misc/translator/text-extract.pl

   1 #!/usr/bin/perl
   2 use HTML::Tree;
   3 use Getopt::Std;
   4 getopt("f:");
   5         my $tree = HTML::TreeBuilder->new; # empty tree
   6
   7         $tree->parse_file($opt_f);
   8         sub give_id {
   9                 my $x = $_[0];
  10                 foreach my $c ($x->content_list) {
  11                         next if (ref($c) && $c->tag() eq "~comment");
  12                         next if (ref($c) && $c->tag() eq "script");
  13                         next if (ref($c) && $c->tag() eq "style");
  14                         if (!ref($c)) {
  15                                 print "$c\n";
  16                         }
  17                         if (ref($c) && $c->attr('alt')) {
  18                                 print $c->attr('alt')."\n";
  19                         }
  20                         if (ref($c) && $c->attr('title')) {
  21                                 print $c->attr('title')."\n";
  22                         }
  23                         if (ref($c) && $c->tag() eq "input" && $c->attr('value')) {
  24                                 print $c->attr('value')."\n";
  25                         }
  26                         if (ref($c) && $c->tag() eq 'meta') {
  27                                 print $c->attr('content')."\n ";
  28                         }
  29                         give_id($c) if ref $c; # ignore text nodes
  30                 }
  31         };
  32         give_id($tree);
  33         $tree = $tree->delete;