/[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 51 by dpavlin, Tue Jan 20 18:40:06 2004 UTC revision 69 by dpavlin, Thu Mar 18 23:07:21 2004 UTC
# Line 1  Line 1 
1  sub filter {  sub filter {
2          my $contents = shift || return;          my $contents = shift || return;
3    
4            my $verbose = 0;
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 37  sub filter { Line 40  sub filter {
40    
41          }          }
42    
43            # is second argument collection?
44            my $collection = shift || return $contents;
45    
46            # construct new title (from various parts of DocBook if available)
47            my $new_title;
48    
49            if ($contents =~ m,<!--SafTocEntry="([^"]+)"-->,is) {
50                    $new_title = $1;
51            } elsif ($contents =~ m,<(h\d)\sclass="docPartTitle"[^>]*>(.+?)<\1>,is) {
52                    $new_title = $2;
53            } elsif ($contents =~ m,<(h\d)\sclass="docChapterTitle"[^>]*>(.+?)<\1>,is) {
54                    $new_title = $2;
55            } elsif ($contents =~ m,<(h\d)\sclass="docSection1Title"[^>]*>(.+?)<\1>,is) {
56                    $new_title = $2;
57            } elsif ($contents =~ m,<(h\d)\sclass="chapter"[^>]*>(.+?)<\1>,is) {
58                    $new_title = $2;
59            } elsif ($contents =~ m,<(h\d)\sclass="sect1"[^>]*>(.+?)<\1>,is) {
60                    $new_title = $2;
61            } else {
62                    if ($contents =~ m,<title>([^<]+)</title>,is) {
63                            $new_title = $1;
64                    } elsif ($contents =~ m,<h\d[^>]*>([^<]+)</h\d>,is) {
65                            $new_title = $1;
66                    }
67            }
68    
69            if ($new_title) {
70                    # nuke html in title
71                    $new_title =~ s/<br>\s+/: /gs;
72                    $new_title =~ s/<\/*[^>]+>//gs;
73            
74                    # check if new title is same as collection name
75                    my ($a,$b) = ($new_title,$collection);
76                    $a =~ s/([^a-zA-Z])+/ /gs;
77                    $b =~ s/([^a-zA-Z])+/ /gs;
78                    if ($a =~ m/$b/i) {
79                            $new_title = $collection;
80                    } else {
81                            $new_title = "$collection :: $new_title";
82                    }
83            } else {
84                    # fall-back to collection title
85                    $new_title = $collection;
86            }
87    
88            $new_title =~ s/\s\s+/ /g;
89    
90            if ($contents =~ s,<title>(.*)</title>,<title>$new_title</title>,is) {
91                    print STDERR "replace title '$1' with '$new_title'\n" if ($verbose);
92            } elsif ($contents =~ s,(<head>),$1<title>$new_title</title>,is) {
93                    print STDERR "adding title '$new_title' after <head>\n" if ($verbose);
94            
95            } elsif ($contents =~ s,(<html>),$1<head><title>$new_title</title></head>,is) {
96                    print STDERR "adding title '$new_title' after <html>\n" if ($verbose);
97    
98            } elsif ($contents =~ s,^,<title>$new_title</title>,)  {
99                    print STDERR "adding new title '$new_title'\n" if ($verbose);
100            } else {
101                    print STDERR "WARNING: filter couldn't add new title '$new_title' anywhere!";
102            }
103    
104          return $contents;          return $contents;
105  }  }
106    

Legend:
Removed from v.51  
changed lines
  Added in v.69

  ViewVC Help
Powered by ViewVC 1.1.26