/[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 65 by dpavlin, Wed Mar 17 12:19:14 2004 UTC revision 71 by dpavlin, Sat Apr 3 15:15:36 2004 UTC
# Line 22  sub filter { Line 22  sub filter {
22          # remove comments between <html> and <head> texi2html inserts them          # remove comments between <html> and <head> texi2html inserts them
23          # there and swish can't find document title then (libxml or swish bug?)          # there and swish can't find document title then (libxml or swish bug?)
24          while ($contents =~ s/(<html>.*)<!--.*?-->(.*<head>)/$1$2/msi) { };          while ($contents =~ s/(<html>.*)<!--.*?-->(.*<head>)/$1$2/msi) { };
25            
26            # remove empty lines before/after <html>
27            $contents =~ s/^\s+(<html>)/$1/is;
28            $contents =~ s/(<\/html>)\s+$/$1/is;
29            # remove cr
30            $contents =~ s/\r//gs;
31    
32            # remove SQL Magazine header and footer
33            $contents =~ s/<!-- begin topnav area -->.+?<!-- end topnav area -->/<\/table>/is;
34            $contents =~ s/<!--Begin Footer-->.+?<\/table>/<\/table>/is;
35    
36          # remote TPJ left column          # remote TPJ left column
37          if ($contents =~ s,<!-- BEGIN LEFT SIDE BAR CELL -->.+?<!-- END LEFT SIDE BAR CELL -->,,isg) {          if ($contents =~ s,<!-- BEGIN LEFT SIDE BAR CELL -->.+?<!-- END LEFT SIDE BAR CELL -->,,isg) {
# Line 89  sub filter { Line 99  sub filter {
99    
100          if ($contents =~ s,<title>(.*)</title>,<title>$new_title</title>,is) {          if ($contents =~ s,<title>(.*)</title>,<title>$new_title</title>,is) {
101                  print STDERR "replace title '$1' with '$new_title'\n" if ($verbose);                  print STDERR "replace title '$1' with '$new_title'\n" if ($verbose);
102          } elsif (          } elsif ($contents =~ s,(<head>),$1<title>$new_title</title>,is) {
103                          # try to insert after <head>, <html> or at top                  print STDERR "adding title '$new_title' after <head>\n" if ($verbose);
104                  $contents =~ s,(<head>),$1<title>$new_title</title>,is ||          
105                  $contents =~ s,(<html>),$1<title>$new_title</title>,is ||          } elsif ($contents =~ s,(<html>),$1<head><title>$new_title</title></head>,is) {
106                          $contents =~ s,^,<title>$new_title</title>, ) {                  print STDERR "adding title '$new_title' after <html>\n" if ($verbose);
107    
108            } elsif ($contents =~ s,^,<title>$new_title</title>,)  {
109                  print STDERR "adding new title '$new_title'\n" if ($verbose);                  print STDERR "adding new title '$new_title'\n" if ($verbose);
110            } else {
111                    print STDERR "WARNING: filter couldn't add new title '$new_title' anywhere!";
112          }          }
113    
114          return $contents;          return $contents;
Legend:
Removed from v.65  
changed lines
  Added in v.71
  ViewVC Help
Powered by ViewVC 1.1.26