[PATCH v2] Pull gitweb If-Modified-Since handling out into its own function and use for snapshots.

From: W. Trevor King <wking@drexel.edu>
Date: Mar 20, 2012, 16:40 UTC
Message-ID: <20120320164056.GA23093@odin.tremily.us>
In-Reply-To: <m37gyf4hc3.fsf@localhost.localdomain>

The current gitweb only generates Last-Modified and handles If-Modified-Since headers for the git_feed action. This patch breaks the Last-Modified and If-Modified-Since handling code out from git_feed into a new function modified_since. This makes the code easy to reuse for other actions where it is appropriate and adds the code to do that to git_snapshot.

Signed-off-by: W Trevor King <wking@drexel.edu>                                 
---
Thanks for the feedback, Jakub.  I believe I've addressed your points
in this revised patch.

Changed since v1:
- Fixed impersonal/personal commit message
- Simplified modified_since() -> die_if_unmodified().  Now it only
  takes a $latest_epoch argument and we set fewer headers.  It also
  jump straight to DONE_GITWEB if the page is unmodified.
- Ran tests (which passed this time).  Somehow I missed t/README
  earlier.

I'd add tests for my patch, but I'd have to have commit timestamps from the test repository, and I don't understand the test framework well enough to know how to do that cleanly.

Things that haven't changed:

Show 8 quoted lines

> > @@ -7038,9 +7071,11 @@ sub git_snapshot {
> >  	}
> >  
> >  	$filename =~ s/(["\\])/\\$1/g;
> > +	my %latest_date = parse_date($co{'committer_epoch'}, $co{'committer_tz'});
> ...
> This simply has no way of working, as %co variable you use here is
> not defined anywhere in git_snapshot() subroutine.

I defined it earlier:

Show 8 quoted lines

> > @@ -7029,6 +7054,14 @@ sub git_snapshot {
> >
> >   my ($name, $prefix) = snapshot_name($project, $hash);
> >   my $filename = "$name$known_snapshot_formats{$format}{'suffix'}";
> > +
> > + my %co = parse_commit($hash) or die_error(404, "Unknown commit object");
> > + if (! modified_since($known_snapshot_formats{$format}{'type'},
> > +                      $co{'committer_epoch'},

While modified_since() (now die_if_unmodified()) handles the 304 case, we also want to set Last-Modified for 200s. That's what the latter code is trying to do.

> What if there is no commit, for example if we are rewuesting snapshot
> of a tree by its SHA-1?  We need to be able to deal with sich
> situation, if only by not handling Last-Modified / If-Modified-Since.

The parse_commit() call should handle these cases.

diff --git a/gitweb/gitweb.perl b/gitweb/gitweb.perl
index a8b5fad..b944351 100755
--- a/gitweb/gitweb.perl
+++ b/gitweb/gitweb.perl
@@ -7003,6 +7003,28 @@ sub snapshot_name {
 	return wantarray ? ($name, $name) : $name;
 }
 
+sub die_if_unmodified {
+	my ($latest_epoch) = @_;
+	our $cgi;
+
+	my $if_modified = $cgi->http('IF_MODIFIED_SINCE');
+	if (defined $if_modified) {
+		my $since;
+		if (eval { require HTTP::Date; 1; }) {
+			$since = HTTP::Date::str2time($if_modified);
+		} elsif (eval { require Time::ParseDate; 1; }) {
+			$since = Time::ParseDate::parsedate($if_modified, GMT => 1);
+		}
+		if (defined $since && $latest_epoch <= $since) {
+			my %latest_date = parse_date($latest_epoch);
+			print $cgi->header(
+				-last_modified => $latest_date{'rfc2822'},
+				-status => '304 Not Modified');
+			goto DONE_GITWEB;
+		}
+	}
+}
+
 sub git_snapshot {
 	my $format = $input_params{'snapshot_format'};
 	if (!@snapshot_fmts) {
@@ -7029,6 +7051,10 @@ sub git_snapshot {
 
 	my ($name, $prefix) = snapshot_name($project, $hash);
 	my $filename = "$name$known_snapshot_formats{$format}{'suffix'}";
+
+	my %co = parse_commit($hash) or die_error(404, "Unknown commit object");
+	die_if_unmodified($co{'committer_epoch'});
+
 	my $cmd = quote_command(
 		git_cmd(), 'archive',
 		"--format=$known_snapshot_formats{$format}{'format'}",
@@ -7038,9 +7064,11 @@ sub git_snapshot {
 	}
 
 	$filename =~ s/(["\\])/\\$1/g;
+	my %latest_date = parse_date($co{'committer_epoch'}, $co{'committer_tz'});
 	print $cgi->header(
 		-type => $known_snapshot_formats{$format}{'type'},
 		-content_disposition => 'inline; filename="' . $filename . '"',
+		-last_modified => $latest_date{'rfc2822'},
 		-status => '200 OK');
 
 	open my $fd, "-|", $cmd
@@ -7820,24 +7848,8 @@ sub git_feed {
 	if (defined($commitlist[0])) {
 		%latest_commit = %{$commitlist[0]};
 		my $latest_epoch = $latest_commit{'committer_epoch'};
+		die_if_unmodified($latest_epoch);
 		%latest_date   = parse_date($latest_epoch, $latest_commit{'comitter_tz'});
-		my $if_modified = $cgi->http('IF_MODIFIED_SINCE');
-		if (defined $if_modified) {
-			my $since;
-			if (eval { require HTTP::Date; 1; }) {
-				$since = HTTP::Date::str2time($if_modified);
-			} elsif (eval { require Time::ParseDate; 1; }) {
-				$since = Time::ParseDate::parsedate($if_modified, GMT => 1);
-			}
-			if (defined $since && $latest_epoch <= $since) {
-				print $cgi->header(
-					-type => $content_type,
-					-charset => 'utf-8',
-					-last_modified => $latest_date{'rfc2822'},
-					-status => '304 Not Modified');
-				return;
-			}
-		}
 		print $cgi->header(
 			-type => $content_type,
 			-charset => 'utf-8',
--                                                                              
1.7.3.4