/[Grep]/bin/reindex.pl
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 /bin/reindex.pl

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

revision 110 by dpavlin, Wed Mar 14 20:02:19 2007 UTC revision 128 by dpavlin, Sun Apr 29 00:48:04 2007 UTC
# Line 7  use strict; Line 7  use strict;
7  use lib 'lib';  use lib 'lib';
8    
9  use Jifty;  use Jifty;
 use Lucene;  
10  use Grep::Search;  use Grep::Search;
11  use Data::Dump qw/dump/;  use Data::Dump qw/dump/;
12    use Text::DeDuper;
13    use Encode;
14    
15    my $remove_duplicate = 1;
16    
17    $|=1;
18    
19  BEGIN { Jifty->new; };  BEGIN { Jifty->new; };
20    
21  my $coll = Grep::Model::ItemCollection->new( results_are_readable => 1 );  my $system_user = Grep::CurrentUser->superuser;
22    my $coll = Grep::Model::ItemCollection->new( results_are_readable => 1, current_user => $system_user );
23  $coll->unlimit;  $coll->unlimit;
24    
25  print "indexing ", $coll->count, " items ";  print "indexing ", $coll->count, " items ";
26    
27  my $search = Grep::Search->new();  my $search = Grep::Search->new();
28    my $deduper = Text::DeDuper->new();
29    
30    my ( $total, $duplicates ) = ( 0, 0 );
31    
32  while ( my $i = $coll->next ) {  while ( my $i = $coll->next ) {
33    
34          $search->add( $i, $i->in_feed->owner->id );          my $c = encode('utf-8', $i->content);
35    
36          print $i->id, ' ';          if ( $remove_duplicate && $deduper->find_similar( $c ) ) {
37                    $i->delete;
38                    print "-",$i->id,"- ";
39                    $duplicates++;
40            } else {
41                    $search->add( $i, $i->in_feed->owner->id );
42                    print $i->id;
43                    $deduper->add_doc( $i->id, $c );
44                    print ' ';
45            }
46            $total++;
47  }  }
48    
49  print "\n";  print "$total records indexed", $remove_duplicate ? " ($duplicates duplicates)" : "", "\n";
50    
51  $search->finish;  $search->finish;

Legend:
Removed from v.110  
changed lines
  Added in v.128

  ViewVC Help
Powered by ViewVC 1.1.26