/[swish]/trunk/spider/filter.pm
This is repository of my old source code which isn't updated any more. Go to git.rot13.org for current projects!
ViewVC logotype

Diff of /trunk/spider/filter.pm

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 56 by dpavlin, Fri Jan 23 13:10:40 2004 UTC revision 57 by dpavlin, Sun Jan 25 16:49:50 2004 UTC
# Line 1  Line 1 
1  sub filter {  sub filter {
2          my $contents = shift || return;          my $contents = shift || return;
3    
4    #       my $verbose = 1;
5            
6          # if you don't want content to be indexed, include it in          # if you don't want content to be indexed, include it in
7          # <noindex> foobar </noindex> tags or surround it with comments          # <noindex> foobar </noindex> tags or surround it with comments
8          # <!-- noindex --> foobar <!-- /noindex -->          # <!-- noindex --> foobar <!-- /noindex -->
# Line 76  sub filter { Line 79  sub filter {
79    
80          $new_title =~ s/\s\s+/ /g;          $new_title =~ s/\s\s+/ /g;
81    
82          if ($contents =~ s,<title>(.*)</title>,$new_title,is) {          if ($contents =~ s,<title>(.*)</title>,<title>$new_title</title>,is) {
83                  print STDERR "replace title '$1' with '$new_title'\n";                  print STDERR "replace title '$1' with '$new_title'\n" if ($verbose);
84          } else {          } elsif (
85                  print STDERR "adding new title '$new_title'\n";                          # try to insert after <head>, <html> or at top
                 # try to insert after <head>, <html> or at top  
86                  $contents =~ s,(<head>),$1<title>$new_title</title>,is ||                  $contents =~ s,(<head>),$1<title>$new_title</title>,is ||
87                  $contents =~ s,(<html>),$1<title>$new_title</title>,is ||                  $contents =~ s,(<html>),$1<title>$new_title</title>,is ||
88                  $contents =~ s,^,<title>$new_title</title>,;                          $contents =~ s,^,<title>$new_title</title>, ) {
89                    print STDERR "adding new title '$new_title'\n" if ($verbose);
90          }          }
91    
92          return $contents;          return $contents;

Legend:
Removed from v.56  
changed lines
  Added in v.57

  ViewVC Help
Powered by ViewVC 1.1.26