[popfile-commit] engine/Classifier MailParse.pm,1.90,1.91

[<ssc...@us...>]

Update of /cvsroot/popfile/engine/Classifier
In directory sc8-pr-cvs1:/tmp/cvs-serv3131/Classifier

Modified Files:
	MailParse.pm 
Log Message:
fixed possible handle leak in archive code. made spaced word detector more aggressive at word edges and updated spaced tests

Index: MailParse.pm
===================================================================
RCS file: /cvsroot/popfile/engine/Classifier/MailParse.pm,v
retrieving revision 1.90
retrieving revision 1.91
diff -C2 -d -r1.90 -r1.91
*** MailParse.pm	25 Feb 2003 02:04:12 -0000	1.90
--- MailParse.pm	27 Feb 2003 09:49:38 -0000	1.91
***************
*** 257,265 ****
              #       without using the encoded content printer or modifying $self->{ut}
  
!             foreach my $space (' ', '\'', '*', '^', '`' ){
!                 while ( $line =~ s/ (([A-Z]\Q$space\E){2,15}[A-Z])( |\Q$space\E|[!\?])/ /i ) {
!                     my $word = $1;
                      $word    =~ s/\Q$space\E//g;
!                     update_word( $self, $word, $encoded, ' ', ' ', '' );
                      increment_word( $self, 'trick:spacedout' );
                  }
--- 257,267 ----
              #       without using the encoded content printer or modifying $self->{ut}
  
!             foreach my $space (' ', '\'', '*', '^', '`', '  ', '\38' ){
!                 while ( $line =~ s/( |^)(([A-Z]\Q$space\E){2,15}[A-Z])( |\Q$space\E|[!\?])/ /i ) {
!                     my $word = $2;
!                     print "$word ->" if $self->{debug};
                      $word    =~ s/\Q$space\E//g;
!                     print "$word\n" if $self->{debug};
!                     update_word( $self, $word, $encoded, ' ', ' ', $prefix);
                      increment_word( $self, 'trick:spacedout' );
                  }