Disclaimer:Please excuse me if this is not the right place to post this, but I couldn’t figure the right place. (not Code section right?).

Earlier release of the text searcher module that you might be using has a ‘serious’ bug in it. You can find full discussion of the bug on this forum page. The bug basically wouldn’t allow you to delete indexed data of any document in your database. Moreover, if you are using any database other than MySQL you might also receive database error on erroneous use of a column whenever you invoke the delete_document() method. I thought of making a note of this on PerlMonks in hopes this may prevent needless hours of debugging and extra cause for a headache on your part if you are using this module in any of your existing applications.

For reference, here’s the patch to cure the bug:

diff -ur --exclude=CVS D:\vlad\programming\perl\lib_dev\DBIx-FullTextS
+earch-0.61/lib/DBIx/FullTextSearch/String.pm D:\vlad\programming\perl
+\lib_dev\DBIx-FullTextSearch-0.61_patched/lib/DBIx/FullTextSearch/Str
+ing.pm
--- D:\vlad\programming\perl\lib_dev\DBIx-FullTextSearch-0.61/lib/DBIx
+/FullTextSearch/String.pm    Fri Mar 23 17:00:38 2001
+++ D:\vlad\programming\perl\lib_dev\DBIx-FullTextSearch-0.61_patched/
+lib/DBIx/FullTextSearch/String.pm    Wed May 01 14:38:44 2002
@@ -56,6 +56,12 @@
     $self->SUPER::index_document($id, $data);
 }
 
+sub delete_document {
+  my ($self, $doc_id) = @_;
+  
+  $self->SUPER::delete_document($self->get_id_for_name($doc_id));
+}
+
 sub contains_hashref {
     my $self = shift;
     my $res = $self->SUPER::contains_hashref(@_);
[download]