X-Git-Url: http://git.vanrenterghem.biz/git.ikiwiki.info.git/blobdiff_plain/c2f01e320df8a27c0ac16fb38fba099925dbd010..b6774bc6afac48e2c0a122647edc1485e7174a6e:/IkiWiki/Plugin/git.pm

diff --git a/IkiWiki/Plugin/git.pm b/IkiWiki/Plugin/git.pm
index f5101d904..7a6a55495 100644
--- a/IkiWiki/Plugin/git.pm
+++ b/IkiWiki/Plugin/git.pm
@@ -5,11 +5,12 @@ use warnings;
 use strict;
 use IkiWiki;
 use Encode;
+use File::Path qw{remove_tree};
+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 $git_dir=undef;
 
 sub import {
 	hook(type => "checkconfig", id => "git", call => \&checkconfig);
@@ -29,6 +30,8 @@ sub import {
 	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 () {
@@ -151,11 +154,57 @@ sub genwrapper {
 	}
 }
 
-sub safe_git (&@) {
-	# Start a child process safely without resorting /bin/sh.
-	# Return command output or success state (in scalar context).
+my @git_dir_stack;
+my $prefix;
+
+sub in_git_dir ($$) {
+	unshift @git_dir_stack, shift;
+	my @ret=shift->();
+	shift @git_dir_stack;
+	$prefix=undef;
+	return @ret;
+}
+
+# Loosely based on git-new-workdir from git contrib.
+sub create_temp_working_dir ($$) {
+	my $rootdir = shift;
+	my $branch = shift;
+	my $working = "$rootdir/.git/ikiwiki-temp-working";
+	remove_tree($working);
+
+	foreach my $dir ("", ".git") {
+		if (!mkdir("$working/$dir")) {
+			error("Unable to create $working/$dir: $!");
+		}
+	}
+
+	# Hooks are deliberately not included: we will commit to the temporary
+	# branch that is used in the temporary working tree, and we don't want
+	# to run the post-commit hook there.
+	#
+	# logs/refs is not included because we don't use the reflog.
+	# remotes, rr-cache, svn are similarly excluded.
+	foreach my $link ("config", "refs", "objects", "info", "packed-refs") {
+		if (!symlink("../../$link", "$working/.git/$link")) {
+			error("Unable to create symlink $working/.git/$link: $!");
+		}
+	}
+
+	open (my $out, '>', "$working/.git/HEAD") or
+		error("failed to write $working.git/HEAD: $!");
+	print $out "ref: refs/heads/$branch\n" or
+		error("failed to write $working.git/HEAD: $!");
+	close $out or
+		error("failed to write $working.git/HEAD: $!");
+	return $working;
+}
 
-	my ($error_handler, @cmdline) = @_;
+sub safe_git {
+	# 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 %params = @_;
 
 	my $pid = open my $OUT, "-|";
 
@@ -164,15 +213,20 @@ sub safe_git (&@) {
 	if (!$pid) {
 		# In child.
 		# Git commands want to be in wc.
-		if (! defined $git_dir) {
+		if (! @git_dir_stack) {
 			chdir $config{srcdir}
 			    or error("cannot chdir to $config{srcdir}: $!");
 		}
 		else {
-			chdir $git_dir
-			    or error("cannot chdir to $git_dir: $!");
+			chdir $git_dir_stack[0]
+			    or error("cannot chdir to $git_dir_stack[0]: $!");
+		}
+
+		if ($params{stdout}) {
+			open(STDOUT, '>&', $params{stdout}) or error("Cannot reopen stdout: $!");
 		}
-		exec @cmdline or error("Cannot exec '@cmdline': $!");
+
+		exec @{$params{cmdline}} or error("Cannot exec '@{$params{cmdline}}': $!");
 	}
 	# In parent.
 
@@ -187,20 +241,40 @@ sub safe_git (&@) {
 
 		chomp;
 
-		push @lines, $_;
+		if (! defined $params{data_handler}) {
+			push @lines, $_;
+		}
+		else {
+			last unless $params{data_handler}->($_);
+		}
 	}
 
 	close $OUT;
 
-	$error_handler->("'@cmdline' failed: $!") if $? && $error_handler;
+	$params{error_handler}->("'@{$params{cmdline}}' failed: $!") if $? && $params{error_handler};
 
 	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 ($@) { safe_git(error_handler => \&error, cmdline => \@_) }
+sub run_or_cry ($@) { safe_git(error_handler => sub { warn @_ }, cmdline => \@_) }
+sub run_or_non ($@) { safe_git(cmdline => \@_) }
+
+sub ensure_committer {
+	if (! length $ENV{GIT_AUTHOR_NAME} || ! length $ENV{GIT_COMMITTER_NAME}) {
+		my $name = join('', run_or_non("git", "config", "user.name"));
+		if (! length $name) {
+			run_or_die("git", "config", "user.name", "IkiWiki");
+		}
+	}
 
+	if (! length $ENV{GIT_AUTHOR_EMAIL} || ! length $ENV{GIT_COMMITTER_EMAIL}) {
+		my $email = join('', run_or_non("git", "config", "user.email"));
+		if (! length $email) {
+			run_or_die("git", "config", "user.email", "ikiwiki.info");
+		}
+	}
+}
 
 sub merge_past ($$$) {
 	# Unlike with Subversion, Git cannot make a 'svn merge -rN:M file'.
@@ -239,6 +313,8 @@ sub merge_past ($$$) {
 	my @undo;      # undo stack for cleanup in case of an error
 	my $conflict;  # file content with conflict markers
 
+	ensure_committer();
+
 	eval {
 		# Hide local changes from Git by renaming the modified file.
 		# Relative paths must be converted to absolute for renaming.
@@ -296,8 +372,6 @@ sub merge_past ($$$) {
 	return $conflict;
 }
 
-{
-my $prefix;
 sub decode_git_file ($) {
 	my $file=shift;
 
@@ -319,7 +393,6 @@ sub decode_git_file ($) {
 
 	return decode("utf8", $file);
 }
-}
 
 sub parse_diff_tree ($) {
 	# Parse the raw diff tree chunk and return the info hash.
@@ -327,8 +400,8 @@ sub parse_diff_tree ($) {
 	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.
@@ -392,6 +465,16 @@ sub parse_diff_tree ($) {
 	}
 	shift @{ $dt_ref } if $dt_ref->[0] =~ /^$/;
 
+	$ci{details} = [parse_changed_files($dt_ref)];
+
+	return \%ci;
+}
+
+sub parse_changed_files {
+	my $dt_ref = shift;
+
+	my @files;
+
 	# Modified files.
 	while (my $line = shift @{ $dt_ref }) {
 		if ($line =~ m{^
@@ -409,7 +492,7 @@ sub parse_diff_tree ($) {
 			my $status = shift(@tmp);
 
 			if (length $file) {
-				push @{ $ci{'details'} }, {
+				push @files, {
 					'file'      => decode_git_file($file),
 					'sha1_from' => $sha1_from[0],
 					'sha1_to'   => $sha1_to,
@@ -423,7 +506,7 @@ sub parse_diff_tree ($) {
 		last;
 	}
 
-	return \%ci;
+	return @files;
 }
 
 sub git_commit_info ($;$) {
@@ -436,7 +519,7 @@ sub git_commit_info ($;$) {
 
 	my @raw_lines = run_or_die('git', 'log', @opts,
 		'--pretty=raw', '--raw', '--abbrev=40', '--always', '-c',
-		'-r', $sha1, '--', '.');
+		'-r', $sha1, '--no-renames', '--', '.');
 
 	my @ci;
 	while (my $parsed = parse_diff_tree(\@raw_lines)) {
@@ -448,25 +531,70 @@ sub git_commit_info ($;$) {
 	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.
 
+	ensure_committer();
+
 	if (length $config{gitorigin_branch}) {
 		run_or_cry('git', 'pull', '--prune', $config{gitorigin_branch});
 	}
@@ -477,7 +605,7 @@ sub rcs_prepedit ($) {
 	# 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 (@) {
@@ -488,24 +616,27 @@ sub rcs_commit (@) {
 
 	# Check to see if the page has been changed by someone else since
 	# rcs_prepedit was called.
-	my $cur    = git_sha1($params{file});
-	my ($prev) = $params{token} =~ /^($sha1_pattern)$/; # untaint
+	my $cur    = git_sha1_file($params{file});
+	my $prev;
+	if (defined $params{token}) {
+		($prev) = $params{token} =~ /^($sha1_pattern)$/; # untaint
+	}
 
 	if (defined $cur && defined $prev && $cur ne $prev) {
 		my $conflict = merge_past($prev, $params{file}, $dummy_commit_msg);
 		return $conflict if defined $conflict;
 	}
 
-	rcs_add($params{file});
-	return rcs_commit_staged(
-		message => $params{message},
-		session => $params{session},
-	);
+	return rcs_commit_helper(@_);
 }
 
 sub rcs_commit_staged (@) {
 	# Commits all staged changes. Changes can be staged using rcs_add,
 	# rcs_remove, and rcs_rename.
+	return rcs_commit_helper(@_);
+}
+
+sub rcs_commit_helper (@) {
 	my %params=@_;
 	
 	my %env=%ENV;
@@ -519,39 +650,55 @@ sub rcs_commit_staged (@) {
 		elsif (defined $params{session}->remote_addr()) {
 			$u=$params{session}->remote_addr();
 		}
-		if (defined $u) {
+		if (length $u) {
 			$u=encode_utf8($u);
 			$ENV{GIT_AUTHOR_NAME}=$u;
 		}
+		else {
+			$u = 'anonymous';
+		}
 		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) {
+		if (length $u) {
 			$ENV{GIT_AUTHOR_EMAIL}="$u\@web";
 		}
+		else {
+			$ENV{GIT_AUTHOR_EMAIL}='anonymous@web';
+		}
 	}
 
+	ensure_committer();
+
 	$params{message} = IkiWiki::possibly_foolish_untaint($params{message});
 	my @opts;
 	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 {
 			$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', $params{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});
 		}
 	}
 	
@@ -564,7 +711,9 @@ sub rcs_add ($) {
 
 	my ($file) = @_;
 
-	run_or_cry('git', 'add', $file);
+	ensure_committer();
+
+	run_or_cry('git', 'add', '--', $file);
 }
 
 sub rcs_remove ($) {
@@ -572,13 +721,17 @@ sub rcs_remove ($) {
 
 	my ($file) = @_;
 
-	run_or_cry('git', 'rm', '-f', $file);
+	ensure_committer();
+
+	run_or_cry('git', 'rm', '-f', '--', $file);
 }
 
 sub rcs_rename ($$) {
 	my ($src, $dest) = @_;
 
-	run_or_cry('git', 'mv', '-f', $src, $dest);
+	ensure_committer();
+
+	run_or_cry('git', 'mv', '-f', '--', $src, $dest);
 }
 
 sub rcs_recentchanges ($) {
@@ -602,9 +755,12 @@ sub rcs_recentchanges ($) {
 		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/\[\[sha1_to\]\]/$detail->{'sha1_to'}/go;
@@ -661,15 +817,23 @@ sub rcs_recentchanges ($) {
 	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(
+		error_handler => undef,
+		data_handler => $addlines,
+		cmdline => ["git", "show", $sha1],
+	);
 	if (wantarray) {
 		return @lines;
 	}
@@ -688,7 +852,7 @@ sub findtimes ($$) {
 	if (! keys %time_cache) {
 		my $date;
 		foreach my $line (run_or_die('git', 'log',
-				'--pretty=format:%ct',
+				'--pretty=format:%at',
 				'--name-only', '--relative')) {
 			if (! defined $date && $line =~ /^(\d+)$/) {
 				$date=$line;
@@ -750,6 +914,7 @@ sub git_find_root {
 }
 
 sub git_parse_changes {
+	my $reverted = shift;
 	my @changes = @_;
 
 	my ($subdir, $rootdir) = git_find_root();
@@ -770,11 +935,11 @@ sub git_parse_changes {
 				$mode=$detail->{'mode_to'};
 			}
 			elsif ($detail->{'status'} =~ /^[AM]+\d*$/) {
-				$action="add";
+				$action= $reverted ? "remove" : "add";
 				$mode=$detail->{'mode_to'};
 			}
 			elsif ($detail->{'status'} =~ /^[DAM]+\d*/) {
-				$action="remove";
+				$action= $reverted ? "add" : "remove";
 				$mode=$detail->{'mode_from'};
 			}
 			else {
@@ -797,12 +962,12 @@ sub git_parse_changes {
 				eval q{use File::Temp};
 				die $@ if $@;
 				my $fh;
-				($fh, $path)=File::Temp::tempfile("XXXXXXXXXX", UNLINK => 1);
-				my $cmd = "cd $git_dir && ".
-				          "git show $detail->{sha1_to} > '$path'";
-				if (system($cmd) != 0) {
-					error("failed writing temp file '$path'.");
-				}
+				($fh, $path)=File::Temp::tempfile(undef, UNLINK => 1);
+				safe_git(
+					error_handler => sub { error("failed writing temp file '$path': ".shift."."); },
+					stdout => $fh,
+					cmdline => ['git', 'show', $detail->{sha1_to}],
+				);
 			}
 
 			push @rets, {
@@ -832,9 +997,9 @@ sub rcs_receive () {
 		# (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.
-		$git_dir=".";
-		push @rets, git_parse_changes(git_commit_info($oldrev."..".$newrev));
-		$git_dir=undef;
+		in_git_dir(".", sub {
+			push @rets, git_parse_changes(0, git_commit_info($oldrev."..".$newrev));
+		});
 	}
 
 	return reverse @rets;
@@ -844,24 +1009,83 @@ sub rcs_preprevert ($) {
 	my $rev=shift;
 	my ($sha1) = $rev =~ /^($sha1_pattern)$/; # untaint
 
+	my @undo;      # undo stack for cleanup in case of an error
+
+	ensure_committer();
+
 	# Examine changes from root of git repo, not from any subdir,
 	# in order to see all changes.
 	my ($subdir, $rootdir) = git_find_root();
-	$git_dir=$rootdir;
-	my @commits=git_commit_info($sha1, 1);
-	$git_dir=undef;
+	return in_git_dir($rootdir, sub {
+		my @commits=git_commit_info($sha1, 1);
+	
+		if (! @commits) {
+			error "unknown commit"; # just in case
+		}
 
-	if (! @commits) {
-		error "unknown commit"; # just in case
-	}
+		# 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 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");
-	}
+		# Due to the presence of rename-detection, we cannot actually
+		# see what will happen in a revert without trying it.
+		# But we can guess, which is enough to rule out most changes
+		# that we won't allow reverting.
+		git_parse_changes(1, @commits);
 
-	return git_parse_changes(@commits);
+		my $failure;
+		my @ret;
+		eval {
+			my $branch = "ikiwiki_revert_${sha1}"; # supposed to be unique
+
+			push @undo, sub {
+				run_or_cry('git', 'branch', '-D', $branch) if $failure;
+			};
+			if (run_or_non('git', 'rev-parse', '--quiet', '--verify', $branch)) {
+				run_or_non('git', 'branch', '-D', $branch);
+			}
+			run_or_die('git', 'branch', $branch, $config{gitmaster_branch});
+
+			my $working = create_temp_working_dir($rootdir, $branch);
+
+			push @undo, sub {
+				remove_tree($working);
+			};
+
+			in_git_dir($working, sub {
+				run_or_die('git', 'checkout', '--quiet', '--force', $branch);
+				run_or_die('git', 'revert', '--no-commit', $sha1);
+				run_or_die('git', 'commit', '-m', "revert $sha1", '-a');
+			});
+
+			my @raw_lines;
+			@raw_lines = run_or_die('git', 'diff', '--pretty=raw',
+				'--raw', '--abbrev=40', '--always', '--no-renames',
+				"..${branch}");
+
+			my $ci = {
+				details => [parse_changed_files(\@raw_lines)],
+			};
+
+			@ret = git_parse_changes(0, $ci);
+		};
+		$failure = $@;
+
+		# Process undo stack (in reverse order).  By policy cleanup
+		# actions should normally print a warning on failure.
+		while (my $handle = pop @undo) {
+			$handle->();
+		}
+
+		if ($failure) {
+			my $message = sprintf(gettext("Failed to revert commit %s"), $sha1);
+			error("$message\n$failure\n");
+		}
+
+		return @ret;
+	});
 }
 
 sub rcs_revert ($) {
@@ -869,11 +1093,13 @@ sub rcs_revert ($) {
 	my $rev = shift;
 	my ($sha1) = $rev =~ /^($sha1_pattern)$/; # untaint
 
-	if (run_or_non('git', 'revert', '--no-commit', $sha1)) {
+	ensure_committer();
+
+	if (run_or_non('git', 'cherry-pick', '--no-commit', "ikiwiki_revert_$sha1")) {
 		return undef;
 	}
 	else {
-		run_or_die('git', 'reset', '--hard');
+		run_or_non('git', 'branch', '-D', "ikiwiki_revert_$sha1");
 		return sprintf(gettext("Failed to revert commit %s"), $sha1);
 	}
 }