gitweb: Better chopping in commit search results

When searching commit messages (commit search), if matched string is
too long, the generated HTML was munged leading to an ill-formed XHTML
document.

Now gitweb chop leading, trailing and matched parts, HTML escapes
those parts, then composes and marks up match info.  HTML output is
never chopped.  Limiting matched info to 80 columns (with slop) is now
done by dividing remaining characters after chopping match equally to
leading and trailing part, not by chopping composed and HTML marked
output.

Noticed-by: Jean-Baptiste Quenot <jbq@caraldi.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

diff --git a/gitweb/gitweb.perl b/gitweb/gitweb.perl
index 8ed6d04417d4076a54866407459137f7d742cca2..326e27cf88b1158313df704e5fb5c4f8bbd50a01 100755 (executable)
--- a/gitweb/gitweb.perl
+++ b/gitweb/gitweb.perl
@@ -3784,18 +3784,24 @@ sub git_search_grep_body {
                print "<td title=\"$co{'age_string_age'}\"><i>$co{'age_string_date'}</i></td>\n" .
                      "<td><i>" . $author . "</i></td>\n" .
                      "<td>" .
                print "<td title=\"$co{'age_string_age'}\"><i>$co{'age_string_date'}</i></td>\n" .
                      "<td><i>" . $author . "</i></td>\n" .
                      "<td>" .

-                     $cgi->a({-href => href(action=>"commit", hash=>$co{'id'}), -class => "list subject"},
-                              chop_and_escape_str($co{'title'}, 50) . "<br/>");
+                     $cgi->a({-href => href(action=>"commit", hash=>$co{'id'}),
+                              -class => "list subject"},
+                             chop_and_escape_str($co{'title'}, 50) . "<br/>");

                my $comment = $co{'comment'};
                foreach my $line (@$comment) {
                        if ($line =~ m/^(.*)($search_regexp)(.*)$/i) {
                my $comment = $co{'comment'};
                foreach my $line (@$comment) {
                        if ($line =~ m/^(.*)($search_regexp)(.*)$/i) {

-                               my $lead = esc_html($1) || "";
-                               $lead = chop_str($lead, 30, 10);
-                               my $match = esc_html($2) || "";
-                               my $trail = esc_html($3) || "";
-                               $trail = chop_str($trail, 30, 10);
-                               my $text = "$lead<span class=\"match\">$match</span>$trail";
-                               print chop_str($text, 80, 5) . "<br/>\n";
+                               my ($lead, $match, $trail) = ($1, $2, $3);
+                               $match = chop_str($match, 70, 5);       # in case match is very long
+                               my $contextlen = (80 - len($match))/2;  # is left for the remainder
+                               $contextlen = 30 if ($contextlen > 30); # but not too much
+                               $lead  = chop_str($lead,  $contextlen, 10);
+                               $trail = chop_str($trail, $contextlen, 10);
+
+                               $lead  = esc_html($lead);
+                               $match = esc_html($match);
+                               $trail = esc_html($trail);
+
+                               print "$lead<span class=\"match\">$match</span>$trail<br />";

                        }
                }
                print "</td>\n" .
                        }
                }
                print "</td>\n" .