
5109 hash_base=>$hash, hash_parent_base=>$parent,
5110 file_name=>$diff->{'to_file'}, file_parent=>$diff->{'from_file'})},
5111 "diff") .
5112 " | ";
5114 print $cgi->a({-href => href(action=>"blob", hash=>$diff->{'to_id'},
5115 hash_base=>$parent, file_name=>$diff->{'to_file'})},
5116 "blob") . " | ";
5117 if ($have_blame) {
5118 print $cgi->a({-href => href(action=>"blame", hash_base=>$hash,
5119 file_name=>$diff->{'to_file'})},
5120 "blame") . " | ";
5122 print $cgi->a({-href => href(action=>"history", hash_base=>$hash,
5123 file_name=>$diff->{'to_file'})},
5124 "history");
5125 print "</td>\n";
5127 } # we should not encounter Unmerged (U) or Unknown (X) status
5128 print "</tr>\n";
5130 print "</tbody>" if $has_header;
5131 print "</table>\n";
5134 # Print context lines and then rem/add lines in a side-by-side manner.
5135 sub print_sidebyside_diff_lines {
5136 my ($ctx, $rem, $add) = @_;
5138 # print context block before add/rem block
5139 if (@$ctx) {
5140 print join '',
5141 '<div class="chunk_block ctx">',
5142 '<div class="old">',
5143 @$ctx,
5144 '</div>',
5145 '<div class="new">',
5146 @$ctx,
5147 '</div>',
5148 '</div>';
5151 if (!@$add) {
5152 # pure removal
5153 print join '',
5154 '<div class="chunk_block rem">',
5155 '<div class="old">',
5156 @$rem,
5157 '</div>',
5158 '</div>';
5159 } elsif (!@$rem) {
5160 # pure addition
5161 print join '',
5162 '<div class="chunk_block add">',
5163 '<div class="new">',
5164 @$add,
5165 '</div>',
5166 '</div>';
5167 } else {
5168 print join '',
5169 '<div class="chunk_block chg">',
5170 '<div class="old">',
5171 @$rem,
5172 '</div>',
5173 '<div class="new">',
5174 @$add,
5175 '</div>',
5176 '</div>';
5180 # Print context lines and then rem/add lines in inline manner.
5181 sub print_inline_diff_lines {
5182 my ($ctx, $rem, $add) = @_;
5184 print @$ctx, @$rem, @$add;
5187 # Format removed and added line, mark changed part and HTML-format them.
5188 # Implementation is based on contrib/diff-highlight
5189 sub format_rem_add_lines_pair {
5190 my ($rem, $add, $num_parents) = @_;
5192 # We need to untabify lines before split()'ing them;
5193 # otherwise offsets would be invalid.
5194 chomp $rem;
5195 chomp $add;
5196 $rem = untabify($rem);
5197 $add = untabify($add);
5199 my @rem = split(//, $rem);
5200 my @add = split(//, $add);
5201 my ($esc_rem, $esc_add);
5202 # Ignore leading +/- characters for each parent.
5203 my ($prefix_len, $suffix_len) = ($num_parents, 0);
5204 my ($prefix_has_nonspace, $suffix_has_nonspace);
5206 my $shorter = (@rem < @add) ? @rem : @add;
5207 while ($prefix_len < $shorter) {
5208 last if ($rem[$prefix_len] ne $add[$prefix_len]);
5210 $prefix_has_nonspace = 1 if ($rem[$prefix_len] !~ /\s/);
5211 $prefix_len++;
5214 while ($prefix_len + $suffix_len < $shorter) {
5215 last if ($rem[-1 - $suffix_len] ne $add[-1 - $suffix_len]);
5217 $suffix_has_nonspace = 1 if ($rem[-1 - $suffix_len] !~ /\s/);
5218 $suffix_len++;
5221 # Mark lines that are different from each other, but have some common
5222 # part that isn't whitespace. If lines are completely different, don't
5223 # mark them because that would make output unreadable, especially if
5224 # diff consists of multiple lines.
5225 if ($prefix_has_nonspace || $suffix_has_nonspace) {
5226 $esc_rem = esc_html_hl_regions($rem, 'marked',
5227 [$prefix_len, @rem - $suffix_len], -nbsp=>1);
5228 $esc_add = esc_html_hl_regions($add, 'marked',
5229 [$prefix_len, @add - $suffix_len], -nbsp=>1);
5230 } else {
5231 $esc_rem = esc_html($rem, -nbsp=>1);
5232 $esc_add = esc_html($add, -nbsp=>1);
5235 return format_diff_line(\$esc_rem, 'rem'),
5236 format_diff_line(\$esc_add, 'add');
5239 # HTML-format diff context, removed and added lines.
5240 sub format_ctx_rem_add_lines {
5241 my ($ctx, $rem, $add, $num_parents) = @_;
5242 my (@new_ctx, @new_rem, @new_add);
5243 my $can_highlight = 0;
5244 my $is_combined = ($num_parents > 1);
5246 # Highlight if every removed line has a corresponding added line.
5247 if (@$add > 0 && @$add == @$rem) {
5248 $can_highlight = 1;
5250 # Highlight lines in combined diff only if the chunk contains
5251 # diff between the same version, e.g.
5253 # - a
5254 # - b
5255 # + c
5256 # + d
5258 # Otherwise the highlightling would be confusing.
5259 if ($is_combined) {
5260 for (my $i = 0; $i < @$add; $i++) {
5261 my $prefix_rem = substr($rem->[$i], 0, $num_parents);
5262 my $prefix_add = substr($add->[$i], 0, $num_parents);
5264 $prefix_rem =~ s/-/+/g;
5266 if ($prefix_rem ne $prefix_add) {
5267 $can_highlight = 0;
5268 last;
5274 if ($can_highlight) {
5275 for (my $i = 0; $i < @$add; $i++) {
5276 my ($line_rem, $line_add) = format_rem_add_lines_pair(
5277 $rem->[$i], $add->[$i], $num_parents);
5278 push @new_rem, $line_rem;
5279 push @new_add, $line_add;
5281 } else {
5282 @new_rem = map { format_diff_line($_, 'rem') } @$rem;
5283 @new_add = map { format_diff_line($_, 'add') } @$add;
5286 @new_ctx = map { format_diff_line($_, 'ctx') } @$ctx;
5288 return (\@new_ctx, \@new_rem, \@new_add);
5291 # Print context lines and then rem/add lines.
5292 sub print_diff_lines {
5293 my ($ctx, $rem, $add, $diff_style, $num_parents) = @_;
5294 my $is_combined = $num_parents > 1;
5296 ($ctx, $rem, $add) = format_ctx_rem_add_lines($ctx, $rem, $add,
5297 $num_parents);
5299 if ($diff_style eq 'sidebyside' && !$is_combined) {
5300 print_sidebyside_diff_lines($ctx, $rem, $add);
5301 } else {
5302 # default 'inline' style and unknown styles
5303 print_inline_diff_lines($ctx, $rem, $add);
5307 sub print_diff_chunk {
5308 my ($diff_style, $num_parents, $from, $to, @chunk) = @_;
5309 my (@ctx, @rem, @add);
5311 # The class of the previous line.
5312 my $prev_class = '';
5314 return unless @chunk;
5316 # incomplete last line might be among removed or added lines,
5317 # or both, or among context lines: find which
5318 for (my $i = 1; $i < @chunk; $i++) {
5319 if ($chunk[$i][0] eq 'incomplete') {
5320 $chunk[$i][0] = $chunk[$i-1][0];
5324 # guardian
5325 push @chunk, ["", ""];
5327 foreach my $line_info (@chunk) {
5328 my ($class, $line) = @$line_info;
5330 # print chunk headers
5331 if ($class && $class eq 'chunk_header') {
5332 print format_diff_line($line, $class, $from, $to);
5333 next;
5336 ## print from accumulator when have some add/rem lines or end
5337 # of chunk (flush context lines), or when have add and rem
5338 # lines and new block is reached (otherwise add/rem lines could
5339 # be reordered)
5340 if (!$class || ((@rem || @add) && $class eq 'ctx') ||
5341 (@rem && @add && $class ne $prev_class)) {
5342 print_diff_lines(\@ctx, \@rem, \@add,
5343 $diff_style, $num_parents);
5344 @ctx = @rem = @add = ();
5347 ## adding lines to accumulator
5348 # guardian value
5349 last unless $line;
5350 # rem, add or change
5351 if ($class eq 'rem') {
5352 push @rem, $line;
5353 } elsif ($class eq 'add') {
5354 push @add, $line;
5356 # context line
5357 if ($class eq 'ctx') {
5358 push @ctx, $line;
5361 $prev_class = $class;
5365 sub git_patchset_body {
5366 my ($fd, $diff_style, $difftree, $hash, @hash_parents) = @_;
5367 my ($hash_parent) = $hash_parents[0];
5369 my $is_combined = (@hash_parents > 1);
5370 my $patch_idx = 0;
5371 my $patch_number = 0;
5372 my $patch_line;
5373 my $diffinfo;
5374 my $to_name;
5375 my (%from, %to);
5376 my @chunk; # for side-by-side diff
5378 print "<div class=\"patchset\">\n";
5380 # skip to first patch
5381 while ($patch_line = <$fd>) {
5382 chomp $patch_line;
5384 last if ($patch_line =~ m/^diff /);
5387 PATCH:
5388 while ($patch_line) {
5390 # parse "git diff" header line
5391 if ($patch_line =~ m/^diff --git (\"(?:[^\\\"]*(?:\\.[^\\\"]*)*)\"|[^ "]*) (.*)$/) {
5392 # $1 is from_name, which we do not use
5393 $to_name = unquote($2);
5394 $to_name =~ s!^b/!!;
5395 } elsif ($patch_line =~ m/^diff --(cc|combined) ("?.*"?)$/) {
5396 # $1 is 'cc' or 'combined', which we do not use
5397 $to_name = unquote($2);
5398 } else {
5399 $to_name = undef;
5402 # check if current patch belong to current raw line
5403 # and parse raw git-diff line if needed
5404 if (is_patch_split($diffinfo, { 'to_file' => $to_name })) {
5405 # this is continuation of a split patch
5406 print "<div class=\"patch cont\">\n";
5407 } else {
5408 # advance raw git-diff output if needed
5409 $patch_idx++ if defined $diffinfo;
5411 # read and prepare patch information
5412 $diffinfo = parsed_difftree_line($difftree->[$patch_idx]);
5414 # compact combined diff output can have some patches skipped
5415 # find which patch (using pathname of result) we are at now;
5416 if ($is_combined) {
5417 while ($to_name ne $diffinfo->{'to_file'}) {
5418 print "<div class=\"patch\" id=\"patch". ($patch_idx+1) ."\">\n" .
5419 format_diff_cc_simplified($diffinfo, @hash_parents) .
5420 "</div>\n"; # class="patch"
5422 $patch_idx++;
5423 $patch_number++;
5425 last if $patch_idx > $#$difftree;
5426 $diffinfo = parsed_difftree_line($difftree->[$patch_idx]);
5430 # modifies %from, %to hashes
5431 parse_from_to_diffinfo($diffinfo, \%from, \%to, @hash_parents);
5433 # this is first patch for raw difftree line with $patch_idx index
5434 # we index @$difftree array from 0, but number patches from 1
5435 print "<div class=\"patch\" id=\"patch". ($patch_idx+1) ."\">\n";
5438 # git diff header
5439 #assert($patch_line =~ m/^diff /) if DEBUG;
5440 #assert($patch_line !~ m!$/$!) if DEBUG; # is chomp-ed
5441 $patch_number++;
5442 # print "git diff" header
5443 print format_git_diff_header_line($patch_line, $diffinfo,
5444 \%from, \%to);
5446 # print extended diff header
5447 print "<div class=\"diff extended_header\">\n";
5448 EXTENDED_HEADER:
5449 while ($patch_line = <$fd>) {
5450 chomp $patch_line;
5452 last EXTENDED_HEADER if ($patch_line =~ m/^--- |^diff /);
5454 print format_extended_diff_header_line($patch_line, $diffinfo,
5455 \%from, \%to);
5457 print "</div>\n"; # class="diff extended_header"
5459 # from-file/to-file diff header
5460 if (! $patch_line) {
5461 print "</div>\n"; # class="patch"
5462 last PATCH;
5464 next PATCH if ($patch_line =~ m/^diff /);
5465 #assert($patch_line =~ m/^---/) if DEBUG;
5467 my $last_patch_line = $patch_line;
5468 $patch_line = <$fd>;
5469 chomp $patch_line;
5470 #assert($patch_line =~ m/^\+\+\+/) if DEBUG;
5472 print format_diff_from_to_header($last_patch_line, $patch_line,
5473 $diffinfo, \%from, \%to,
5474 @hash_parents);
5476 # the patch itself
5477 LINE:
5478 while ($patch_line = <$fd>) {
5479 chomp $patch_line;
5481 next PATCH if ($patch_line =~ m/^diff /);
5483 my $class = diff_line_class($patch_line, \%from, \%to);
5485 if ($class eq 'chunk_header') {
5486 print_diff_chunk($diff_style, scalar @hash_parents, \%from, \%to, @chunk);
5487 @chunk = ();
5490 push @chunk, [ $class, $patch_line ];
5493 } continue {
5494 if (@chunk) {
5495 print_diff_chunk($diff_style, scalar @hash_parents, \%from, \%to, @chunk);
5496 @chunk = ();
5498 print "</div>\n"; # class="patch"
5501 # for compact combined (--cc) format, with chunk and patch simplification
5502 # the patchset might be empty, but there might be unprocessed raw lines
5503 for (++$patch_idx if $patch_number > 0;
5504 $patch_idx < @$difftree;
5505 ++$patch_idx) {
5506 # read and prepare patch information
5507 $diffinfo = parsed_difftree_line($difftree->[$patch_idx]);
5509 # generate anchor for "patch" links in difftree / whatchanged part
5510 print "<div class=\"patch\" id=\"patch". ($patch_idx+1) ."\">\n" .
5511 format_diff_cc_simplified($diffinfo, @hash_parents) .
5512 "</div>\n"; # class="patch"
5514 $patch_number++;
5517 if ($patch_number == 0) {
5518 if (@hash_parents > 1) {
5519 print "<div class=\"diff nodifferences\">Trivial merge</div>\n";
5520 } else {
5521 print "<div class=\"diff nodifferences\">No differences found</div>\n";
5525 print "</div>\n"; # class="patchset"
5528 # . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5530 sub git_project_search_form {
5531 my ($searchtext, $search_use_regexp) = @_;
5533 my $limit = '';
5534 if ($project_filter) {
5535 $limit = " in '$project_filter/'";
5538 print "<div class=\"projsearch\">\n";
5539 print $cgi->start_form(-method => 'get', -action => $my_uri) .
5540 $cgi->hidden(-name => 'a', -value => 'project_list') . "\n";
5541 print $cgi->hidden(-name => 'pf', -value => $project_filter). "\n"
5542 if (defined $project_filter);
5543 print $cgi->textfield(-name => 's', -value => $searchtext,
5544 -title => "Search project by name and description$limit",
5545 -size => 60) . "\n" .
5546 "<span title=\"Extended regular expression\">" .
5547 $cgi->checkbox(-name => 'sr', -value => 1, -label => 're',
5548 -checked => $search_use_regexp) .
5549 "</span>\n" .
5550 $cgi->submit(-name => 'btnS', -value => 'Search') .
5551 $cgi->end_form() . "\n" .
5552 $cgi->a({-href => href(project => undef, searchtext => undef,
5553 project_filter => $project_filter)},
5554 esc_html("List all projects$limit")) . "<br />\n";
5555 print "</div>\n";
5558 # entry for given @keys needs filling if at least one of keys in list
5559 # is not present in %$project_info
5560 sub project_info_needs_filling {
5561 my ($project_info, @keys) = @_;
5563 # return List::MoreUtils::any { !exists $project_info->{$_} } @keys;
5564 foreach my $key (@keys) {
5565 if (!exists $project_info->{$key}) {
5566 return 1;
5569 return;
5572 # fills project list info (age, description, owner, category, forks, etc.)
5573 # for each project in the list, removing invalid projects from
5574 # returned list, or fill only specified info.
5576 # Invalid projects are removed from the returned list if and only if you
5577 # ask 'age' or 'age_string' to be filled, because they are the only fields
5578 # that run unconditionally git command that requires repository, and
5579 # therefore do always check if project repository is invalid.
5581 # USAGE:
5582 # * fill_project_list_info(\@project_list, 'descr_long', 'ctags')
5583 # ensures that 'descr_long' and 'ctags' fields are filled
5584 # * @project_list = fill_project_list_info(\@project_list)
5585 # ensures that all fields are filled (and invalid projects removed)
5587 # NOTE: modifies $projlist, but does not remove entries from it
5588 sub fill_project_list_info {
5589 my ($projlist, @wanted_keys) = @_;
5590 my @projects;
5591 my $filter_set = sub { return @_; };
5592 if (@wanted_keys) {
5593 my %wanted_keys = map { $_ => 1 } @wanted_keys;
5594 $filter_set = sub { return grep { $wanted_keys{$_} } @_; };
5597 my $show_ctags = gitweb_check_feature('ctags');
5598 PROJECT:
5599 foreach my $pr (@$projlist) {
5600 if (project_info_needs_filling($pr, $filter_set->('age', 'age_string'))) {
5601 my (@activity) = git_get_last_activity($pr->{'path'});
5602 unless (@activity) {
5603 next PROJECT;
5605 ($pr->{'age'}, $pr->{'age_string'}) = @activity;
5607 if (project_info_needs_filling($pr, $filter_set->('descr', 'descr_long'))) {
5608 my $descr = git_get_project_description($pr->{'path'}) || "";
5609 $descr = to_utf8($descr);
5610 $pr->{'descr_long'} = $descr;
5611 $pr->{'descr'} = chop_str($descr, $projects_list_description_width, 5);
5613 if (project_info_needs_filling($pr, $filter_set->('owner'))) {
5614 $pr->{'owner'} = git_get_project_owner("$pr->{'path'}") || "";
5616 if ($show_ctags &&
5617 project_info_needs_filling($pr, $filter_set->('ctags'))) {
5618 $pr->{'ctags'} = git_get_project_ctags($pr->{'path'});
5620 if ($projects_list_group_categories &&
5621 project_info_needs_filling($pr, $filter_set->('category'))) {
5622 my $cat = git_get_project_category($pr->{'path'}) ||
5623 $project_list_default_category;
5624 $pr->{'category'} = to_utf8($cat);
5627 push @projects, $pr;
5630 return @projects;
5633 sub sort_projects_list {
5634 my ($projlist, $order) = @_;
5636 sub order_str {
5637 my $key = shift;
5638 return sub { $a->{$key} cmp $b->{$key} };
5641 sub order_num_then_undef {
5642 my $key = shift;
5643 return sub {
5644 defined $a->{$key} ?
5645 (defined $b->{$key} ? $a->{$key} <=> $b->{$key} : -1) :
5646 (defined $b->{$key} ? 1 : 0)
5650 my %orderings = (
5651 project => order_str('path'),
5652 descr => order_str('descr_long'),
5653 owner => order_str('owner'),
5654 age => order_num_then_undef('age'),
5657 my $ordering = $orderings{$order};
5658 return defined $ordering ? sort $ordering @$projlist : @$projlist;
5661 # returns a hash of categories, containing the list of project
5662 # belonging to each category
5663 sub build_projlist_by_category {
5664 my ($projlist, $from, $to) = @_;
5665 my %categories;
5667 $from = 0 unless defined $from;
5668 $to = $#$projlist if (!defined $to || $#$projlist < $to);
5670 for (my $i = $from; $i <= $to; $i++) {
5671 my $pr = $projlist->[$i];
5672 push @{$categories{ $pr->{'category'} }}, $pr;
5675 return wantarray ? %categories : \%categories;
5678 # print 'sort by' <th> element, generating 'sort by $name' replay link
5679 # if that order is not selected
5680 sub print_sort_th {
5681 print format_sort_th(@_);
5684 sub format_sort_th {
5685 my ($name, $order, $header) = @_;
5686 my $sort_th = "";
5687 $header ||= ucfirst($name);
5689 if ($order eq $name) {
5690 $sort_th .= "<th>$header</th>\n";
5691 } else {
5692 $sort_th .= "<th>" .
5693 $cgi->a({-href => href(-replay=>1, order=>$name),
5694 -class => "header"}, $header) .
5695 "</th>\n";
5698 return $sort_th;
5701 sub git_project_list_rows {
5702 my ($projlist, $from, $to, $check_forks) = @_;
5704 $from = 0 unless defined $from;
5705 $to = $#$projlist if (!defined $to || $#$projlist < $to);
5707 my $alternate = 1;
5708 for (my $i = $from; $i <= $to; $i++) {
5709 my $pr = $projlist->[$i];
5711 if ($alternate) {
5712 print "<tr class=\"dark\">\n";
5713 } else {
5714 print "<tr class=\"light\">\n";
5716 $alternate ^= 1;
5718 if ($check_forks) {
5719 print "<td>";
5720 if ($pr->{'forks'}) {
5721 my $nforks = scalar @{$pr->{'forks'}};
5722 if ($nforks > 0) {
5723 print $cgi->a({-href => href(project=>$pr->{'path'}, action=>"forks"),
5724 -title => "$nforks forks"}, "+");
5725 } else {
5726 print $cgi->span({-title => "$nforks forks"}, "+");
5729 print "</td>\n";
5731 print "<td>" . $cgi->a({-href => href(project=>$pr->{'path'}, action=>"summary"),
5732 -class => "list"},
5733 esc_html_match_hl($pr->{'path'}, $search_regexp)) .
5734 "</td>\n" .
5735 "<td>" . $cgi->a({-href => href(project=>$pr->{'path'}, action=>"summary"),
5736 -class => "list",
5737 -title => $pr->{'descr_long'}},
5738 $search_regexp
5739 ? esc_html_match_hl_chopped($pr->{'descr_long'},
5740 $pr->{'descr'}, $search_regexp)
5741 : esc_html($pr->{'descr'})) .
5742 "</td>\n";
5743 unless ($omit_owner) {
5744 print "<td><i>" . chop_and_escape_str($pr->{'owner'}, 15) . "</i></td>\n";
5746 unless ($omit_age_column) {
5747 print "<td class=\"". age_class($pr->{'age'}) . "\">" .
5748 (defined $pr->{'age_string'} ? $pr->{'age_string'} : "No commits") . "</td>\n";
5750 print"<td class=\"link\">" .
5751 $cgi->a({-href => href(project=>$pr->{'path'}, action=>"summary")}, "summary") . " | " .
5752 $cgi->a({-href => href(project=>$pr->{'path'}, action=>"shortlog")}, "shortlog") . " | " .
5753 $cgi->a({-href => href(project=>$pr->{'path'}, action=>"log")}, "log") . " | " .
5754 $cgi->a({-href => href(project=>$pr->{'path'}, action=>"tree")}, "tree") .
5755 ($pr->{'forks'} ? " | " . $cgi->a({-href => href(project=>$pr->{'path'}, action=>"forks")}, "forks") : '') .
5756 "</td>\n" .
5757 "</tr>\n";
5761 sub git_project_list_body {
5762 # actually uses global variable $project
5763 my ($projlist, $order, $from, $to, $extra, $no_header) = @_;
5764 my @projects = @$projlist;
5766 my $check_forks = gitweb_check_feature('forks');
5767 my $show_ctags = gitweb_check_feature('ctags');
5768 my $tagfilter = $show_ctags ? $input_params{'ctag'} : undef;
5769 $check_forks = undef
5770 if ($tagfilter || $search_regexp);
5772 # filtering out forks before filling info allows to do less work
5773 @projects = filter_forks_from_projects_list(\@projects)
5774 if ($check_forks);
5775 # search_projects_list pre-fills required info
5776 @projects = search_projects_list(\@projects,
5777 'search_regexp' => $search_regexp,
5778 'tagfilter' => $tagfilter)
5779 if ($tagfilter || $search_regexp);
5780 # fill the rest
5781 my @all_fields = ('descr', 'descr_long', 'ctags', 'category');
5782 push @all_fields, ('age', 'age_string') unless($omit_age_column);
5783 push @all_fields, 'owner' unless($omit_owner);
5784 @projects = fill_project_list_info(\@projects, @all_fields);
5786 $order ||= $default_projects_order;
5787 $from = 0 unless defined $from;
5788 $to = $#projects if (!defined $to || $#projects < $to);
5790 # short circuit
5791 if ($from > $to) {
5792 print "<center>\n".
5793 "<b>No such projects found</b><br />\n".
5794 "Click ".$cgi->a({-href=>href(project=>undef)},"here")." to view all projects<br />\n".
5795 "</center>\n<br />\n";
5796 return;
5799 @projects = sort_projects_list(\@projects, $order);
5801 if ($show_ctags) {
5802 my $ctags = git_gather_all_ctags(\@projects);
5803 my $cloud = git_populate_project_tagcloud($ctags);
5804 print git_show_project_tagcloud($cloud, 64);
5807 print "<table class=\"project_list\">\n";
5808 unless ($no_header) {
5809 print "<tr>\n";
5810 if ($check_forks) {
5811 print "<th></th>\n";
5813 print_sort_th('project', $order, 'Project');
5814 print_sort_th('descr', $order, 'Description');
5815 print_sort_th('owner', $order, 'Owner') unless $omit_owner;
5816 print_sort_th('age', $order, 'Last Change') unless $omit_age_column;
5817 print "<th></th>\n" . # for links
5818 "</tr>\n";
5821 if ($projects_list_group_categories) {
5822 # only display categories with projects in the $from-$to window
5823 @projects = sort {$a->{'category'} cmp $b->{'category'}} @projects[$from..$to];
5824 my %categories = build_projlist_by_category(\@projects, $from, $to);
5825 foreach my $cat (sort keys %categories) {
5826 unless ($cat eq "") {
5827 print "<tr>\n";
5828 if ($check_forks) {
5829 print "<td></td>\n";
5831 print "<td class=\"category\" colspan=\"5\">".esc_html($cat)."</td>\n";
5832 print "</tr>\n";
5835 git_project_list_rows($categories{$cat}, undef, undef, $check_forks);
5837 } else {
5838 git_project_list_rows(\@projects, $from, $to, $check_forks);
5841 if (defined $extra) {
5842 print "<tr>\n";
5843 if ($check_forks) {
5844 print "<td></td>\n";
5846 print "<td colspan=\"5\">$extra</td>\n" .
5847 "</tr>\n";
5849 print "</table>\n";
5852 sub git_log_body {
5853 # uses global variable $project
5854 my ($commitlist, $from, $to, $refs, $extra) = @_;
5856 $from = 0 unless defined $from;
5857 $to = $#{$commitlist} if (!defined $to || $#{$commitlist} < $to);
5859 for (my $i = 0; $i <= $to; $i++) {
5860 my %co = %{$commitlist->[$i]};
5861 next if !%co;
5862 my $commit = $co{'id'};
5863 my $ref = format_ref_marker($refs, $commit);
5864 git_print_header_div('commit',
5865 "<span class=\"age\">$co{'age_string'}</span>" .
5866 esc_html($co{'title'}) . $ref,
5867 $commit);
5868 print "<div class=\"title_text\">\n" .
5869 "<div class=\"log_link\">\n" .
5870 $cgi->a({-href => href(action=>"commit", hash=>$commit)}, "commit") .
5871 " | " .
5872 $cgi->a({-href => href(action=>"commitdiff", hash=>$commit)}, "commitdiff") .
5873 " | " .
5874 $cgi->a({-href => href(action=>"tree", hash=>$commit, hash_base=>$commit)}, "tree") .
5875 "<br/>\n" .
5876 "</div>\n";
5877 git_print_authorship(\%co, -tag => 'span');
5878 print "<br/>\n</div>\n";
5880 print "<div class=\"log_body\">\n";
5881 git_print_log($co{'comment'}, -final_empty_line=> 1);
5882 print "</div>\n";
5884 if ($extra) {
5885 print "<div class=\"page_nav\">\n";
5886 print "$extra\n";
5887 print "</div>\n";
5891 sub git_shortlog_body {
5892 # uses global variable $project
5893 my ($commitlist, $from, $to, $refs, $extra) = @_;
5895 $from = 0 unless defined $from;
5896 $to = $#{$commitlist} if (!defined $to || $#{$commitlist} < $to);
5898 print "<table class=\"shortlog\">\n";
5899 my $alternate = 1;
5900 for (my $i = $from; $i <= $to; $i++) {
5901 my %co = %{$commitlist->[$i]};
5902 my $commit = $co{'id'};
5903 my $ref = format_ref_marker($refs, $commit);
5904 if ($alternate) {
5905 print "<tr class=\"dark\">\n";
5906 } else {
5907 print "<tr class=\"light\">\n";
5909 $alternate ^= 1;
5910 # git_summary() used print "<td><i>$co{'age_string'}</i></td>\n" .
5911 print "<td title=\"$co{'age_string_age'}\"><i>$co{'age_string_date'}</i></td>\n" .
5912 format_author_html('td', \%co, 10) . "<td>";
5913 print format_subject_html($co{'title'}, $co{'title_short'},
5914 href(action=>"commit", hash=>$commit), $ref);
5915 print "</td>\n" .
5916 "<td class=\"link\">" .
5917 $cgi->a({-href => href(action=>"commit", hash=>$commit)}, "commit") . " | " .
5918 $cgi->a({-href => href(action=>"commitdiff", hash=>$commit)}, "commitdiff") . " | " .
5919 $cgi->a({-href => href(action=>"tree", hash=>$commit, hash_base=>$commit)}, "tree");
5920 my $snapshot_links = format_snapshot_links($commit);
5921 if (defined $snapshot_links) {
5922 print " | " . $snapshot_links;
5924 print "</td>\n" .
5925 "</tr>\n";
5927 if (defined $extra) {
5928 print "<tr>\n" .
5929 "<td colspan=\"4\">$extra</td>\n" .
5930 "</tr>\n";
5932 print "</table>\n";
5935 sub git_history_body {
5936 # Warning: assumes constant type (blob or tree) during history
5937 my ($commitlist, $from, $to, $refs, $extra,
5938 $file_name, $file_hash, $ftype) = @_;
5940 $from = 0 unless defined $from;
5941 $to = $#{$commitlist} unless (defined $to && $to <= $#{$commitlist});
5943 print "<table class=\"history\">\n";
5944 my $alternate = 1;
5945 for (my $i = $from; $i <= $to; $i++) {
5946 my %co = %{$commitlist->[$i]};
5947 if (!%co) {
5948 next;
5950 my $commit = $co{'id'};
5952 my $ref = format_ref_marker($refs, $commit);
5954 if ($alternate) {
5955 print "<tr class=\"dark\">\n";
5956 } else {
5957 print "<tr class=\"light\">\n";
5959 $alternate ^= 1;
5960 print "<td title=\"$co{'age_string_age'}\"><i>$co{'age_string_date'}</i></td>\n" .
5961 # shortlog: format_author_html('td', \%co, 10)
5962 format_author_html('td', \%co, 15, 3) . "<td>";
5963 # originally git_history used chop_str($co{'title'}, 50)
5964 print format_subject_html($co{'title'}, $co{'title_short'},
5965 href(action=>"commit", hash=>$commit), $ref);
5966 print "</td>\n" .
5967 "<td class=\"link\">" .
5968 $cgi->a({-href => href(action=>$ftype, hash_base=>$commit, file_name=>$file_name)}, $ftype) . " | " .
5969 $cgi->a({-href => href(action=>"commitdiff", hash=>$commit)}, "commitdiff");
5971 if ($ftype eq 'blob') {
5972 my $blob_current = $file_hash;
5973 my $blob_parent = git_get_hash_by_path($commit, $file_name);
5974 if (defined $blob_current && defined $blob_parent &&
5975 $blob_current ne $blob_parent) {
5976 print " | " .
5977 $cgi->a({-href => href(action=>"blobdiff",
5978 hash=>$blob_current, hash_parent=>$blob_parent,
5979 hash_base=>$hash_base, hash_parent_base=>$commit,
5980 file_name=>$file_name)},
5981 "diff to current");
5984 print "</td>\n" .
5985 "</tr>\n";
5987 if (defined $extra) {
5988 print "<tr>\n" .
5989 "<td colspan=\"4\">$extra</td>\n" .
5990 "</tr>\n";
5992 print "</table>\n";
5995 sub git_tags_body {
5996 # uses global variable $project
5997 my ($taglist, $from, $to, $extra) = @_;
5998 $from = 0 unless defined $from;
5999 $to = $#{$taglist} if (!defined $to || $#{$taglist} < $to);
6001 print "<table class=\"tags\">\n";
6002 my $alternate = 1;
6003 for (my $i = $from; $i <= $to; $i++) {
6004 my $entry = $taglist->[$i];
6005 my %tag = %$entry;
6006 my $comment = $tag{'subject'};
6007 my $comment_short;
6008 if (defined $comment) {
6009 $comment_short = chop_str($comment, 30, 5);
6011 if ($alternate) {
6012 print "<tr class=\"dark\">\n";
6013 } else {
6014 print "<tr class=\"light\">\n";
6016 $alternate ^= 1;
6017 if (defined $tag{'age'}) {
6018 print "<td><i>$tag{'age'}</i></td>\n";
6019 } else {
6020 print "<td></td>\n";
6022 print "<td>" .
6023 $cgi->a({-href => href(action=>$tag{'reftype'}, hash=>$tag{'refid'}),
6024 -class => "list name"}, esc_html($tag{'name'})) .
6025 "</td>\n" .
6026 "<td>";
6027 if (defined $comment) {
6028 print format_subject_html($comment, $comment_short,
6029 href(action=>"tag", hash=>$tag{'id'}));
6031 print "</td>\n" .
6032 "<td class=\"selflink\">";
6033 if ($tag{'type'} eq "tag") {
6034 print $cgi->a({-href => href(action=>"tag", hash=>$tag{'id'})}, "tag");
6035 } else {
6036 print "&nbsp;";
6038 print "</td>\n" .
6039 "<td class=\"link\">" . " | " .
6040 $cgi->a({-href => href(action=>$tag{'reftype'}, hash=>$tag{'refid'})}, $tag{'reftype'});
6041 if ($tag{'reftype'} eq "commit") {
6042 print " | " . $cgi->a({-href => href(action=>"shortlog", hash=>$tag{'fullname'})}, "shortlog") .
6043 " | " . $cgi->a({-href => href(action=>"log", hash=>$tag{'fullname'})}, "log");
6044 } elsif ($tag{'reftype'} eq "blob") {
6045 print " | " . $cgi->a({-href => href(action=>"blob_plain", hash=>$tag{'refid'})}, "raw");
6047 print "</td>\n" .
6048 "</tr>";
6050 if (defined $extra) {
6051 print "<tr>\n" .
6052 "<td colspan=\"5\">$extra</td>\n" .
6053 "</tr>\n";
6055 print "</table>\n";
6058 sub git_heads_body {
6059 # uses global variable $project
6060 my ($headlist, $head_at, $from, $to, $extra) = @_;
6061 $from = 0 unless defined $from;
6062 $to = $#{$headlist} if (!defined $to || $#{$headlist} < $to);
6064 print "<table class=\"heads\">\n";
6065 my $alternate = 1;
6066 for (my $i = $from; $i <= $to; $i++) {
6067 my $entry = $headlist->[$i];
6068 my %ref = %$entry;
6069 my $curr = defined $head_at && $ref{'id'} eq $head_at;
6070 if ($alternate) {
6071 print "<tr class=\"dark\">\n";
6072 } else {
6073 print "<tr class=\"light\">\n";
6075 $alternate ^= 1;
6076 print "<td><i>$ref{'age'}</i></td>\n" .
6077 ($curr ? "<td class=\"current_head\">" : "<td>") .
6078 $cgi->a({-href => href(action=>"shortlog", hash=>$ref{'fullname'}),
6079 -class => "list name"},esc_html($ref{'name'})) .
6080 "</td>\n" .
6081 "<td class=\"link\">" .
6082 $cgi->a({-href => href(action=>"shortlog", hash=>$ref{'fullname'})}, "shortlog") . " | " .
6083 $cgi->a({-href => href(action=>"log", hash=>$ref{'fullname'})}, "log") . " | " .
6084 $cgi->a({-href => href(action=>"tree", hash=>$ref{'fullname'}, hash_base=>$ref{'fullname'})}, "tree") .
6085 "</td>\n" .
6086 "</tr>";
6088 if (defined $extra) {
6089 print "<tr>\n" .
6090 "<td colspan=\"3\">$extra</td>\n" .
6091 "</tr>\n";
6093 print "</table>\n";
6096 # Display a single remote block
6097 sub git_remote_block {
6098 my ($remote, $rdata, $limit, $head) = @_;
6100 my $heads = $rdata->{'heads'};
6101 my $fetch = $rdata->{'fetch'};
6102 my $push = $rdata->{'push'};
6104 my $urls_table = "<table class=\"projects_list\">\n" ;
6106 if (defined $fetch) {
6107 if ($fetch eq $push) {
6108 $urls_table .= format_repo_url("URL", $fetch);
6109 } else {
6110 $urls_table .= format_repo_url("Fetch URL", $fetch);
6111 $urls_table .= format_repo_url("Push URL", $push) if defined $push;
6113 } elsif (defined $push) {
6114 $urls_table .= format_repo_url("Push URL", $push);
6115 } else {
6116 $urls_table .= format_repo_url("", "No remote URL");
6119 $urls_table .= "</table>\n";
6121 my $dots;
6122 if (defined $limit && $limit < @$heads) {
6123 $dots = $cgi->a({-href => href(action=>"remotes", hash=>$remote)}, "...");
6126 print $urls_table;
6127 git_heads_body($heads, $head, 0, $limit, $dots);
6130 # Display a list of remote names with the respective fetch and push URLs
6131 sub git_remotes_list {
6132 my ($remotedata, $limit) = @_;
6133 print "<table class=\"heads\">\n";
6134 my $alternate = 1;
6135 my @remotes = sort keys %$remotedata;
6137 my $limited = $limit && $limit < @remotes;
6139 $#remotes = $limit - 1 if $limited;
6141 while (my $remote = shift @remotes) {
6142 my $rdata = $remotedata->{$remote};
6143 my $fetch = $rdata->{'fetch'};
6144 my $push = $rdata->{'push'};
6145 if ($alternate) {
6146 print "<tr class=\"dark\">\n";
6147 } else {
6148 print "<tr class=\"light\">\n";
6150 $alternate ^= 1;
6151 print "<td>" .
6152 $cgi->a({-href=> href(action=>'remotes', hash=>$remote),
6153 -class=> "list name"},esc_html($remote)) .
6154 "</td>";
6155 print "<td class=\"link\">" .
6156 (defined $fetch ? $cgi->a({-href=> $fetch}, "fetch") : "fetch") .
6157 " | " .
6158 (defined $push ? $cgi->a({-href=> $push}, "push") : "push") .
6159 "</td>";
6161 print "</tr>\n";
6164 if ($limited) {
6165 print "<tr>\n" .
6166 "<td colspan=\"3\">" .
6167 $cgi->a({-href => href(action=>"remotes")}, "...") .
6168 "</td>\n" . "</tr>\n";
6171 print "</table>";
6174 # Display remote heads grouped by remote, unless there are too many
6175 # remotes, in which case we only display the remote names
6176 sub git_remotes_body {
6177 my ($remotedata, $limit, $head) = @_;
6178 if ($limit and $limit < keys %$remotedata) {
6179 git_remotes_list($remotedata, $limit);
6180 } else {
6181 fill_remote_heads($remotedata);
6182 while (my ($remote, $rdata) = each %$remotedata) {
6183 git_print_section({-class=>"remote", -id=>$remote},
6184 ["remotes", $remote, $remote], sub {
6185 git_remote_block($remote, $rdata, $limit, $head);
6191 sub git_search_message {
6192 my %co = @_;
6194 my $greptype;
6195 if ($searchtype eq 'commit') {
6196 $greptype = "--grep=";
6197 } elsif ($searchtype eq 'author') {
6198 $greptype = "--author=";
6199 } elsif ($searchtype eq 'committer') {
6200 $greptype = "--committer=";
6202 $greptype .= $searchtext;
6203 my @commitlist = parse_commits($hash, 101, (100 * $page), undef,
6204 $greptype, '--regexp-ignore-case',
6205 $search_use_regexp ? '--extended-regexp' : '--fixed-strings');
6207 my $paging_nav = '';
6208 if ($page > 0) {
6209 $paging_nav .=
6210 $cgi->a({-href => href(-replay=>1, page=>undef)},
6211 "first") .
6212 " &sdot; " .
6213 $cgi->a({-href => href(-replay=>1, page=>$page-1),
6214 -accesskey => "p", -title => "Alt-p"}, "prev");
6215 } else {
6216 $paging_nav .= "first &sdot; prev";
6218 my $next_link = '';
6219 if ($#commitlist >= 100) {
6220 $next_link =
6221 $cgi->a({-href => href(-replay=>1, page=>$page+1),
6222 -accesskey => "n", -title => "Alt-n"}, "next");
6223 $paging_nav .= " &sdot; $next_link";
6224 } else {
6225 $paging_nav .= " &sdot; next";
6228 git_header_html();
6230 git_print_page_nav('','', $hash,$co{'tree'},$hash, $paging_nav);
6231 git_print_header_div('commit', esc_html($co{'title'}), $hash);
6232 if ($page == 0 && !@commitlist) {
6233 print "<p>No match.</p>\n";
6234 } else {
6235 git_search_grep_body(\@commitlist, 0, 99, $next_link);
6238 git_footer_html();
6241 sub git_search_changes {
6242 my %co = @_;
6244 local $/ = "\n";
6245 open my $fd, '-|', git_cmd(), '--no-pager', 'log', @diff_opts,
6246 '--pretty=format:%H', '--no-abbrev', '--raw', "-S$searchtext",
6247 ($search_use_regexp ? '--pickaxe-regex' : ())
6248 or die_error(500, "Open git-log failed");
6250 git_header_html();
6252 git_print_page_nav('','', $hash,$co{'tree'},$hash);
6253 git_print_header_div('commit', esc_html($co{'title'}), $hash);
6255 print "<table class=\"pickaxe search\">\n";
6256 my $alternate = 1;
6257 undef %co;
6258 my @files;
6259 while (my $line = <$fd>) {
6260 chomp $line;
6261 next unless $line;
6263 my %set = parse_difftree_raw_line($line);
6264 if (defined $set{'commit'}) {
6265 # finish previous commit
6266 if (%co) {
6267 print "</td>\n" .
6268 "<td class=\"link\">" .
6269 $cgi->a({-href => href(action=>"commit", hash=>$co{'id'})},
6270 "commit") .
6271 " | " .
6272 $cgi->a({-href => href(action=>"tree", hash=>$co{'tree'},
6273 hash_base=>$co{'id'})},
6274 "tree") .
6275 "</td>\n" .
6276 "</tr>\n";
6279 if ($alternate) {
6280 print "<tr class=\"dark\">\n";
6281 } else {
6282 print "<tr class=\"light\">\n";
6284 $alternate ^= 1;
6285 %co = parse_commit($set{'commit'});
6286 my $author = chop_and_escape_str($co{'author_name'}, 15, 5);
6287 print "<td title=\"$co{'age_string_age'}\"><i>$co{'age_string_date'}</i></td>\n" .
6288 "<td><i>$author</i></td>\n" .
6289 "<td>" .
6290 $cgi->a({-href => href(action=>"commit", hash=>$co{'id'}),
6291 -class => "list subject"},
6292 chop_and_escape_str($co{'title'}, 50) . "<br/>");
6293 } elsif (defined $set{'to_id'}) {
6294 next if ($set{'to_id'} =~ m/^0{40}$/);
6296 print $cgi->a({-href => href(action=>"blob", hash_base=>$co{'id'},
6297 hash=>$set{'to_id'}, file_name=>$set{'to_file'}),
6298 -class => "list"},
6299 "<span class=\"match\">" . esc_path($set{'file'}) . "</span>") .
6300 "<br/>\n";
6303 close $fd;
6305 # finish last commit (warning: repetition!)
6306 if (%co) {
6307 print "</td>\n" .
6308 "<td class=\"link\">" .
6309 $cgi->a({-href => href(action=>"commit", hash=>$co{'id'})},
6310 "commit") .
6311 " | " .
6312 $cgi->a({-href => href(action=>"tree", hash=>$co{'tree'},
6313 hash_base=>$co{'id'})},
6314 "tree") .
6315 "</td>\n" .
6316 "</tr>\n";
6319 print "</table>\n";
6321 git_footer_html();
6324 sub git_search_files {
6325 my %co = @_;
6327 local $/ = "\n";
6328 open my $fd, "-|", git_cmd(), 'grep', '-n', '-z',
6329 $search_use_regexp ? ('-E', '-i') : '-F',
6330 $searchtext, $co{'tree'}
6331 or die_error(500, "Open git-grep failed");
6333 git_header_html();
6335 git_print_page_nav('','', $hash,$co{'tree'},$hash);
6336 git_print_header_div('commit', esc_html($co{'title'}), $hash);
6338 print "<table class=\"grep_search\">\n";
6339 my $alternate = 1;
6340 my $matches = 0;
6341 my $lastfile = '';
6342 my $file_href;
6343 while (my $line = <$fd>) {
6344 chomp $line;
6345 my ($file, $lno, $ltext, $binary);
6346 last if ($matches++ > 1000);
6347 if ($line =~ /^Binary file (.+) matches$/) {
6348 $file = $1;
6349 $binary = 1;
6350 } else {
6351 ($file, $lno, $ltext) = split(/\0/, $line, 3);
6352 $file =~ s/^$co{'tree'}://;
6354 if ($file ne $lastfile) {
6355 $lastfile and print "</td></tr>\n";
6356 if ($alternate++) {
6357 print "<tr class=\"dark\">\n";
6358 } else {
6359 print "<tr class=\"light\">\n";
6361 $file_href = href(action=>"blob", hash_base=>$co{'id'},
6362 file_name=>$file);
6363 print "<td class=\"list\">".
6364 $cgi->a({-href => $file_href, -class => "list"}, esc_path($file));
6365 print "</td><td>\n";
6366 $lastfile = $file;
6368 if ($binary) {
6369 print "<div class=\"binary\">Binary file</div>\n";
6370 } else {
6371 $ltext = untabify($ltext);
6372 if ($ltext =~ m/^(.*)($search_regexp)(.*)$/i) {
6373 $ltext = esc_html($1, -nbsp=>1);
6374 $ltext .= '<span class="match">';
6375 $ltext .= esc_html($2, -nbsp=>1);
6376 $ltext .= '</span>';
6377 $ltext .= esc_html($3, -nbsp=>1);
6378 } else {
6379 $ltext = esc_html($ltext, -nbsp=>1);
6381 print "<div class=\"pre\">" .
6382 $cgi->a({-href => $file_href.'#l'.$lno,
6383 -class => "linenr"}, sprintf('%4i', $lno)) .
6384 ' ' . $ltext . "</div>\n";
6387 if ($lastfile) {
6388 print "</td></tr>\n";
6389 if ($matches > 1000) {
6390 print "<div class=\"diff nodifferences\">Too many matches, listing trimmed</div>\n";
6392 } else {
6393 print "<div class=\"diff nodifferences\">No matches found</div>\n";
6395 close $fd;
6397 print "</table>\n";
6399 git_footer_html();
6402 sub git_search_grep_body {
6403 my ($commitlist, $from, $to, $extra) = @_;
6404 $from = 0 unless defined $from;
6405 $to = $#{$commitlist} if (!defined $to || $#{$commitlist} < $to);
6407 print "<table class=\"commit_search\">\n";
6408 my $alternate = 1;
6409 for (my $i = $from; $i <= $to; $i++) {
6410 my %co = %{$commitlist->[$i]};
6411 if (!%co) {
6412 next;
6414 my $commit = $co{'id'};
6415 if ($alternate) {
6416 print "<tr class=\"dark\">\n";
6417 } else {
6418 print "<tr class=\"light\">\n";
6420 $alternate ^= 1;
6421 print "<td title=\"$co{'age_string_age'}\"><i>$co{'age_string_date'}</i></td>\n" .
6422 format_author_html('td', \%co, 15, 5) .
6423 "<td>" .
6424 $cgi->a({-href => href(action=>"commit", hash=>$co{'id'}),
6425 -class => "list subject"},
6426 chop_and_escape_str($co{'title'}, 50) . "<br/>");
6427 my $comment = $co{'comment'};
6428 foreach my $line (@$comment) {
6429 if ($line =~ m/^(.*?)($search_regexp)(.*)$/i) {
6430 my ($lead, $match, $trail) = ($1, $2, $3);
6431 $match = chop_str($match, 70, 5, 'center');
6432 my $contextlen = int((80 - length($match))/2);
6433 $contextlen = 30 if ($contextlen > 30);
6434 $lead = chop_str($lead, $contextlen, 10, 'left');
6435 $trail = chop_str($trail, $contextlen, 10, 'right');
6437 $lead = esc_html($lead);
6438 $match = esc_html($match);
6439 $trail = esc_html($trail);
6441 print "$lead<span class=\"match\">$match</span>$trail<br />";
6444 print "</td>\n" .
6445 "<td class=\"link\">" .
6446 $cgi->a({-href => href(action=>"commit", hash=>$co{'id'})}, "commit") .
6447 " | " .
6448 $cgi->a({-href => href(action=>"commitdiff", hash=>$co{'id'})}, "commitdiff") .
6449 " | " .
6450 $cgi->a({-href => href(action=>"tree", hash=>$co{'tree'}, hash_base=>$co{'id'})}, "tree");
6451 print "</td>\n" .
6452 "</tr>\n";
6454 if (defined $extra) {
6455 print "<tr>\n" .
6456 "<td colspan=\"3\">$extra</td>\n" .
6457 "</tr>\n";
6459 print "</table>\n";
6462 ## ======================================================================
6463 ## ======================================================================
6464 ## actions
6466 sub git_project_list {
6467 my $order = $input_params{'order'};
6468 if (defined $order && $order !~ m/none|project|descr|owner|age/) {
6469 die_error(400, "Unknown order parameter");
6472 my @list = git_get_projects_list($project_filter, $strict_export);
6473 if (!@list) {
6474 die_error(404, "No projects found");
6477 git_header_html();
6478 if (defined $home_text && -f $home_text) {
6479 print "<div class=\"index_include\">\n";
6480 insert_file($home_text);
6481 print "</div>\n";
6484 git_project_search_form($searchtext, $search_use_regexp);
6485 git_project_list_body(\@list, $order);
6486 git_footer_html();
6489 sub git_forks {
6490 my $order = $input_params{'order'};
6491 if (defined $order && $order !~ m/none|project|descr|owner|age/) {
6492 die_error(400, "Unknown order parameter");
6495 my $filter = $project;
6496 $filter =~ s/\.git$//;
6497 my @list = git_get_projects_list($filter);
6498 if (!@list) {
6499 die_error(404, "No forks found");
6502 git_header_html();
6503 git_print_page_nav('','');
6504 git_print_header_div('summary', "$project forks");
6505 git_project_list_body(\@list, $order);
6506 git_footer_html();
6509 sub git_project_index {
6510 my @projects = git_get_projects_list($project_filter, $strict_export);
6511 if (!@projects) {
6512 die_error(404, "No projects found");
6515 print $cgi->header(
6516 -type => 'text/plain',
6517 -charset => 'utf-8',
6518 -content_disposition => 'inline; filename="index.aux"');
6520 foreach my $pr (@projects) {
6521 if (!exists $pr->{'owner'}) {
6522 $pr->{'owner'} = git_get_project_owner("$pr->{'path'}");
6525 my ($path, $owner) = ($pr->{'path'}, $pr->{'owner'});
6526 # quote as in CGI::Util::encode, but keep the slash, and use '+' for ' '
6527 $path =~ s/([^a-zA-Z0-9_.\-\/ ])/sprintf("%%%02X", ord($1))/eg;
6528 $owner =~ s/([^a-zA-Z0-9_.\-\/ ])/sprintf("%%%02X", ord($1))/eg;
6529 $path =~ s/ /\+/g;
6530 $owner =~ s/ /\+/g;
6532 print "$path $owner\n";
6536 sub git_summary {
6537 my $descr = git_get_project_description($project) || "none";
6538 my %co = parse_commit("HEAD");
6539 my %cd = %co ? parse_date($co{'committer_epoch'}, $co{'committer_tz'}) : ();
6540 my $head = $co{'id'};
6541 my $remote_heads = gitweb_check_feature('remote_heads');
6543 my $owner = git_get_project_owner($project);
6545 my $refs = git_get_references();
6546 # These get_*_list functions return one more to allow us to see if
6547 # there are more ...
6548 my @taglist = git_get_tags_list(16);
6549 my @headlist = git_get_heads_list(16);
6550 my %remotedata = $remote_heads ? git_get_remotes_list() : ();
6551 my @forklist;
6552 my $check_forks = gitweb_check_feature('forks');
6554 if ($check_forks) {
6555 # find forks of a project
6556 my $filter = $project;
6557 $filter =~ s/\.git$//;
6558 @forklist = git_get_projects_list($filter);
6559 # filter out forks of forks
6560 @forklist = filter_forks_from_projects_list(\@forklist)
6561 if (@forklist);
6564 git_header_html();
6565 git_print_page_nav('summary','', $head);
6567 print "<div class=\"title\">&nbsp;</div>\n";
6568 print "<table class=\"projects_list\">\n" .
6569 "<tr id=\"metadata_desc\"><td>description</td><td>" . esc_html($descr) . "</td></tr>\n";
6570 if ($owner and not $omit_owner) {
6571 print "<tr id=\"metadata_owner\"><td>owner</td><td>" . esc_html($owner) . "</td></tr>\n";
6573 if (defined $cd{'rfc2822'}) {
6574 print "<tr id=\"metadata_lchange\"><td>last change</td>" .
6575 "<td>".format_timestamp_html(\%cd)."</td></tr>\n";
6578 # use per project git URL list in $projectroot/$project/cloneurl
6579 # or make project git URL from git base URL and project name
6580 my $url_tag = "URL";
6581 my @url_list = git_get_project_url_list($project);
6582 @url_list = map { "$_/$project" } @git_base_url_list unless @url_list;
6583 foreach my $git_url (@url_list) {
6584 next unless $git_url;
6585 print format_repo_url($url_tag, $git_url);
6586 $url_tag = "";
6589 # Tag cloud
6590 my $show_ctags = gitweb_check_feature('ctags');
6591 if ($show_ctags) {
6592 my $ctags = git_get_project_ctags($project);
6593 if (%$ctags) {
6594 # without ability to add tags, don't show if there are none
6595 my $cloud = git_populate_project_tagcloud($ctags);
6596 print "<tr id=\"metadata_ctags\">" .
6597 "<td>content tags</td>" .
6598 "<td>".git_show_project_tagcloud($cloud, 48)."</td>" .
6599 "</tr>\n";
6603 print "</table>\n";
6605 # If XSS prevention is on, we don't include README.html.
6606 # TODO: Allow a readme in some safe format.
6607 if (!$prevent_xss && -s "$projectroot/$project/README.html") {
6608 print "<div class=\"title\">readme</div>\n" .
6609 "<div class=\"readme\">\n";
6610 insert_file("$projectroot/$project/README.html");
6611 print "\n</div>\n"; # class="readme"
6614 # we need to request one more than 16 (0..15) to check if
6615 # those 16 are all
6616 my @commitlist = $head ? parse_commits($head, 17) : ();
6617 if (@commitlist) {
6618 git_print_header_div('shortlog');
6619 git_shortlog_body(\@commitlist, 0, 15, $refs,
6620 $#commitlist <= 15 ? undef :
6621 $cgi->a({-href => href(action=>"shortlog")}, "..."));
6624 if (@taglist) {
6625 git_print_header_div('tags');
6626 git_tags_body(\@taglist, 0, 15,
6627 $#taglist <= 15 ? undef :
6628 $cgi->a({-href => href(action=>"tags")}, "..."));
6631 if (@headlist) {
6632 git_print_header_div('heads');
6633 git_heads_body(\@headlist, $head, 0, 15,
6634 $#headlist <= 15 ? undef :
6635 $cgi->a({-href => href(action=>"heads")}, "..."));
6638 if (%remotedata) {
6639 git_print_header_div('remotes');
6640 git_remotes_body(\%remotedata, 15, $head);
6643 if (@forklist) {
6644 git_print_header_div('forks');
6645 git_project_list_body(\@forklist, 'age', 0, 15,
6646 $#forklist <= 15 ? undef :
6647 $cgi->a({-href => href(action=>"forks")}, "..."),
6648 'no_header');
6651 git_footer_html();
6654 sub git_tag {
6655 my %tag = parse_tag($hash);
6657 if (! %tag) {
6658 die_error(404, "Unknown tag object");
6661 my $head = git_get_head_hash($project);
6662 git_header_html();
6663 git_print_page_nav('','', $head,undef,$head);
6664 git_print_header_div('commit', esc_html($tag{'name'}), $hash);
6665 print "<div class=\"title_text\">\n" .
6666 "<table class=\"object_header\">\n" .
6667 "<tr>\n" .
6668 "<td>object</td>\n" .
6669 "<td>" . $cgi->a({-class => "list", -href => href(action=>$tag{'type'}, hash=>$tag{'object'})},
6670 $tag{'object'}) . "</td>\n" .
6671 "<td class=\"link\">" . $cgi->a({-href => href(action=>$tag{'type'}, hash=>$tag{'object'})},
6672 $tag{'type'}) . "</td>\n" .
6673 "</tr>\n";
6674 if (defined($tag{'author'})) {
6675 git_print_authorship_rows(\%tag, 'author');
6677 print "</table>\n\n" .
6678 "</div>\n";
6679 print "<div class=\"page_body\">";
6680 my $comment = $tag{'comment'};
6681 foreach my $line (@$comment) {
6682 chomp $line;
6683 print esc_html($line, -nbsp=>1) . "<br/>\n";
6685 print "</div>\n";
6686 git_footer_html();
6689 sub git_blame_common {
6690 my $format = shift || 'porcelain';
6691 if ($format eq 'porcelain' && $input_params{'javascript'}) {
6692 $format = 'incremental';
6693 $action = 'blame_incremental'; # for page title etc
6696 # permissions
6697 gitweb_check_feature('blame')
6698 or die_error(403, "Blame view not allowed");
6700 # error checking
6701 die_error(400, "No file name given") unless $file_name;
6702 $hash_base ||= git_get_head_hash($project);
6703 die_error(404, "Couldn't find base commit") unless $hash_base;
6704 my %co = parse_commit($hash_base)
6705 or die_error(404, "Commit not found");
6706 my $ftype = "blob";
6707 if (!defined $hash) {
6708 $hash = git_get_hash_by_path($hash_base, $file_name, "blob")
6709 or die_error(404, "Error looking up file");
6710 } else {
6711 $ftype = git_get_type($hash);
6712 if ($ftype !~ "blob") {
6713 die_error(400, "Object is not a blob");
6717 my $fd;
6718 if ($format eq 'incremental') {
6719 # get file contents (as base)
6720 open $fd, "-|", git_cmd(), 'cat-file', 'blob', $hash
6721 or die_error(500, "Open git-cat-file failed");
6722 } elsif ($format eq 'data') {
6723 # run git-blame --incremental
6724 open $fd, "-|", git_cmd(), "blame", "--incremental",
6725 $hash_base, "--", $file_name
6726 or die_error(500, "Open git-blame --incremental failed");
6727 } else {
6728 # run git-blame --porcelain
6729 open $fd, "-|", git_cmd(), "blame", '-p',
6730 $hash_base, '--', $file_name
6731 or die_error(500, "Open git-blame --porcelain failed");
6733 binmode $fd, ':utf8';
6735 # incremental blame data returns early
6736 if ($format eq 'data') {
6737 print $cgi->header(
6738 -type=>"text/plain", -charset => "utf-8",
6739 -status=> "200 OK");
6740 local $| = 1; # output autoflush
6741 while (my $line = <$fd>) {
6742 print to_utf8($line);
6744 close $fd
6745 or print "ERROR $!\n";
6747 print 'END';
6748 if (defined $t0 && gitweb_check_feature('timed')) {
6749 print ' '.
6750 tv_interval($t0, [ gettimeofday() ]).
6751 ' '.$number_of_git_cmds;
6753 print "\n";
6755 return;
6758 # page header
6759 git_header_html();
6760 my $formats_nav =
6761 $cgi->a({-href => href(action=>"blob", -replay=>1)},
6762 "blob") .
6763 " | ";
6764 if ($format eq 'incremental') {
6765 $formats_nav .=
6766 $cgi->a({-href => href(action=>"blame", javascript=>0, -replay=>1)},
6767 "blame") . " (non-incremental)";
6768 } else {
6769 $formats_nav .=
6770 $cgi->a({-href => href(action=>"blame_incremental", -replay=>1)},
6771 "blame") . " (incremental)";
6773 $formats_nav .=
6774 " | " .
6775 $cgi->a({-href => href(action=>"history", -replay=>1)},
6776 "history") .
6777 " | " .
6778 $cgi->a({-href => href(action=>$action, file_name=>$file_name)},
6779 "HEAD");
6780 git_print_page_nav('','', $hash_base,$co{'tree'},$hash_base, $formats_nav);
6781 git_print_header_div('commit', esc_html($co{'title'}), $hash_base);
6782 git_print_page_path($file_name, $ftype, $hash_base);
6784 # page body
6785 if ($format eq 'incremental') {
6786 print "<noscript>\n<div class=\"error\"><center><b>\n".
6787 "This page requires JavaScript to run.\n Use ".
6788 $cgi->a({-href => href(action=>'blame',javascript=>0,-replay=>1)},
6789 'this page').
6790 " instead.\n".
6791 "</b></center></div>\n</noscript>\n";
6793 print qq!<div id="progress_bar" style="width: 100%; background-color: yellow"></div>\n!;
6796 print qq!<div class="page_body">\n!;
6797 print qq!<div id="progress_info">... / ...</div>\n!
6798 if ($format eq 'incremental');
6799 print qq!<table id="blame_table" class="blame" width="100%">\n!.
6800 #qq!<col width="5.5em" /><col width="2.5em" /><col width="*" />\n!.
6801 qq!<thead>\n!.
6802 qq!<tr><th>Commit</th><th>Line</th><th>Data</th></tr>\n!.
6803 qq!</thead>\n!.
6804 qq!<tbody>\n!;
6806 my @rev_color = qw(light dark);
6807 my $num_colors = scalar(@rev_color);
6808 my $current_color = 0;
6810 if ($format eq 'incremental') {
6811 my $color_class = $rev_color[$current_color];
6813 #contents of a file
6814 my $linenr = 0;
6815 LINE:
6816 while (my $line = <$fd>) {
6817 chomp $line;
6818 $linenr++;
6820 print qq!<tr id="l$linenr" class="$color_class">!.
6821 qq!<td class="sha1"><a href=""> </a></td>!.
6822 qq!<td class="linenr">!.
6823 qq!<a class="linenr" href="">$linenr</a></td>!;
6824 print qq!<td class="pre">! . esc_html($line) . "</td>\n";
6825 print qq!</tr>\n!;
6828 } else { # porcelain, i.e. ordinary blame
6829 my %metainfo = (); # saves information about commits
6831 # blame data
6832 LINE:
6833 while (my $line = <$fd>) {
6834 chomp $line;
6835 # the header: <SHA-1> <src lineno> <dst lineno> [<lines in group>]
6836 # no <lines in group> for subsequent lines in group of lines
6837 my ($full_rev, $orig_lineno, $lineno, $group_size) =
6838 ($line =~ /^([0-9a-f]{40}) (\d+) (\d+)(?: (\d+))?$/);
6839 if (!exists $metainfo{$full_rev}) {
6840 $metainfo{$full_rev} = { 'nprevious' => 0 };
6842 my $meta = $metainfo{$full_rev};
6843 my $data;
6844 while ($data = <$fd>) {
6845 chomp $data;
6846 last if ($data =~ s/^\t//); # contents of line
6847 if ($data =~ /^(\S+)(?: (.*))?$/) {
6848 $meta->{$1} = $2 unless exists $meta->{$1};
6850 if ($data =~ /^previous /) {
6851 $meta->{'nprevious'}++;
6854 my $short_rev = substr($full_rev, 0, 8);
6855 my $author = $meta->{'author'};
6856 my %date =
6857 parse_date($meta->{'author-time'}, $meta->{'author-tz'});
6858 my $date = $date{'iso-tz'};
6859 if ($group_size) {
6860 $current_color = ($current_color + 1) % $num_colors;
6862 my $tr_class = $rev_color[$current_color];
6863 $tr_class .= ' boundary' if (exists $meta->{'boundary'});
6864 $tr_class .= ' no-previous' if ($meta->{'nprevious'} == 0);
6865 $tr_class .= ' multiple-previous' if ($meta->{'nprevious'} > 1);
6866 print "<tr id=\"l$lineno\" class=\"$tr_class\">\n";
6867 if ($group_size) {
6868 print "<td class=\"sha1\"";
6869 print " title=\"". esc_html($author) . ", $date\"";
6870 print " rowspan=\"$group_size\"" if ($group_size > 1);
6871 print ">";
6872 print $cgi->a({-href => href(action=>"commit",
6873 hash=>$full_rev,
6874 file_name=>$file_name)},
6875 esc_html($short_rev));
6876 if ($group_size >= 2) {
6877 my @author_initials = ($author =~ /\b([[:upper:]])\B/g);
6878 if (@author_initials) {
6879 print "<br />" .
6880 esc_html(join('', @author_initials));
6881 # or join('.', ...)
6884 print "</td>\n";
6886 # 'previous' <sha1 of parent commit> <filename at commit>
6887 if (exists $meta->{'previous'} &&
6888 $meta->{'previous'} =~ /^([a-fA-F0-9]{40}) (.*)$/) {
6889 $meta->{'parent'} = $1;
6890 $meta->{'file_parent'} = unquote($2);
6892 my $linenr_commit =
6893 exists($meta->{'parent'}) ?
6894 $meta->{'parent'} : $full_rev;
6895 my $linenr_filename =
6896 exists($meta->{'file_parent'}) ?
6897 $meta->{'file_parent'} : unquote($meta->{'filename'});
6898 my $blamed = href(action => 'blame',
6899 file_name => $linenr_filename,
6900 hash_base => $linenr_commit);
6901 print "<td class=\"linenr\">";
6902 print $cgi->a({ -href => "$blamed#l$orig_lineno",
6903 -class => "linenr" },
6904 esc_html($lineno));
6905 print "</td>";
6906 print "<td class=\"pre\">" . esc_html($data) . "</td>\n";
6907 print "</tr>\n";
6908 } # end while
6912 # footer
6913 print "</tbody>\n".
6914 "</table>\n"; # class="blame"
6915 print "</div>\n"; # class="blame_body"
6916 close $fd
6917 or print "Reading blob failed\n";
6919 git_footer_html();
6922 sub git_blame {
6923 git_blame_common();
6926 sub git_blame_incremental {
6927 git_blame_common('incremental');
6930 sub git_blame_data {
6931 git_blame_common('data');
6934 sub git_tags {
6935 my $head = git_get_head_hash($project);
6936 git_header_html();
6937 git_print_page_nav('','', $head,undef,$head,format_ref_views('tags'));
6938 git_print_header_div('summary', $project);
6940 my @tagslist = git_get_tags_list();
6941 if (@tagslist) {
6942 git_tags_body(\@tagslist);
6944 git_footer_html();
6947 sub git_heads {
6948 my $head = git_get_head_hash($project);
6949 git_header_html();
6950 git_print_page_nav('','', $head,undef,$head,format_ref_views('heads'));
6951 git_print_header_div('summary', $project);
6953 my @headslist = git_get_heads_list();
6954 if (@headslist) {
6955 git_heads_body(\@headslist, $head);
6957 git_footer_html();
6960 # used both for single remote view and for list of all the remotes
6961 sub git_remotes {
6962 gitweb_check_feature('remote_heads')
6963 or die_error(403, "Remote heads view is disabled");
6965 my $head = git_get_head_hash($project);
6966 my $remote = $input_params{'hash'};
6968 my $remotedata = git_get_remotes_list($remote);
6969 die_error(500, "Unable to get remote information") unless defined $remotedata;
6971 unless (%$remotedata) {
6972 die_error(404, defined $remote ?
6973 "Remote $remote not found" :
6974 "No remotes found");
6977 git_header_html(undef, undef, -action_extra => $remote);
6978 git_print_page_nav('', '', $head, undef, $head,
6979 format_ref_views($remote ? '' : 'remotes'));
6981 fill_remote_heads($remotedata);
6982 if (defined $remote) {
6983 git_print_header_div('remotes', "$remote remote for $project");
6984 git_remote_block($remote, $remotedata->{$remote}, undef, $head);
6985 } else {
6986 git_print_header_div('summary', "$project remotes");
6987 git_remotes_body($remotedata, undef, $head);
6990 git_footer_html();
6993 sub git_blob_plain {
6994 my $type = shift;
6995 my $expires;
6997 if (!defined $hash) {
6998 if (defined $file_name) {
6999 my $base = $hash_base || git_get_head_hash($project);
7000 $hash = git_get_hash_by_path($base, $file_name, "blob")
7001 or die_error(404, "Cannot find file");
7002 } else {
7003 die_error(400, "No file name defined");
7005 } elsif ($hash =~ m/^[0-9a-fA-F]{40}$/) {
7006 # blobs defined by non-textual hash id's can be cached
7007 $expires = "+1d";
7010 open my $fd, "-|", git_cmd(), "cat-file", "blob", $hash
7011 or die_error(500, "Open git-cat-file blob '$hash' failed");
7013 # content-type (can include charset)
7014 $type = blob_contenttype($fd, $file_name, $type);
7016 # "save as" filename, even when no $file_name is given
7017 my $save_as = "$hash";
7018 if (defined $file_name) {
7019 $save_as = $file_name;
7020 } elsif ($type =~ m/^text\//) {
7021 $save_as .= '.txt';
7024 # With XSS prevention on, blobs of all types except a few known safe
7025 # ones are served with "Content-Disposition: attachment" to make sure
7026 # they don't run in our security domain. For certain image types,
7027 # blob view writes an <img> tag referring to blob_plain view, and we
7028 # want to be sure not to break that by serving the image as an
7029 # attachment (though Firefox 3 doesn't seem to care).
7030 my $sandbox = $prevent_xss &&
7031 $type !~ m!^(?:text/[a-z]+|image/(?:gif|png|jpeg))(?:[ ;]|$)!;
7033 # serve text/* as text/plain
7034 if ($prevent_xss &&
7035 ($type =~ m!^text/[a-z]+\b(.*)$! ||
7036 ($type =~ m!^[a-z]+/[a-z]\+xml\b(.*)$! && -T $fd))) {
7037 my $rest = $1;
7038 $rest = defined $rest ? $rest : '';
7039 $type = "text/plain$rest";
7042 print $cgi->header(
7043 -type => $type,
7044 -expires => $expires,
7045 -content_disposition =>
7046 ($sandbox ? 'attachment' : 'inline')
7047 . '; filename="' . $save_as . '"');
7048 local $/ = undef;
7049 binmode STDOUT, ':raw';
7050 print <$fd>;
7051 binmode STDOUT, ':utf8'; # as set at the beginning of gitweb.cgi
7052 close $fd;
7055 sub git_blob {
7056 my $expires;
7058 if (!defined $hash) {
7059 if (defined $file_name) {
7060 my $base = $hash_base || git_get_head_hash($project);
7061 $hash = git_get_hash_by_path($base, $file_name, "blob")
7062 or die_error(404, "Cannot find file");
7063 } else {
7064 die_error(400, "No file name defined");
7066 } elsif ($hash =~ m/^[0-9a-fA-F]{40}$/) {
7067 # blobs defined by non-textual hash id's can be cached
7068 $expires = "+1d";
7071 my $have_blame = gitweb_check_feature('blame');
7072 open my $fd, "-|", git_cmd(), "cat-file", "blob", $hash
7073 or die_error(500, "Couldn't cat $file_name, $hash");
7074 my $mimetype = blob_mimetype($fd, $file_name);
7075 # use 'blob_plain' (aka 'raw') view for files that cannot be displayed
7076 if ($mimetype !~ m!^(?:text/|image/(?:gif|png|jpeg)$)! && -B $fd) {
7077 close $fd;
7078 return git_blob_plain($mimetype);
7080 # we can have blame only for text/* mimetype
7081 $have_blame &&= ($mimetype =~ m!^text/!);
7083 my $highlight = gitweb_check_feature('highlight');
7084 my $syntax = guess_file_syntax($highlight, $file_name);
7085 $fd = run_highlighter($fd, $highlight, $syntax);
7087 git_header_html(undef, $expires);
7088 my $formats_nav = '';
7089 if (defined $hash_base && (my %co = parse_commit($hash_base))) {
7090 if (defined $file_name) {
7091 if ($have_blame) {
7092 $formats_nav .=
7093 $cgi->a({-href => href(action=>"blame", -replay=>1)},
7094 "blame") .
7095 " | ";
7097 $formats_nav .=
7098 $cgi->a({-href => href(action=>"history", -replay=>1)},
7099 "history") .
7100 " | " .
7101 $cgi->a({-href => href(action=>"blob_plain", -replay=>1)},
7102 "raw") .
7103 " | " .
7104 $cgi->a({-href => href(action=>"blob",
7105 hash_base=>"HEAD", file_name=>$file_name)},
7106 "HEAD");
7107 } else {
7108 $formats_nav .=
7109 $cgi->a({-href => href(action=>"blob_plain", -replay=>1)},
7110 "raw");
7112 git_print_page_nav('','', $hash_base,$co{'tree'},$hash_base, $formats_nav);
7113 git_print_header_div('commit', esc_html($co{'title'}), $hash_base);
7114 } else {
7115 print "<div class=\"page_nav\">\n" .
7116 "<br/><br/></div>\n" .
7117 "<div class=\"title\">".esc_html($hash)."</div>\n";
7119 git_print_page_path($file_name, "blob", $hash_base);
7120 print "<div class=\"page_body\">\n";
7121 if ($mimetype =~ m!^image/!) {
7122 print qq!<img class="blob" type="!.esc_attr($mimetype).qq!"!;
7123 if ($file_name) {
7124 print qq! alt="!.esc_attr($file_name).qq!" title="!.esc_attr($file_name).qq!"!;
7126 print qq! src="! .
7127 href(action=>"blob_plain", hash=>$hash,
7128 hash_base=>$hash_base, file_name=>$file_name) .
7129 qq!" />\n!;
7130 } else {
7131 my $nr;
7132 while (my $line = <$fd>) {
7133 chomp $line;
7134 $nr++;
7135 $line = untabify($line);
7136 printf qq!<div class="pre"><a id="l%i" href="%s#l%i" class="linenr">%4i</a> %s</div>\n!,
7137 $nr, esc_attr(href(-replay => 1)), $nr, $nr,
7138 $highlight ? sanitize($line) : esc_html($line, -nbsp=>1);
7141 close $fd
7142 or print "Reading blob failed.\n";
7143 print "</div>";
7144 git_footer_html();
7147 sub git_tree {
7148 if (!defined $hash_base) {
7149 $hash_base = "HEAD";
7151 if (!defined $hash) {
7152 if (defined $file_name) {
7153 $hash = git_get_hash_by_path($hash_base, $file_name, "tree");
7154 } else {
7155 $hash = $hash_base;
7158 die_error(404, "No such tree") unless defined($hash);
7160 my $show_sizes = gitweb_check_feature('show-sizes');
7161 my $have_blame = gitweb_check_feature('blame');
7163 my @entries = ();
7165 local $/ = "\0";
7166 open my $fd, "-|", git_cmd(), "ls-tree", '-z',
7167 ($show_sizes ? '-l' : ()), @extra_options, $hash
7168 or die_error(500, "Open git-ls-tree failed");
7169 @entries = map { chomp; $_ } <$fd>;
7170 close $fd
7171 or die_error(404, "Reading tree failed");
7174 my $refs = git_get_references();
7175 my $ref = format_ref_marker($refs, $hash_base);
7176 git_header_html();
7177 my $basedir = '';
7178 if (defined $hash_base && (my %co = parse_commit($hash_base))) {
7179 my @views_nav = ();
7180 if (defined $file_name) {
7181 push @views_nav,
7182 $cgi->a({-href => href(action=>"history", -replay=>1)},
7183 "history"),
7184 $cgi->a({-href => href(action=>"tree",
7185 hash_base=>"HEAD", file_name=>$file_name)},
7186 "HEAD"),
7188 my $snapshot_links = format_snapshot_links($hash);
7189 if (defined $snapshot_links) {
7190 # FIXME: Should be available when we have no hash base as well.
7191 push @views_nav, $snapshot_links;
7193 git_print_page_nav('tree','', $hash_base, undef, undef,
7194 join(' | ', @views_nav));
7195 git_print_header_div('commit', esc_html($co{'title'}) . $ref, $hash_base);
7196 } else {
7197 undef $hash_base;
7198 print "<div class=\"page_nav\">\n";
7199 print "<br/><br/></div>\n";
7200 print "<div class=\"title\">".esc_html($hash)."</div>\n";
7202 if (defined $file_name) {
7203 $basedir = $file_name;
7204 if ($basedir ne '' && substr($basedir, -1) ne '/') {
7205 $basedir .= '/';
7207 git_print_page_path($file_name, 'tree', $hash_base);
7209 print "<div class=\"page_body\">\n";
7210 print "<table class=\"tree\">\n";
7211 my $alternate = 1;
7212 # '..' (top directory) link if possible
7213 if (defined $hash_base &&
7214 defined $file_name && $file_name =~ m![^/]+$!) {
7215 if ($alternate) {
7216 print "<tr class=\"dark\">\n";
7217 } else {
7218 print "<tr class=\"light\">\n";
7220 $alternate ^= 1;
7222 my $up = $file_name;
7223 $up =~ s!/?[^/]+$!!;
7224 undef $up unless $up;
7225 # based on git_print_tree_entry
7226 print '<td class="mode">' . mode_str('040000') . "</td>\n";
7227 print '<td class="size">&nbsp;</td>'."\n" if $show_sizes;
7228 print '<td class="list">';
7229 print $cgi->a({-href => href(action=>"tree",
7230 hash_base=>$hash_base,
7231 file_name=>$up)},
7232 "..");
7233 print "</td>\n";
7234 print "<td class=\"link\"></td>\n";
7236 print "</tr>\n";
7238 foreach my $line (@entries) {
7239 my %t = parse_ls_tree_line($line, -z => 1, -l => $show_sizes);
7241 if ($alternate) {
7242 print "<tr class=\"dark\">\n";
7243 } else {
7244 print "<tr class=\"light\">\n";
7246 $alternate ^= 1;
7248 git_print_tree_entry(\%t, $basedir, $hash_base, $have_blame);
7250 print "</tr>\n";
7252 print "</table>\n" .
7253 "</div>";
7254 git_footer_html();
7257 sub sanitize_for_filename {
7258 my $name = shift;
7260 $name =~ s!/!-!g;
7261 $name =~ s/[^[:alnum:]_.-]//g;
7263 return $name;
7266 sub snapshot_name {
7267 my ($project, $hash) = @_;
7269 # path/to/project.git -> project
7270 # path/to/project/.git -> project
7271 my $name = to_utf8($project);
7272 $name =~ s,([^/])/*\.git$,$1,;
7273 $name = sanitize_for_filename(basename($name));
7275 my $ver = $hash;
7276 if ($hash =~ /^[0-9a-fA-F]+$/) {
7277 # shorten SHA-1 hash
7278 my $full_hash = git_get_full_hash($project, $hash);
7279 if ($full_hash =~ /^$hash/ && length($hash) > 7) {
7280 $ver = git_get_short_hash($project, $hash);
7282 } elsif ($hash =~ m!^refs/tags/(.*)$!) {
7283 # tags don't need shortened SHA-1 hash
7284 $ver = $1;
7285 } else {
7286 # branches and other need shortened SHA-1 hash
7287 my $strip_refs = join '|', map { quotemeta } get_branch_refs();
7288 if ($hash =~ m!^refs/($strip_refs|remotes)/(.*)$!) {
7289 my $ref_dir = (defined $1) ? $1 : '';
7290 $ver = $2;
7292 $ref_dir = sanitize_for_filename($ref_dir);
7293 # for refs neither in heads nor remotes we want to
7294 # add a ref dir to archive name
7295 if ($ref_dir ne '' and $ref_dir ne 'heads' and $ref_dir ne 'remotes') {
7296 $ver = $ref_dir . '-' . $ver;
7299 $ver .= '-' . git_get_short_hash($project, $hash);
7301 # special case of sanitization for filename - we change
7302 # slashes to dots instead of dashes
7303 # in case of hierarchical branch names
7304 $ver =~ s!/!.!g;
7305 $ver =~ s/[^[:alnum:]_.-]//g;
7307 # name = project-version_string
7308 $name = "$name-$ver";
7310 return wantarray ? ($name, $name) : $name;
7313 sub exit_if_unmodified_since {
7314 my ($latest_epoch) = @_;
7315 our $cgi;
7317 my $if_modified = $cgi->http('IF_MODIFIED_SINCE');
7318 if (defined $if_modified) {
7319 my $since;
7320 if (eval { require HTTP::Date; 1; }) {
7321 $since = HTTP::Date::str2time($if_modified);
7322 } elsif (eval { require Time::ParseDate; 1; }) {
7323 $since = Time::ParseDate::parsedate($if_modified, GMT => 1);
7325 if (defined $since && $latest_epoch <= $since) {
7326 my %latest_date = parse_date($latest_epoch);
7327 print $cgi->header(
7328 -last_modified => $latest_date{'rfc2822'},
7329 -status => '304 Not Modified');
7330 goto DONE_GITWEB;
7335 sub git_snapshot {
7336 my $format = $input_params{'snapshot_format'};
7337 if (!@snapshot_fmts) {
7338 die_error(403, "Snapshots not allowed");
7340 # default to first supported snapshot format
7341 $format ||= $snapshot_fmts[0];
7342 if ($format !~ m/^[a-z0-9]+$/) {
7343 die_error(400, "Invalid snapshot format parameter");
7344 } elsif (!exists($known_snapshot_formats{$format})) {
7345 die_error(400, "Unknown snapshot format");
7346 } elsif ($known_snapshot_formats{$format}{'disabled'}) {
7347 die_error(403, "Snapshot format not allowed");
7348 } elsif (!grep($_ eq $format, @snapshot_fmts)) {
7349 die_error(403, "Unsupported snapshot format");
7352 my $type = git_get_type("$hash^{}");
7353 if (!$type) {
7354 die_error(404, 'Object does not exist');
7355 } elsif ($type eq 'blob') {
7356 die_error(400, 'Object is not a tree-ish');
7359 my ($name, $prefix) = snapshot_name($project, $hash);
7360 my $filename = "$name$known_snapshot_formats{$format}{'suffix'}";
7362 my %co = parse_commit($hash);
7363 exit_if_unmodified_since($co{'committer_epoch'}) if %co;
7365 my $cmd = quote_command(
7366 git_cmd(), 'archive',
7367 "--format=$known_snapshot_formats{$format}{'format'}",
7368 "--prefix=$prefix/", $hash);
7369 if (exists $known_snapshot_formats{$format}{'compressor'}) {
7370 $cmd .= ' | ' . quote_command(@{$known_snapshot_formats{$format}{'compressor'}});
7373 $filename =~ s/(["\\])/\\$1/g;
7374 my %latest_date;
7375 if (%co) {
7376 %latest_date = parse_date($co{'committer_epoch'}, $co{'committer_tz'});
7379 print $cgi->header(
7380 -type => $known_snapshot_formats{$format}{'type'},
7381 -content_disposition => 'inline; filename="' . $filename . '"',
7382 %co ? (-last_modified => $latest_date{'rfc2822'}) : (),
7383 -status => '200 OK');
7385 open my $fd, "-|", $cmd
7386 or die_error(500, "Execute git-archive failed");
7387 binmode STDOUT, ':raw';
7388 print <$fd>;
7389 binmode STDOUT, ':utf8'; # as set at the beginning of gitweb.cgi
7390 close $fd;
7393 sub git_log_generic {
7394 my ($fmt_name, $body_subr, $base, $parent, $file_name, $file_hash) = @_;
7396 my $head = git_get_head_hash($project);
7397 if (!defined $base) {
7398 $base = $head;
7400 if (!defined $page) {
7401 $page = 0;
7403 my $refs = git_get_references();
7405 my $commit_hash = $base;
7406 if (defined $parent) {
7407 $commit_hash = "$parent..$base";
7409 my @commitlist =
7410 parse_commits($commit_hash, 101, (100 * $page),
7411 defined $file_name ? ($file_name, "--full-history") : ());
7413 my $ftype;
7414 if (!defined $file_hash && defined $file_name) {
7415 # some commits could have deleted file in question,
7416 # and not have it in tree, but one of them has to have it
7417 for (my $i = 0; $i < @commitlist; $i++) {
7418 $file_hash = git_get_hash_by_path($commitlist[$i]{'id'}, $file_name);
7419 last if defined $file_hash;
7422 if (defined $file_hash) {
7423 $ftype = git_get_type($file_hash);
7425 if (defined $file_name && !defined $ftype) {
7426 die_error(500, "Unknown type of object");
7428 my %co;
7429 if (defined $file_name) {
7430 %co = parse_commit($base)
7431 or die_error(404, "Unknown commit object");
7435 my $paging_nav = format_paging_nav($fmt_name, $page, $#commitlist >= 100);
7436 my $next_link = '';
7437 if ($#commitlist >= 100) {
7438 $next_link =
7439 $cgi->a({-href => href(-replay=>1, page=>$page+1),
7440 -accesskey => "n", -title => "Alt-n"}, "next");
7442 my $patch_max = gitweb_get_feature('patches');
7443 if ($patch_max && !defined $file_name) {
7444 if ($patch_max < 0 || @commitlist <= $patch_max) {
7445 $paging_nav .= " &sdot; " .
7446 $cgi->a({-href => href(action=>"patches", -replay=>1)},
7447 "patches");
7451 git_header_html();
7452 git_print_page_nav($fmt_name,'', $hash,$hash,$hash, $paging_nav);
7453 if (defined $file_name) {
7454 git_print_header_div('commit', esc_html($co{'title'}), $base);
7455 } else {
7456 git_print_header_div('summary', $project)
7458 git_print_page_path($file_name, $ftype, $hash_base)
7459 if (defined $file_name);
7461 $body_subr->(\@commitlist, 0, 99, $refs, $next_link,
7462 $file_name, $file_hash, $ftype);
7464 git_footer_html();
7467 sub git_log {
7468 git_log_generic('log', \&git_log_body,
7469 $hash, $hash_parent);
7472 sub git_commit {
7473 $hash ||= $hash_base || "HEAD";
7474 my %co = parse_commit($hash)
7475 or die_error(404, "Unknown commit object");
7477 my $parent = $co{'parent'};
7478 my $parents = $co{'parents'}; # listref
7480 # we need to prepare $formats_nav before any parameter munging
7481 my $formats_nav;
7482 if (!defined $parent) {
7483 # --root commitdiff
7484 $formats_nav .= '(initial)';
7485 } elsif (@$parents == 1) {
7486 # single parent commit
7487 $formats_nav .=
7488 '(parent: ' .
7489 $cgi->a({-href => href(action=>"commit",
7490 hash=>$parent)},
7491 esc_html(substr($parent, 0, 7))) .
7492 ')';
7493 } else {
7494 # merge commit
7495 $formats_nav .=
7496 '(merge: ' .
7497 join(' ', map {
7498 $cgi->a({-href => href(action=>"commit",
7499 hash=>$_)},
7500 esc_html(substr($_, 0, 7)));
7501 } @$parents ) .
7502 ')';
7504 if (gitweb_check_feature('patches') && @$parents <= 1) {
7505 $formats_nav .= " | " .
7506 $cgi->a({-href => href(action=>"patch", -replay=>1)},
7507 "patch");
7510 if (!defined $parent) {
7511 $parent = "--root";
7513 my @difftree;
7514 open my $fd, "-|", git_cmd(), "diff-tree", '-r', "--no-commit-id",
7515 @diff_opts,
7516 (@$parents <= 1 ? $parent : '-c'),
7517 $hash, "--"
7518 or die_error(500, "Open git-diff-tree failed");
7519 @difftree = map { chomp; $_ } <$fd>;
7520 close $fd or die_error(404, "Reading git-diff-tree failed");
7522 # non-textual hash id's can be cached
7523 my $expires;
7524 if ($hash =~ m/^[0-9a-fA-F]{40}$/) {
7525 $expires = "+1d";
7527 my $refs = git_get_references();
7528 my $ref = format_ref_marker($refs, $co{'id'});
7530 git_header_html(undef, $expires);
7531 git_print_page_nav('commit', '',
7532 $hash, $co{'tree'}, $hash,
7533 $formats_nav);
7535 if (defined $co{'parent'}) {
7536 git_print_header_div('commitdiff', esc_html($co{'title'}) . $ref, $hash);
7537 } else {
7538 git_print_header_div('tree', esc_html($co{'title'}) . $ref, $co{'tree'}, $hash);
7540 print "<div class=\"title_text\">\n" .
7541 "<table class=\"object_header\">\n";
7542 git_print_authorship_rows(\%co);
7543 print "<tr><td>commit</td><td class=\"sha1\">$co{'id'}</td></tr>\n";
7544 print "<tr>" .
7545 "<td>tree</td>" .
7546 "<td class=\"sha1\">" .
7547 $cgi->a({-href => href(action=>"tree", hash=>$co{'tree'}, hash_base=>$hash),
7548 class => "list"}, $co{'tree'}) .
7549 "</td>" .
7550 "<td class=\"link\">" .
7551 $cgi->a({-href => href(action=>"tree", hash=>$co{'tree'}, hash_base=>$hash)},
7552 "tree");
7553 my $snapshot_links = format_snapshot_links($hash);
7554 if (defined $snapshot_links) {
7555 print " | " . $snapshot_links;
7557 print "</td>" .
7558 "</tr>\n";
7560 foreach my $par (@$parents) {
7561 print "<tr>" .
7562 "<td>parent</td>" .
7563 "<td class=\"sha1\">" .
7564 $cgi->a({-href => href(action=>"commit", hash=>$par),
7565 class => "list"}, $par) .
7566 "</td>" .
7567 "<td class=\"link\">" .
7568 $cgi->a({-href => href(action=>"commit", hash=>$par)}, "commit") .
7569 " | " .
7570 $cgi->a({-href => href(action=>"commitdiff", hash=>$hash, hash_parent=>$par)}, "diff") .
7571 "</td>" .
7572 "</tr>\n";
7574 print "</table>".
7575 "</div>\n";
7577 print "<div class=\"page_body\">\n";
7578 git_print_log($co{'comment'});
7579 print "</div>\n";
7581 git_difftree_body(\@difftree, $hash, @$parents);
7583 git_footer_html();
7586 sub git_object {
7587 # object is defined by:
7588 # - hash or hash_base alone
7589 # - hash_base and file_name
7590 my $type;
7592 # - hash or hash_base alone
7593 if ($hash || ($hash_base && !defined $file_name)) {
7594 my $object_id = $hash || $hash_base;
7596 open my $fd, "-|", quote_command(
7597 git_cmd(), 'cat-file', '-t', $object_id) . ' 2> /dev/null'
7598 or die_error(404, "Object does not exist");
7599 $type = <$fd>;
7600 defined $type && chomp $type;
7601 close $fd
7602 or die_error(404, "Object does not exist");
7604 # - hash_base and file_name
7605 } elsif ($hash_base && defined $file_name) {
7606 $file_name =~ s,/+$,,;
7608 system(git_cmd(), "cat-file", '-e', $hash_base) == 0
7609 or die_error(404, "Base object does not exist");
7611 # here errors should not happen
7612 open my $fd, "-|", git_cmd(), "ls-tree", $hash_base, "--", $file_name
7613 or die_error(500, "Open git-ls-tree failed");
7614 my $line = <$fd>;
7615 close $fd;
7617 #'100644 blob 0fa3f3a66fb6a137f6ec2c19351ed4d807070ffa panic.c'
7618 unless ($line && $line =~ m/^([0-9]+) (.+) ([0-9a-fA-F]{40})\t/) {
7619 die_error(404, "File or directory for given base does not exist");
7621 $type = $2;
7622 $hash = $3;
7623 } else {
7624 die_error(400, "Not enough information to find object");
7627 print $cgi->redirect(-uri => href(action=>$type, -full=>1,
7628 hash=>$hash, hash_base=>$hash_base,
7629 file_name=>$file_name),
7630 -status => '302 Found');
7633 sub git_blobdiff {
7634 my $format = shift || 'html';
7635 my $diff_style = $input_params{'diff_style'} || 'inline';
7637 my $fd;
7638 my @difftree;
7639 my %diffinfo;
7640 my $expires;
7642 # preparing $fd and %diffinfo for git_patchset_body
7643 # new style URI
7644 if (defined $hash_base && defined $hash_parent_base) {
7645 if (defined $file_name) {
7646 # read raw output
7647 open $fd, "-|", git_cmd(), "diff-tree", '-r', @diff_opts,
7648 $hash_parent_base, $hash_base,
7649 "--", (defined $file_parent ? $file_parent : ()), $file_name
7650 or die_error(500, "Open git-diff-tree failed");
7651 @difftree = map { chomp; $_ } <$fd>;
7652 close $fd
7653 or die_error(404, "Reading git-diff-tree failed");
7654 @difftree
7655 or die_error(404, "Blob diff not found");
7657 } elsif (defined $hash &&
7658 $hash =~ /[0-9a-fA-F]{40}/) {
7659 # try to find filename from $hash
7661 # read filtered raw output
7662 open $fd, "-|", git_cmd(), "diff-tree", '-r', @diff_opts,
7663 $hash_parent_base, $hash_base, "--"
7664 or die_error(500, "Open git-diff-tree failed");
7665 @difftree =
7666 # ':100644 100644 03b21826... 3b93d5e7... M ls-files.c'
7667 # $hash == to_id
7668 grep { /^:[0-7]{6} [0-7]{6} [0-9a-fA-F]{40} $hash/ }
7669 map { chomp; $_ } <$fd>;
7670 close $fd
7671 or die_error(404, "Reading git-diff-tree failed");
7672 @difftree
7673 or die_error(404, "Blob diff not found");
7675 } else {
7676 die_error(400, "Missing one of the blob diff parameters");
7679 if (@difftree > 1) {
7680 die_error(400, "Ambiguous blob diff specification");
7683 %diffinfo = parse_difftree_raw_line($difftree[0]);
7684 $file_parent ||= $diffinfo{'from_file'} || $file_name;
7685 $file_name ||= $diffinfo{'to_file'};
7687 $hash_parent ||= $diffinfo{'from_id'};
7688 $hash ||= $diffinfo{'to_id'};
7690 # non-textual hash id's can be cached
7691 if ($hash_base =~ m/^[0-9a-fA-F]{40}$/ &&
7692 $hash_parent_base =~ m/^[0-9a-fA-F]{40}$/) {
7693 $expires = '+1d';
7696 # open patch output
7697 open $fd, "-|", git_cmd(), "diff-tree", '-r', @diff_opts,
7698 '-p', ($format eq 'html' ? "--full-index" : ()),
7699 $hash_parent_base, $hash_base,
7700 "--", (defined $file_parent ? $file_parent : ()), $file_name
7701 or die_error(500, "Open git-diff-tree failed");
7704 # old/legacy style URI -- not generated anymore since 1.4.3.
7705 if (!%diffinfo) {
7706 die_error('404 Not Found', "Missing one of the blob diff parameters")
7709 # header
7710 if ($format eq 'html') {
7711 my $formats_nav =
7712 $cgi->a({-href => href(action=>"blobdiff_plain", -replay=>1)},
7713 "raw");
7714 $formats_nav .= diff_style_nav($diff_style);
7715 git_header_html(undef, $expires);
7716 if (defined $hash_base && (my %co = parse_commit($hash_base))) {
7717 git_print_page_nav('','', $hash_base,$co{'tree'},$hash_base, $formats_nav);
7718 git_print_header_div('commit', esc_html($co{'title'}), $hash_base);
7719 } else {
7720 print "<div class=\"page_nav\"><br/>$formats_nav<br/></div>\n";
7721 print "<div class=\"title\">".esc_html("$hash vs $hash_parent")."</div>\n";
7723 if (defined $file_name) {
7724 git_print_page_path($file_name, "blob", $hash_base);
7725 } else {
7726 print "<div class=\"page_path\"></div>\n";
7729 } elsif ($format eq 'plain') {
7730 print $cgi->header(
7731 -type => 'text/plain',
7732 -charset => 'utf-8',
7733 -expires => $expires,
7734 -content_disposition => 'inline; filename="' . "$file_name" . '.patch"');
7736 print "X-Git-Url: " . $cgi->self_url() . "\n\n";
7738 } else {
7739 die_error(400, "Unknown blobdiff format");
7742 # patch
7743 if ($format eq 'html') {
7744 print "<div class=\"page_body\">\n";
7746 git_patchset_body($fd, $diff_style,
7747 [ \%diffinfo ], $hash_base, $hash_parent_base);
7748 close $fd;
7750 print "</div>\n"; # class="page_body"
7751 git_footer_html();
7753 } else {
7754 while (my $line = <$fd>) {
7755 $line =~ s!a/($hash|$hash_parent)!'a/'.esc_path($diffinfo{'from_file'})!eg;
7756 $line =~ s!b/($hash|$hash_parent)!'b/'.esc_path($diffinfo{'to_file'})!eg;
7758 print $line;
7760 last if $line =~ m!^\+\+\+!;
7762 local $/ = undef;
7763 print <$fd>;
7764 close $fd;
7768 sub git_blobdiff_plain {
7769 git_blobdiff('plain');
7772 # assumes that it is added as later part of already existing navigation,
7773 # so it returns "| foo | bar" rather than just "foo | bar"
7774 sub diff_style_nav {
7775 my ($diff_style, $is_combined) = @_;
7776 $diff_style ||= 'inline';
7778 return "" if ($is_combined);
7780 my @styles = (inline => 'inline', 'sidebyside' => 'side by side');
7781 my %styles = @styles;
7782 @styles =
7783 @styles[ map { $_ * 2 } 0..$#styles/2 ];
7785 return join '',
7786 map { " | ".$_ }
7787 map {
7788 $_ eq $diff_style ? $styles{$_} :
7789 $cgi->a({-href => href(-replay=>1, diff_style => $_)}, $styles{$_})
7790 } @styles;
7793 sub git_commitdiff {
7794 my %params = @_;
7795 my $format = $params{-format} || 'html';
7796 my $diff_style = $input_params{'diff_style'} || 'inline';
7798 my ($patch_max) = gitweb_get_feature('patches');
7799 if ($format eq 'patch') {
7800 die_error(403, "Patch view not allowed") unless $patch_max;
7803 $hash ||= $hash_base || "HEAD";
7804 my %co = parse_commit($hash)
7805 or die_error(404, "Unknown commit object");
7807 # choose format for commitdiff for merge
7808 if (! defined $hash_parent && @{$co{'parents'}} > 1) {
7809 $hash_parent = '--cc';
7811 # we need to prepare $formats_nav before almost any parameter munging
7812 my $formats_nav;
7813 if ($format eq 'html') {
7814 $formats_nav =
7815 $cgi->a({-href => href(action=>"commitdiff_plain", -replay=>1)},
7816 "raw");
7817 if ($patch_max && @{$co{'parents'}} <= 1) {
7818 $formats_nav .= " | " .
7819 $cgi->a({-href => href(action=>"patch", -replay=>1)},
7820 "patch");
7822 $formats_nav .= diff_style_nav($diff_style, @{$co{'parents'}} > 1);
7824 if (defined $hash_parent &&
7825 $hash_parent ne '-c' && $hash_parent ne '--cc') {
7826 # commitdiff with two commits given
7827 my $hash_parent_short = $hash_parent;
7828 if ($hash_parent =~ m/^[0-9a-fA-F]{40}$/) {
7829 $hash_parent_short = substr($hash_parent, 0, 7);
7831 $formats_nav .=
7832 ' (from';
7833 for (my $i = 0; $i < @{$co{'parents'}}; $i++) {
7834 if ($co{'parents'}[$i] eq $hash_parent) {
7835 $formats_nav .= ' parent ' . ($i+1);
7836 last;
7839 $formats_nav .= ': ' .
7840 $cgi->a({-href => href(-replay=>1,
7841 hash=>$hash_parent, hash_base=>undef)},
7842 esc_html($hash_parent_short)) .
7843 ')';
7844 } elsif (!$co{'parent'}) {
7845 # --root commitdiff
7846 $formats_nav .= ' (initial)';
7847 } elsif (scalar @{$co{'parents'}} == 1) {
7848 # single parent commit
7849 $formats_nav .=
7850 ' (parent: ' .
7851 $cgi->a({-href => href(-replay=>1,
7852 hash=>$co{'parent'}, hash_base=>undef)},
7853 esc_html(substr($co{'parent'}, 0, 7))) .
7854 ')';
7855 } else {
7856 # merge commit
7857 if ($hash_parent eq '--cc') {
7858 $formats_nav .= ' | ' .
7859 $cgi->a({-href => href(-replay=>1,
7860 hash=>$hash, hash_parent=>'-c')},
7861 'combined');
7862 } else { # $hash_parent eq '-c'
7863 $formats_nav .= ' | ' .
7864 $cgi->a({-href => href(-replay=>1,
7865 hash=>$hash, hash_parent=>'--cc')},
7866 'compact');
7868 $formats_nav .=
7869 ' (merge: ' .
7870 join(' ', map {
7871 $cgi->a({-href => href(-replay=>1,
7872 hash=>$_, hash_base=>undef)},
7873 esc_html(substr($_, 0, 7)));
7874 } @{$co{'parents'}} ) .
7875 ')';
7879 my $hash_parent_param = $hash_parent;
7880 if (!defined $hash_parent_param) {
7881 # --cc for multiple parents, --root for parentless
7882 $hash_parent_param =
7883 @{$co{'parents'}} > 1 ? '--cc' : $co{'parent'} || '--root';
7886 # read commitdiff
7887 my $fd;
7888 my @difftree;
7889 if ($format eq 'html') {
7890 open $fd, "-|", git_cmd(), "diff-tree", '-r', @diff_opts,
7891 "--no-commit-id", "--patch-with-raw", "--full-index",
7892 $hash_parent_param, $hash, "--"
7893 or die_error(500, "Open git-diff-tree failed");
7895 while (my $line = <$fd>) {
7896 chomp $line;
7897 # empty line ends raw part of diff-tree output
7898 last unless $line;
7899 push @difftree, scalar parse_difftree_raw_line($line);
7902 } elsif ($format eq 'plain') {
7903 open $fd, "-|", git_cmd(), "diff-tree", '-r', @diff_opts,
7904 '-p', $hash_parent_param, $hash, "--"
7905 or die_error(500, "Open git-diff-tree failed");
7906 } elsif ($format eq 'patch') {
7907 # For commit ranges, we limit the output to the number of
7908 # patches specified in the 'patches' feature.
7909 # For single commits, we limit the output to a single patch,
7910 # diverging from the git-format-patch default.
7911 my @commit_spec = ();
7912 if ($hash_parent) {
7913 if ($patch_max > 0) {
7914 push @commit_spec, "-$patch_max";
7916 push @commit_spec, '-n', "$hash_parent..$hash";
7917 } else {
7918 if ($params{-single}) {
7919 push @commit_spec, '-1';
7920 } else {
7921 if ($patch_max > 0) {
7922 push @commit_spec, "-$patch_max";
7924 push @commit_spec, "-n";
7926 push @commit_spec, '--root', $hash;
7928 open $fd, "-|", git_cmd(), "format-patch", @diff_opts,
7929 '--encoding=utf8', '--stdout', @commit_spec
7930 or die_error(500, "Open git-format-patch failed");
7931 } else {
7932 die_error(400, "Unknown commitdiff format");
7935 # non-textual hash id's can be cached
7936 my $expires;
7937 if ($hash =~ m/^[0-9a-fA-F]{40}$/) {
7938 $expires = "+1d";
7941 # write commit message
7942 if ($format eq 'html') {
7943 my $refs = git_get_references();
7944 my $ref = format_ref_marker($refs, $co{'id'});
7946 git_header_html(undef, $expires);
7947 git_print_page_nav('commitdiff','', $hash,$co{'tree'},$hash, $formats_nav);
7948 git_print_header_div('commit', esc_html($co{'title'}) . $ref, $hash);
7949 print "<div class=\"title_text\">\n" .
7950 "<table class=\"object_header\">\n";
7951 git_print_authorship_rows(\%co);
7952 print "</table>".
7953 "</div>\n";
7954 print "<div class=\"page_body\">\n";
7955 if (@{$co{'comment'}} > 1) {
7956 print "<div class=\"log\">\n";
7957 git_print_log($co{'comment'}, -final_empty_line=> 1, -remove_title => 1);
7958 print "</div>\n"; # class="log"
7961 } elsif ($format eq 'plain') {
7962 my $refs = git_get_references("tags");
7963 my $tagname = git_get_rev_name_tags($hash);
7964 my $filename = basename($project) . "-$hash.patch";
7966 print $cgi->header(
7967 -type => 'text/plain',
7968 -charset => 'utf-8',
7969 -expires => $expires,
7970 -content_disposition => 'inline; filename="' . "$filename" . '"');
7971 my %ad = parse_date($co{'author_epoch'}, $co{'author_tz'});
7972 print "From: " . to_utf8($co{'author'}) . "\n";
7973 print "Date: $ad{'rfc2822'} ($ad{'tz_local'})\n";
7974 print "Subject: " . to_utf8($co{'title'}) . "\n";
7976 print "X-Git-Tag: $tagname\n" if $tagname;
7977 print "X-Git-Url: " . $cgi->self_url() . "\n\n";
7979 foreach my $line (@{$co{'comment'}}) {
7980 print to_utf8($line) . "\n";
7982 print "---\n\n";
7983 } elsif ($format eq 'patch') {
7984 my $filename = basename($project) . "-$hash.patch";
7986 print $cgi->header(
7987 -type => 'text/plain',
7988 -charset => 'utf-8',
7989 -expires => $expires,
7990 -content_disposition => 'inline; filename="' . "$filename" . '"');
7993 # write patch
7994 if ($format eq 'html') {
7995 my $use_parents = !defined $hash_parent ||
7996 $hash_parent eq '-c' || $hash_parent eq '--cc';
7997 git_difftree_body(\@difftree, $hash,
7998 $use_parents ? @{$co{'parents'}} : $hash_parent);
7999 print "<br/>\n";
8001 git_patchset_body($fd, $diff_style,
8002 \@difftree, $hash,
8003 $use_parents ? @{$co{'parents'}} : $hash_parent);
8004 close $fd;
8005 print "</div>\n"; # class="page_body"
8006 git_footer_html();
8008 } elsif ($format eq 'plain') {
8009 local $/ = undef;
8010 print <$fd>;
8011 close $fd
8012 or print "Reading git-diff-tree failed\n";
8013 } elsif ($format eq 'patch') {
8014 local $/ = undef;
8015 print <$fd>;
8016 close $fd
8017 or print "Reading git-format-patch failed\n";
8021 sub git_commitdiff_plain {
8022 git_commitdiff(-format => 'plain');
8025 # format-patch-style patches
8026 sub git_patch {
8027 git_commitdiff(-format => 'patch', -single => 1);
8030 sub git_patches {
8031 git_commitdiff(-format => 'patch');
8034 sub git_history {
8035 git_log_generic('history', \&git_history_body,
8036 $hash_base, $hash_parent_base,
8037 $file_name, $hash);
8040 sub git_search {
8041 $searchtype ||= 'commit';
8043 # check if appropriate features are enabled
8044 gitweb_check_feature('search')
8045 or die_error(403, "Search is disabled");
8046 if ($searchtype eq 'pickaxe') {
8047 # pickaxe may take all resources of your box and run for several minutes
8048 # with every query - so decide by yourself how public you make this feature
8049 gitweb_check_feature('pickaxe')
8050 or die_error(403, "Pickaxe search is disabled");
8052 if ($searchtype eq 'grep') {
8053 # grep search might be potentially CPU-intensive, too
8054 gitweb_check_feature('grep')
8055 or die_error(403, "Grep search is disabled");
8058 if (!defined $searchtext) {
8059 die_error(400, "Text field is empty");
8061 if (!defined $hash) {
8062 $hash = git_get_head_hash($project);
8064 my %co = parse_commit($hash);
8065 if (!%co) {
8066 die_error(404, "Unknown commit object");
8068 if (!defined $page) {
8069 $page = 0;
8072 if ($searchtype eq 'commit' ||
8073 $searchtype eq 'author' ||
8074 $searchtype eq 'committer') {
8075 git_search_message(%co);
8076 } elsif ($searchtype eq 'pickaxe') {
8077 git_search_changes(%co);
8078 } elsif ($searchtype eq 'grep') {
8079 git_search_files(%co);
8080 } else {
8081 die_error(400, "Unknown search type");
8085 sub git_search_help {
8086 git_header_html();
8087 git_print_page_nav('','', $hash,$hash,$hash);
8088 print <<EOT;
8089 <p><strong>Pattern</strong> is by default a normal string that is matched precisely (but without
8090 regard to case, except in the case of pickaxe). However, when you check the <em>re</em> checkbox,
8091 the pattern entered is recognized as the POSIX extended
8092 <a href="https://en.wikipedia.org/wiki/Regular_expression">regular expression</a> (also case
8093 insensitive).</p>
8094 <dl>
8095 <dt><b>commit</b></dt>
8096 <dd>The commit messages and authorship information will be scanned for the given pattern.</dd>
8098 my $have_grep = gitweb_check_feature('grep');
8099 if ($have_grep) {
8100 print <<EOT;
8101 <dt><b>grep</b></dt>
8102 <dd>All files in the currently selected tree (HEAD unless you are explicitly browsing
8103 a different one) are searched for the given pattern. On large trees, this search can take
8104 a while and put some strain on the server, so please use it with some consideration. Note that
8105 due to git-grep peculiarity, currently if regexp mode is turned off, the matches are
8106 case-sensitive.</dd>
8109 print <<EOT;
8110 <dt><b>author</b></dt>
8111 <dd>Name and e-mail of the change author and date of birth of the patch will be scanned for the given pattern.</dd>
8112 <dt><b>committer</b></dt>
8113 <dd>Name and e-mail of the committer and date of commit will be scanned for the given pattern.</dd>
8115 my $have_pickaxe = gitweb_check_feature('pickaxe');
8116 if ($have_pickaxe) {
8117 print <<EOT;
8118 <dt><b>pickaxe</b></dt>
8119 <dd>All commits that caused the string to appear or disappear from any file (changes that
8120 added, removed or "modified" the string) will be listed. This search can take a while and
8121 takes a lot of strain on the server, so please use it wisely. Note that since you may be
8122 interested even in changes just changing the case as well, this search is case sensitive.</dd>
8125 print "</dl>\n";
8126 git_footer_html();
8129 sub git_shortlog {
8130 git_log_generic('shortlog', \&git_shortlog_body,
8131 $hash, $hash_parent);
8134 ## ......................................................................
8135 ## feeds (RSS, Atom; OPML)
8137 sub git_feed {
8138 my $format = shift || 'atom';
8139 my $have_blame = gitweb_check_feature('blame');
8141 # Atom: http://www.atomenabled.org/developers/syndication/
8142 # RSS: http://www.notestips.com/80256B3A007F2692/1/NAMO5P9UPQ
8143 if ($format ne 'rss' && $format ne 'atom') {
8144 die_error(400, "Unknown web feed format");
8147 # log/feed of current (HEAD) branch, log of given branch, history of file/directory
8148 my $head = $hash || 'HEAD';
8149 my @commitlist = parse_commits($head, 150, 0, $file_name);
8151 my %latest_commit;
8152 my %latest_date;
8153 my $content_type = "application/$format+xml";
8154 if (defined $cgi->http('HTTP_ACCEPT') &&
8155 $cgi->Accept('text/xml') > $cgi->Accept($content_type)) {
8156 # browser (feed reader) prefers text/xml
8157 $content_type = 'text/xml';
8159 if (defined($commitlist[0])) {
8160 %latest_commit = %{$commitlist[0]};
8161 my $latest_epoch = $latest_commit{'committer_epoch'};
8162 exit_if_unmodified_since($latest_epoch);
8163 %latest_date = parse_date($latest_epoch, $latest_commit{'committer_tz'});
8165 print $cgi->header(
8166 -type => $content_type,
8167 -charset => 'utf-8',
8168 %latest_date ? (-last_modified => $latest_date{'rfc2822'}) : (),
8169 -status => '200 OK');
8171 # Optimization: skip generating the body if client asks only
8172 # for Last-Modified date.
8173 return if ($cgi->request_method() eq 'HEAD');
8175 # header variables
8176 my $title = "$site_name - $project/$action";
8177 my $feed_type = 'log';
8178 if (defined $hash) {
8179 $title .= " - '$hash'";
8180 $feed_type = 'branch log';
8181 if (defined $file_name) {
8182 $title .= " :: $file_name";
8183 $feed_type = 'history';
8185 } elsif (defined $file_name) {
8186 $title .= " - $file_name";
8187 $feed_type = 'history';
8189 $title .= " $feed_type";
8190 $title = esc_html($title);
8191 my $descr = git_get_project_description($project);
8192 if (defined $descr) {
8193 $descr = esc_html($descr);
8194 } else {
8195 $descr = "$project " .
8196 ($format eq 'rss' ? 'RSS' : 'Atom') .
8197 " feed";
8199 my $owner = git_get_project_owner($project);
8200 $owner = esc_html($owner);
8202 #header
8203 my $alt_url;
8204 if (defined $file_name) {
8205 $alt_url = href(-full=>1, action=>"history", hash=>$hash, file_name=>$file_name);
8206 } elsif (defined $hash) {
8207 $alt_url = href(-full=>1, action=>"log", hash=>$hash);
8208 } else {
8209 $alt_url = href(-full=>1, action=>"summary");
8211 print qq!<?xml version="1.0" encoding="utf-8"?>\n!;
8212 if ($format eq 'rss') {
8213 print <<XML;
8214 <rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/">
8215 <channel>
8217 print "<title>$title</title>\n" .
8218 "<link>$alt_url</link>\n" .
8219 "<description>$descr</description>\n" .
8220 "<language>en</language>\n" .
8221 # project owner is responsible for 'editorial' content
8222 "<managingEditor>$owner</managingEditor>\n";
8223 if (defined $logo || defined $favicon) {
8224 # prefer the logo to the favicon, since RSS
8225 # doesn't allow both
8226 my $img = esc_url($logo || $favicon);
8227 print "<image>\n" .
8228 "<url>$img</url>\n" .
8229 "<title>$title</title>\n" .
8230 "<link>$alt_url</link>\n" .
8231 "</image>\n";
8233 if (%latest_date) {
8234 print "<pubDate>$latest_date{'rfc2822'}</pubDate>\n";
8235 print "<lastBuildDate>$latest_date{'rfc2822'}</lastBuildDate>\n";
8237 print "<generator>gitweb v.$version/$git_version</generator>\n";
8238 } elsif ($format eq 'atom') {
8239 print <<XML;
8240 <feed xmlns="http://www.w3.org/2005/Atom">
8242 print "<title>$title</title>\n" .
8243 "<subtitle>$descr</subtitle>\n" .
8244 '<link rel="alternate" type="text/html" href="' .
8245 $alt_url . '" />' . "\n" .
8246 '<link rel="self" type="' . $content_type . '" href="' .
8247 $cgi->self_url() . '" />' . "\n" .
8248 "<id>" . href(-full=>1) . "</id>\n" .
8249 # use project owner for feed author
8250 "<author><name>$owner</name></author>\n";
8251 if (defined $favicon) {
8252 print "<icon>" . esc_url($favicon) . "</icon>\n";
8254 if (defined $logo) {
8255 # not twice as wide as tall: 72 x 27 pixels
8256 print "<logo>" . esc_url($logo) . "</logo>\n";
8258 if (! %latest_date) {
8259 # dummy date to keep the feed valid until commits trickle in:
8260 print "<updated>1970-01-01T00:00:00Z</updated>\n";
8261 } else {
8262 print "<updated>$latest_date{'iso-8601'}</updated>\n";
8264 print "<generator version='$version/$git_version'>gitweb</generator>\n";
8267 # contents
8268 for (my $i = 0; $i <= $#commitlist; $i++) {
8269 my %co = %{$commitlist[$i]};
8270 my $commit = $co{'id'};
8271 # we read 150, we always show 30 and the ones more recent than 48 hours
8272 if (($i >= 20) && ((time - $co{'author_epoch'}) > 48*60*60)) {
8273 last;
8275 my %cd = parse_date($co{'author_epoch'}, $co{'author_tz'});
8277 # get list of changed files
8278 open my $fd, "-|", git_cmd(), "diff-tree", '-r', @diff_opts,
8279 $co{'parent'} || "--root",
8280 $co{'id'}, "--", (defined $file_name ? $file_name : ())
8281 or next;
8282 my @difftree = map { chomp; $_ } <$fd>;
8283 close $fd
8284 or next;
8286 # print element (entry, item)
8287 my $co_url = href(-full=>1, action=>"commitdiff", hash=>$commit);
8288 if ($format eq 'rss') {
8289 print "<item>\n" .
8290 "<title>" . esc_html($co{'title'}) . "</title>\n" .
8291 "<author>" . esc_html($co{'author'}) . "</author>\n" .
8292 "<pubDate>$cd{'rfc2822'}</pubDate>\n" .
8293 "<guid isPermaLink=\"true\">$co_url</guid>\n" .
8294 "<link>$co_url</link>\n" .
8295 "<description>" . esc_html($co{'title'}) . "</description>\n" .
8296 "<content:encoded>" .
8297 "<![CDATA[\n";
8298 } elsif ($format eq 'atom') {
8299 print "<entry>\n" .
8300 "<title type=\"html\">" . esc_html($co{'title'}) . "</title>\n" .
8301 "<updated>$cd{'iso-8601'}</updated>\n" .
8302 "<author>\n" .
8303 " <name>" . esc_html($co{'author_name'}) . "</name>\n";
8304 if ($co{'author_email'}) {
8305 print " <email>" . esc_html($co{'author_email'}) . "</email>\n";
8307 print "</author>\n" .
8308 # use committer for contributor
8309 "<contributor>\n" .
8310 " <name>" . esc_html($co{'committer_name'}) . "</name>\n";
8311 if ($co{'committer_email'}) {
8312 print " <email>" . esc_html($co{'committer_email'}) . "</email>\n";
8314 print "</contributor>\n" .
8315 "<published>$cd{'iso-8601'}</published>\n" .
8316 "<link rel=\"alternate\" type=\"text/html\" href=\"$co_url\" />\n" .
8317 "<id>$co_url</id>\n" .
8318 "<content type=\"xhtml\" xml:base=\"" . esc_url($my_url) . "\">\n" .
8319 "<div xmlns=\"http://www.w3.org/1999/xhtml\">\n";
8321 my $comment = $co{'comment'};
8322 print "<pre>\n";
8323 foreach my $line (@$comment) {
8324 $line = esc_html($line);
8325 print "$line\n";
8327 print "</pre><ul>\n";
8328 foreach my $difftree_line (@difftree) {
8329 my %difftree = parse_difftree_raw_line($difftree_line);
8330 next if !$difftree{'from_id'};
8332 my $file = $difftree{'file'} || $difftree{'to_file'};
8334 print "<li>" .
8335 "[" .
8336 $cgi->a({-href => href(-full=>1, action=>"blobdiff",
8337 hash=>$difftree{'to_id'}, hash_parent=>$difftree{'from_id'},
8338 hash_base=>$co{'id'}, hash_parent_base=>$co{'parent'},
8339 file_name=>$file, file_parent=>$difftree{'from_file'}),
8340 -title => "diff"}, 'D');
8341 if ($have_blame) {
8342 print $cgi->a({-href => href(-full=>1, action=>"blame",
8343 file_name=>$file, hash_base=>$commit),
8344 -title => "blame"}, 'B');
8346 # if this is not a feed of a file history
8347 if (!defined $file_name || $file_name ne $file) {
8348 print $cgi->a({-href => href(-full=>1, action=>"history",
8349 file_name=>$file, hash=>$commit),
8350 -title => "history"}, 'H');
8352 $file = esc_path($file);
8353 print "] ".
8354 "$file</li>\n";
8356 if ($format eq 'rss') {
8357 print "</ul>]]>\n" .
8358 "</content:encoded>\n" .
8359 "</item>\n";
8360 } elsif ($format eq 'atom') {
8361 print "</ul>\n</div>\n" .
8362 "</content>\n" .
8363 "</entry>\n";
8367 # end of feed
8368 if ($format eq 'rss') {
8369 print "</channel>\n</rss>\n";
8370 } elsif ($format eq 'atom') {
8371 print "</feed>\n";
8375 sub git_rss {
8376 git_feed('rss');
8379 sub git_atom {
8380 git_feed('atom');
8383 sub git_opml {
8384 my @list = git_get_projects_list($project_filter, $strict_export);
8385 if (!@list) {
8386 die_error(404, "No projects found");
8389 print $cgi->header(
8390 -type => 'text/xml',
8391 -charset => 'utf-8',
8392 -content_disposition => 'inline; filename="opml.xml"');
8394 my $title = esc_html($site_name);
8395 my $filter = " within subdirectory ";
8396 if (defined $project_filter) {
8397 $filter .= esc_html($project_filter);
8398 } else {
8399 $filter = "";
8401 print <<XML;
8402 <?xml version="1.0" encoding="utf-8"?>
8403 <opml version="1.0">
8404 <head>
8405 <title>$title OPML Export$filter</title>
8406 </head>
8407 <body>
8408 <outline text="git RSS feeds">
8411 foreach my $pr (@list) {
8412 my %proj = %$pr;
8413 my $head = git_get_head_hash($proj{'path'});
8414 if (!defined $head) {
8415 next;
8417 $git_dir = "$projectroot/$proj{'path'}";
8418 my %co = parse_commit($head);
8419 if (!%co) {
8420 next;
8423 my $path = esc_html(chop_str($proj{'path'}, 25, 5));
8424 my $rss = href('project' => $proj{'path'}, 'action' => 'rss', -full => 1);
8425 my $html = href('project' => $proj{'path'}, 'action' => 'summary', -full => 1);
8426 print "<outline type=\"rss\" text=\"$path\" title=\"$path\" xmlUrl=\"$rss\" htmlUrl=\"$html\"/>\n";
8428 print <<XML;
8429 </outline>
8430 </body>
8431 </opml>