X-Git-Url: http://git.vanrenterghem.biz/git.ikiwiki.info.git/blobdiff_plain/9ef4b5e677919f95dd2aa08ca51a0313c6c5542d..a55752a7bedb83fa5339f232b9a61a60a67251f3:/IkiWiki/Plugin/git.pm diff --git a/IkiWiki/Plugin/git.pm b/IkiWiki/Plugin/git.pm index ad58231e0..75b89e476 100644 --- a/IkiWiki/Plugin/git.pm +++ b/IkiWiki/Plugin/git.pm @@ -5,11 +5,11 @@ use warnings; use strict; use IkiWiki; use Encode; +use URI::Escape q{uri_escape_utf8}; use open qw{:utf8 :std}; my $sha1_pattern = qr/[0-9a-fA-F]{40}/; # pattern to validate Git sha1sums my $dummy_commit_msg = 'dummy commit'; # message to skip in recent changes -my $no_chdir=0; sub import { hook(type => "checkconfig", id => "git", call => \&checkconfig); @@ -25,7 +25,12 @@ sub import { hook(type => "rcs", id => "rcs_recentchanges", call => \&rcs_recentchanges); hook(type => "rcs", id => "rcs_diff", call => \&rcs_diff); hook(type => "rcs", id => "rcs_getctime", call => \&rcs_getctime); + hook(type => "rcs", id => "rcs_getmtime", call => \&rcs_getmtime); hook(type => "rcs", id => "rcs_receive", call => \&rcs_receive); + hook(type => "rcs", id => "rcs_preprevert", call => \&rcs_preprevert); + hook(type => "rcs", id => "rcs_revert", call => \&rcs_revert); + hook(type => "rcs", id => "rcs_find_changes", call => \&rcs_find_changes); + hook(type => "rcs", id => "rcs_get_current_rev", call => \&rcs_get_current_rev); } sub checkconfig () { @@ -40,17 +45,23 @@ sub checkconfig () { push @{$config{wrappers}}, { wrapper => $config{git_wrapper}, wrappermode => (defined $config{git_wrappermode} ? $config{git_wrappermode} : "06755"), + wrapper_background_command => $config{git_wrapper_background_command}, }; } if (defined $config{git_test_receive_wrapper} && - length $config{git_test_receive_wrapper}) { + length $config{git_test_receive_wrapper} && + defined $config{untrusted_committers} && + @{$config{untrusted_committers}}) { push @{$config{wrappers}}, { test_receive => 1, wrapper => $config{git_test_receive_wrapper}, wrappermode => (defined $config{git_wrappermode} ? $config{git_wrappermode} : "06755"), }; } + + # Avoid notes, parser does not handle and they only slow things down. + $ENV{GIT_NOTES_REF}=""; # Run receive test only if being called by the wrapper, and not # when generating same. @@ -65,6 +76,7 @@ sub getsetup () { plugin => { safe => 0, # rcs plugin rebuild => undef, + section => "rcs", }, git_wrapper => { type => "string", @@ -73,6 +85,13 @@ sub getsetup () { safe => 0, # file rebuild => 0, }, + git_wrapper_background_command => { + type => "string", + example => "git push github", + description => "shell command for git_wrapper to run, in the background", + safe => 0, # command + rebuild => 0, + }, git_wrappermode => { type => "string", example => '06755', @@ -96,7 +115,7 @@ sub getsetup () { }, historyurl => { type => "string", - example => "http://git.example.com/gitweb.cgi?p=wiki.git;a=history;f=[[file]]", + example => "http://git.example.com/gitweb.cgi?p=wiki.git;a=history;f=[[file]];hb=HEAD", description => "gitweb url to show file history ([[file]] substituted)", safe => 1, rebuild => 1, @@ -134,11 +153,23 @@ sub genwrapper { } } +my $git_dir=undef; +my $prefix=undef; + +sub in_git_dir ($$) { + $git_dir=shift; + my @ret=shift->(); + $git_dir=undef; + $prefix=undef; + return @ret; +} + sub safe_git (&@) { - # Start a child process safely without resorting /bin/sh. - # Return command output or success state (in scalar context). + # Start a child process safely without resorting to /bin/sh. + # Returns command output (in list content) or success state + # (in scalar context), or runs the specified data handler. - my ($error_handler, @cmdline) = @_; + my ($error_handler, $data_handler, @cmdline) = @_; my $pid = open my $OUT, "-|"; @@ -147,9 +178,13 @@ sub safe_git (&@) { if (!$pid) { # In child. # Git commands want to be in wc. - if (! $no_chdir) { + if (! defined $git_dir) { chdir $config{srcdir} - or error("Cannot chdir to $config{srcdir}: $!"); + or error("cannot chdir to $config{srcdir}: $!"); + } + else { + chdir $git_dir + or error("cannot chdir to $git_dir: $!"); } exec @cmdline or error("Cannot exec '@cmdline': $!"); } @@ -166,7 +201,12 @@ sub safe_git (&@) { chomp; - push @lines, $_; + if (! defined $data_handler) { + push @lines, $_; + } + else { + last unless $data_handler->($_); + } } close $OUT; @@ -176,9 +216,9 @@ sub safe_git (&@) { return wantarray ? @lines : ($? == 0); } # Convenient wrappers. -sub run_or_die ($@) { safe_git(\&error, @_) } -sub run_or_cry ($@) { safe_git(sub { warn @_ }, @_) } -sub run_or_non ($@) { safe_git(undef, @_) } +sub run_or_die ($@) { safe_git(\&error, undef, @_) } +sub run_or_cry ($@) { safe_git(sub { warn @_ }, undef, @_) } +sub run_or_non ($@) { safe_git(undef, undef, @_) } sub merge_past ($$$) { @@ -275,15 +315,36 @@ sub merge_past ($$$) { return $conflict; } -sub parse_diff_tree ($@) { +sub decode_git_file ($) { + my $file=shift; + + # git does not output utf-8 filenames, but instead + # double-quotes them with the utf-8 characters + # escaped as \nnn\nnn. + if ($file =~ m/^"(.*)"$/) { + ($file=$1) =~ s/\\([0-7]{1,3})/chr(oct($1))/eg; + } + + # strip prefix if in a subdir + if (! defined $prefix) { + ($prefix) = run_or_die('git', 'rev-parse', '--show-prefix'); + if (! defined $prefix) { + $prefix=""; + } + } + $file =~ s/^\Q$prefix\E//; + + return decode("utf8", $file); +} + +sub parse_diff_tree ($) { # Parse the raw diff tree chunk and return the info hash. # See git-diff-tree(1) for the syntax. - - my ($prefix, $dt_ref) = @_; + my $dt_ref = shift; # End of stream? - return if !defined @{ $dt_ref } || - !defined @{ $dt_ref }[0] || !length @{ $dt_ref }[0]; + return if ! @{ $dt_ref } || + !defined $dt_ref->[0] || !length $dt_ref->[0]; my %ci; # Header line. @@ -313,8 +374,9 @@ sub parse_diff_tree ($@) { $ci{ "${who}_epoch" } = $epoch; $ci{ "${who}_tz" } = $tz; - if ($name =~ m/^[^<]+\s+<([^@>]+)/) { - $ci{"${who}_username"} = $1; + if ($name =~ m/^([^<]+)\s+<([^@>]+)/) { + $ci{"${who}_name"} = $1; + $ci{"${who}_username"} = $2; } elsif ($name =~ m/^([^<]+)\s+<>$/) { $ci{"${who}_username"} = $1; @@ -362,16 +424,9 @@ sub parse_diff_tree ($@) { my $sha1_to = shift(@tmp); my $status = shift(@tmp); - # git does not output utf-8 filenames, but instead - # double-quotes them with the utf-8 characters - # escaped as \nnn\nnn. - if ($file =~ m/^"(.*)"$/) { - ($file=$1) =~ s/\\([0-7]{1,3})/chr(oct($1))/eg; - } - $file =~ s/^\Q$prefix\E//; if (length $file) { push @{ $ci{'details'} }, { - 'file' => decode("utf8", $file), + 'file' => decode_git_file($file), 'sha1_from' => $sha1_from[0], 'sha1_to' => $sha1_to, 'mode_from' => $mode_from[0], @@ -398,10 +453,9 @@ sub git_commit_info ($;$) { my @raw_lines = run_or_die('git', 'log', @opts, '--pretty=raw', '--raw', '--abbrev=40', '--always', '-c', '-r', $sha1, '--', '.'); - my ($prefix) = run_or_die('git', 'rev-parse', '--show-prefix'); my @ci; - while (my $parsed = parse_diff_tree(($prefix or ""), \@raw_lines)) { + while (my $parsed = parse_diff_tree(\@raw_lines)) { push @ci, $parsed; } @@ -410,24 +464,70 @@ sub git_commit_info ($;$) { return wantarray ? @ci : $ci[0]; } -sub git_sha1 (;$) { - # Return head sha1sum (of given file). - my $file = shift || q{--}; +sub rcs_find_changes ($) { + my $oldrev=shift; + + # Note that git log will sometimes show files being added that + # don't exist. Particularly, git merge -s ours can result in a + # merge commit where some files were not really added. + # This is why the code below verifies that the files really + # exist. + my @raw_lines = run_or_die('git', 'log', + '--pretty=raw', '--raw', '--abbrev=40', '--always', '-c', + '--no-renames', , '--reverse', + '-r', "$oldrev..HEAD", '--', '.'); + + # Due to --reverse, we see changes in chronological order. + my %changed; + my %deleted; + my $nullsha = 0 x 40; + my $newrev=$oldrev; + while (my $ci = parse_diff_tree(\@raw_lines)) { + $newrev=$ci->{sha1}; + foreach my $i (@{$ci->{details}}) { + my $file=$i->{file}; + if ($i->{sha1_to} eq $nullsha) { + if (! -e "$config{srcdir}/$file") { + delete $changed{$file}; + $deleted{$file}=1; + } + } + else { + if (-e "$config{srcdir}/$file") { + delete $deleted{$file}; + $changed{$file}=1; + } + } + } + } + + return (\%changed, \%deleted, $newrev); +} +sub git_sha1_file ($) { + my $file=shift; + git_sha1("--", $file); +} + +sub git_sha1 (@) { # Ignore error since a non-existing file might be given. my ($sha1) = run_or_non('git', 'rev-list', '--max-count=1', 'HEAD', - '--', $file); - if ($sha1) { + '--', @_); + if (defined $sha1) { ($sha1) = $sha1 =~ m/($sha1_pattern)/; # sha1 is untainted now - } else { debug("Empty sha1sum for '$file'.") } - return defined $sha1 ? $sha1 : q{}; + } + return defined $sha1 ? $sha1 : ''; +} + +sub rcs_get_current_rev () { + git_sha1(); } sub rcs_update () { # Update working directory. if (length $config{gitorigin_branch}) { - run_or_cry('git', 'pull', $config{gitorigin_branch}); + run_or_cry('git', 'pull', '--prune', $config{gitorigin_branch}); } } @@ -436,62 +536,89 @@ sub rcs_prepedit ($) { # This will be later used in rcs_commit if a merge is required. my ($file) = @_; - return git_sha1($file); + return git_sha1_file($file); } -sub rcs_commit ($$$;$$) { +sub rcs_commit (@) { # Try to commit the page; returns undef on _success_ and # a version of the page with the rcs's conflict markers on # failure. - - my ($file, $message, $rcstoken, $user, $ipaddr) = @_; + my %params=@_; # Check to see if the page has been changed by someone else since # rcs_prepedit was called. - my $cur = git_sha1($file); - my ($prev) = $rcstoken =~ /^($sha1_pattern)$/; # untaint + my $cur = git_sha1_file($params{file}); + my ($prev) = $params{token} =~ /^($sha1_pattern)$/; # untaint if (defined $cur && defined $prev && $cur ne $prev) { - my $conflict = merge_past($prev, $file, $dummy_commit_msg); + my $conflict = merge_past($prev, $params{file}, $dummy_commit_msg); return $conflict if defined $conflict; } - rcs_add($file); - return rcs_commit_staged($message, $user, $ipaddr); + return rcs_commit_helper(@_); } -sub rcs_commit_staged ($$$) { +sub rcs_commit_staged (@) { # Commits all staged changes. Changes can be staged using rcs_add, # rcs_remove, and rcs_rename. - my ($message, $user, $ipaddr)=@_; + return rcs_commit_helper(@_); +} - # Set the commit author and email to the web committer. +sub rcs_commit_helper (@) { + my %params=@_; + my %env=%ENV; - if (defined $user || defined $ipaddr) { - my $u=encode_utf8(defined $user ? $user : $ipaddr); - $ENV{GIT_AUTHOR_NAME}=$u; - $ENV{GIT_AUTHOR_EMAIL}="$u\@web"; + + if (defined $params{session}) { + # Set the commit author and email based on web session info. + my $u; + if (defined $params{session}->param("name")) { + $u=$params{session}->param("name"); + } + elsif (defined $params{session}->remote_addr()) { + $u=$params{session}->remote_addr(); + } + if (defined $u) { + $u=encode_utf8($u); + $ENV{GIT_AUTHOR_NAME}=$u; + } + if (defined $params{session}->param("nickname")) { + $u=encode_utf8($params{session}->param("nickname")); + $u=~s/\s+/_/g; + $u=~s/[^-_0-9[:alnum:]]+//g; + } + if (defined $u) { + $ENV{GIT_AUTHOR_EMAIL}="$u\@web"; + } } - $message = IkiWiki::possibly_foolish_untaint($message); + $params{message} = IkiWiki::possibly_foolish_untaint($params{message}); my @opts; - if ($message !~ /\S/) { + if ($params{message} !~ /\S/) { # Force git to allow empty commit messages. # (If this version of git supports it.) my ($version)=`git --version` =~ /git version (.*)/; - if ($version ge "1.5.4") { + if ($version ge "1.7.8") { + push @opts, "--allow-empty-message", "--no-edit"; + } + if ($version ge "1.7.2") { + push @opts, "--allow-empty-message"; + } + elsif ($version ge "1.5.4") { push @opts, '--cleanup=verbatim'; } else { - $message.="."; + $params{message}.="."; } } - push @opts, '-q'; - # git commit returns non-zero if file has not been really changed. - # so we should ignore its exit status (hence run_or_non). - if (run_or_non('git', 'commit', @opts, '-m', $message)) { + if (exists $params{file}) { + push @opts, '--', $params{file}; + } + # git commit returns non-zero if nothing really changed. + # So we should ignore its exit status (hence run_or_non). + if (run_or_non('git', 'commit', '-m', $params{message}, '-q', @opts)) { if (length $config{gitorigin_branch}) { - run_or_cry('git', 'push', $config{gitorigin_branch}); + run_or_cry('git', 'push', $config{gitorigin_branch}, $config{gitmaster_branch}); } } @@ -542,9 +669,12 @@ sub rcs_recentchanges ($) { my @pages; foreach my $detail (@{ $ci->{'details'} }) { my $file = $detail->{'file'}; + my $efile = join('/', + map { uri_escape_utf8($_) } split('/', $file) + ); my $diffurl = defined $config{'diffurl'} ? $config{'diffurl'} : ""; - $diffurl =~ s/\[\[file\]\]/$file/go; + $diffurl =~ s/\[\[file\]\]/$efile/go; $diffurl =~ s/\[\[sha1_parent\]\]/$ci->{'parent'}/go; $diffurl =~ s/\[\[sha1_from\]\]/$detail->{'sha1_from'}/go; $diffurl =~ s/\[\[sha1_to\]\]/$detail->{'sha1_to'}/go; @@ -566,7 +696,16 @@ sub rcs_recentchanges ($) { my $user=$ci->{'author_username'}; my $web_commit = ($ci->{'author'} =~ /\@web>/); - + my $nickname; + + # Set nickname only if a non-url author_username is available, + # and author_name is an url. + if ($user !~ /:\/\// && defined $ci->{'author_name'} && + $ci->{'author_name'} =~ /:\/\//) { + $nickname=$user; + $user=$ci->{'author_name'}; + } + # compatability code for old web commit messages if (! $web_commit && defined $messages[0] && @@ -579,6 +718,7 @@ sub rcs_recentchanges ($) { push @rets, { rev => $sha1, user => $user, + nickname => $nickname, committype => $web_commit ? "web" : "git", when => $when, message => [@messages], @@ -591,15 +731,19 @@ sub rcs_recentchanges ($) { return @rets; } -sub rcs_diff ($) { +sub rcs_diff ($;$) { my $rev=shift; + my $maxlines=shift; my ($sha1) = $rev =~ /^($sha1_pattern)$/; # untaint my @lines; - foreach my $line (run_or_non("git", "show", $sha1)) { - if (@lines || $line=~/^diff --git/) { - push @lines, $line."\n"; - } - } + my $addlines=sub { + my $line=shift; + return if defined $maxlines && @lines == $maxlines; + push @lines, $line."\n" + if (@lines || $line=~/^diff --git/); + return 1; + }; + safe_git(undef, $addlines, "git", "show", $sha1); if (wantarray) { return @lines; } @@ -608,23 +752,61 @@ sub rcs_diff ($) { } } +{ +my %time_cache; + +sub findtimes ($$) { + my $file=shift; + my $id=shift; # 0 = mtime ; 1 = ctime + + if (! keys %time_cache) { + my $date; + foreach my $line (run_or_die('git', 'log', + '--pretty=format:%at', + '--name-only', '--relative')) { + if (! defined $date && $line =~ /^(\d+)$/) { + $date=$line; + } + elsif (! length $line) { + $date=undef; + } + else { + my $f=decode_git_file($line); + + if (! $time_cache{$f}) { + $time_cache{$f}[0]=$date; # mtime + } + $time_cache{$f}[1]=$date; # ctime + } + } + } + + return exists $time_cache{$file} ? $time_cache{$file}[$id] : 0; +} + +} + sub rcs_getctime ($) { my $file=shift; - # Remove srcdir prefix - $file =~ s/^\Q$config{srcdir}\E\/?//; - my @sha1s = run_or_non('git', 'rev-list', 'HEAD', '--', $file); - my $ci = git_commit_info($sha1s[$#sha1s], 1); - my $ctime = $ci->{'author_epoch'}; - debug("ctime for '$file': ". localtime($ctime)); + return findtimes($file, 1); +} + +sub rcs_getmtime ($) { + my $file=shift; - return $ctime; + return findtimes($file, 0); } -sub rcs_receive () { +{ +my $ret; +sub git_find_root { # The wiki may not be the only thing in the git repo. # Determine if it is in a subdirectory by examining the srcdir, # and its parents, looking for the .git directory. + + return @$ret if defined $ret; + my $subdir=""; my $dir=$config{srcdir}; while (! -d "$dir/.git") { @@ -635,83 +817,140 @@ sub rcs_receive () { } } + $ret=[$subdir, $dir]; + return @$ret; +} + +} + +sub git_parse_changes { + my $reverted = shift; + my @changes = @_; + + my ($subdir, $rootdir) = git_find_root(); + my @rets; + foreach my $ci (@changes) { + foreach my $detail (@{ $ci->{'details'} }) { + my $file = $detail->{'file'}; + + # check that all changed files are in the subdir + if (length $subdir && + ! ($file =~ s/^\Q$subdir\E//)) { + error sprintf(gettext("you are not allowed to change %s"), $file); + } + + my ($action, $mode, $path); + if ($detail->{'status'} =~ /^[M]+\d*$/) { + $action="change"; + $mode=$detail->{'mode_to'}; + } + elsif ($detail->{'status'} =~ /^[AM]+\d*$/) { + $action= $reverted ? "remove" : "add"; + $mode=$detail->{'mode_to'}; + } + elsif ($detail->{'status'} =~ /^[DAM]+\d*/) { + $action= $reverted ? "add" : "remove"; + $mode=$detail->{'mode_from'}; + } + else { + error "unknown status ".$detail->{'status'}; + } + + # test that the file mode is ok + if ($mode !~ /^100[64][64][64]$/) { + error sprintf(gettext("you cannot act on a file with mode %s"), $mode); + } + if ($action eq "change") { + if ($detail->{'mode_from'} ne $detail->{'mode_to'}) { + error gettext("you are not allowed to change file modes"); + } + } + + # extract attachment to temp file + if (($action eq 'add' || $action eq 'change') && + ! pagetype($file)) { + eval q{use File::Temp}; + die $@ if $@; + my $fh; + ($fh, $path)=File::Temp::tempfile(undef, UNLINK => 1); + my $cmd = "cd $git_dir && ". + "git show $detail->{sha1_to} > '$path'"; + if (system($cmd) != 0) { + error("failed writing temp file '$path'."); + } + } + + push @rets, { + file => $file, + action => $action, + path => $path, + }; + } + } + + return @rets; +} + +sub rcs_receive () { my @rets; while (<>) { chomp; my ($oldrev, $newrev, $refname) = split(' ', $_, 3); - + # only allow changes to gitmaster_branch if ($refname !~ /^refs\/heads\/\Q$config{gitmaster_branch}\E$/) { error sprintf(gettext("you are not allowed to change %s"), $refname); } - + # Avoid chdir when running git here, because the changes # are in the master git repo, not the srcdir repo. - # The pre-recieve hook already puts us in the right place. - $no_chdir=1; - my @changes=git_commit_info($oldrev."..".$newrev); - $no_chdir=0; - - foreach my $ci (@changes) { - foreach my $detail (@{ $ci->{'details'} }) { - my $file = $detail->{'file'}; - - # check that all changed files are in the - # subdir - if (length $subdir && - ! ($file =~ s/^\Q$subdir\E//)) { - error sprintf(gettext("you are not allowed to change %s"), $file); - } + # (Also, if a subdir is involved, we don't want to chdir to + # it and only see changes in it.) + # The pre-receive hook already puts us in the right place. + in_git_dir(".", sub { + push @rets, git_parse_changes(0, git_commit_info($oldrev."..".$newrev)); + }); + } - my ($action, $mode, $path); - if ($detail->{'status'} =~ /^[M]+\d*$/) { - $action="change"; - $mode=$detail->{'mode_to'}; - } - elsif ($detail->{'status'} =~ /^[AM]+\d*$/) { - $action="add"; - $mode=$detail->{'mode_to'}; - } - elsif ($detail->{'status'} =~ /^[DAM]+\d*/) { - $action="remove"; - $mode=$detail->{'mode_from'}; - } - else { - error "unknown status ".$detail->{'status'}; - } - - # test that the file mode is ok - if ($mode !~ /^100[64][64][64]$/) { - error sprintf(gettext("you cannot act on a file with mode %s"), $mode); - } - if ($action eq "change") { - if ($detail->{'mode_from'} ne $detail->{'mode_to'}) { - error gettext("you are not allowed to change file modes"); - } - } - - # extract attachment to temp file - if (($action eq 'add' || $action eq 'change') && - ! pagetype($file)) { - eval q{use File::Temp}; - die $@ if $@; - my $fh; - ($fh, $path)=File::Temp::tempfile("XXXXXXXXXX", UNLINK => 1); - if (system("git show ".$detail->{sha1_to}." > '$path'") != 0) { - error("failed writing temp file"); - } - } + return reverse @rets; +} - push @rets, { - file => $file, - action => $action, - path => $path, - }; - } +sub rcs_preprevert ($) { + my $rev=shift; + my ($sha1) = $rev =~ /^($sha1_pattern)$/; # untaint + + # Examine changes from root of git repo, not from any subdir, + # in order to see all changes. + my ($subdir, $rootdir) = git_find_root(); + in_git_dir($rootdir, sub { + my @commits=git_commit_info($sha1, 1); + + if (! @commits) { + error "unknown commit"; # just in case } - } - return reverse @rets; + # git revert will fail on merge commits. Add a nice message. + if (exists $commits[0]->{parents} && + @{$commits[0]->{parents}} > 1) { + error gettext("you are not allowed to revert a merge"); + } + + git_parse_changes(1, @commits); + }); +} + +sub rcs_revert ($) { + # Try to revert the given rev; returns undef on _success_. + my $rev = shift; + my ($sha1) = $rev =~ /^($sha1_pattern)$/; # untaint + + if (run_or_non('git', 'revert', '--no-commit', $sha1)) { + return undef; + } + else { + run_or_die('git', 'reset', '--hard'); + return sprintf(gettext("Failed to revert commit %s"), $sha1); + } } 1