/[webpac2]/trunk/lib/WebPAC/Normalize.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/lib/WebPAC/Normalize.pm

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

revision 22 by dpavlin, Sun Jul 17 22:48:25 2005 UTC revision 38 by dpavlin, Sat Nov 12 21:21:50 2005 UTC
# Line 2  package WebPAC::Normalize; Line 2  package WebPAC::Normalize;
2    
3  use warnings;  use warnings;
4  use strict;  use strict;
5    use base 'WebPAC::Common';
6  use Data::Dumper;  use Data::Dumper;
7    
8  =head1 NAME  =head1 NAME
# Line 78  Create new normalisation object Line 79  Create new normalisation object
79                          return length($_);                          return length($_);
80                  }, ...                  }, ...
81          },          },
82          db => $webpac_db_obj,          db => $db_obj,
83          lookup_regex => $lookup->regex,          lookup_regex => $lookup->regex,
84            lookup => $lookup_obj,
85    );    );
86    
87  Parametar C<filter> defines user supplied snippets of perl code which can  Parametar C<filter> defines user supplied snippets of perl code which can
88  be use with C<filter{...}> notation.  be use with C<filter{...}> notation.
89    
90  Recommended parametar C<lookup_regex> is used to enable parsing of lookups  Recommended parametar C<lookup_regex> is used to enable parsing of lookups
91  in structures.  in structures. If you pass this parametar, you must also pass C<lookup>
92    which is C<WebPAC::Lookup> object.
93    
94  =cut  =cut
95    
# Line 95  sub new { Line 98  sub new {
98          my $self = {@_};          my $self = {@_};
99          bless($self, $class);          bless($self, $class);
100    
101            my $r = $self->{'lookup_regex'} ? 1 : 0;
102            my $l = $self->{'lookup'} ? 1 : 0;
103    
104            my $log = $self->_get_logger();
105    
106            # those two must be in pair
107            if ( ($r & $l) != ($r || $l) ) {
108                    my $log = $self->_get_logger();
109                    $log->logdie("lookup_regex and lookup must be in pair");
110            }
111    
112            $log->logdie("lookup must be WebPAC::Lookup object") if ($self->{'lookup'} && ! $self->{'lookup'}->isa('WebPAC::Lookup'));
113    
114          $self ? return $self : return undef;          $self ? return $self : return undef;
115  }  }
116    
# Line 128  sub data_structure { Line 144  sub data_structure {
144    
145          if ($self->{'db'}) {          if ($self->{'db'}) {
146                  my @ds = $self->{'db'}->load_ds($rec);                  my @ds = $self->{'db'}->load_ds($rec);
147                  return @ds if (@ds);                  $log->debug("load_ds( rec = ", sub { Dumper($rec) }, ") = ", sub { Dumper(@ds) });
148                    return @ds if ($#ds > 0);
149                    $log->debug("cache miss, creating");
150          }          }
151    
152          undef $self->{'currnet_filename'};          undef $self->{'currnet_filename'};
# Line 153  sub data_structure { Line 171  sub data_structure {
171  #print "field $field [",$self->{'tag'},"] = ",Dumper($self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}});  #print "field $field [",$self->{'tag'},"] = ",Dumper($self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}});
172    
173                  foreach my $tag (@{$self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}}) {                  foreach my $tag (@{$self->{'import_xml'}->{'indexer'}->{$field}->{$self->{'tag'}}}) {
174                          my $format = $tag->{'value'} || $tag->{'content'};                          my $format;
175    
176                            $log->logdie("expected tag HASH and got $tag") unless (ref($tag) eq 'HASH');
177                            $format = $tag->{'value'} || $tag->{'content'};
178    
179                          $log->debug("format: $format");                          $log->debug("format: $format");
180    
# Line 242  sub data_structure { Line 263  sub data_structure {
263                  headline => $self->{'headline'},                  headline => $self->{'headline'},
264          ) if ($self->{'db'});          ) if ($self->{'db'});
265    
266            $log->debug("ds: ", sub { Dumper(@ds) });
267    
268          return @ds;          return @ds;
269    
270  }  }
# Line 438  sub fill_in { Line 461  sub fill_in {
461                  }                  }
462                  # do we have lookups?                  # do we have lookups?
463                  if ($self->{'lookup'}) {                  if ($self->{'lookup'}) {
464                          return $self->lookup($format);                          if ($self->{'lookup'}->can('lookup')) {
465                                    return $self->{'lookup'}->lookup($format);
466                            } else {
467                                    $log->warn("Have lookup object but can't invoke lookup method");
468                            }
469                  } else {                  } else {
470                          return $format;                          return $format;
471                  }                  }
# Line 564  sub apply_format { Line 591  sub apply_format {
591          $log->debug("using format $name [$format] on $data to produce: $out");          $log->debug("using format $name [$format] on $data to produce: $out");
592    
593          if ($self->{'lookup_regex'} && $out =~ $self->{'lookup_regex'}) {          if ($self->{'lookup_regex'} && $out =~ $self->{'lookup_regex'}) {
594                  return $self->lookup($out);                  return $self->{'lookup'}->lookup($out);
595          } else {          } else {
596                  return $out;                  return $out;
597          }          }

Legend:
Removed from v.22  
changed lines
  Added in v.38

  ViewVC Help
Powered by ViewVC 1.1.26