/[webpac]/trunk2/all2all.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 /trunk2/all2all.pl

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

revision 389 by dpavlin, Tue Jul 20 17:15:48 2004 UTC revision 413 by dpavlin, Tue Sep 7 18:02:06 2004 UTC
# Line 12  use Carp; Line 12  use Carp;
12    
13  use lib './lib';  use lib './lib';
14  use WebPAC;  use WebPAC;
15    use WebPAC::jsFind;
16    use WebPAC::Index;
17    
18  my $webpac = new WebPAC(  my $webpac = new WebPAC(
19          code_page => 'ISO-8859-2',          code_page => 'ISO-8859-2',
20          limit_mfn => 100,  #       limit_mfn => 500,
21  #       debug => 1,  #       debug => 1,
22  ) || die;  ) || die;
23    
24  my $log = $webpac->_get_logger();  my $log = $webpac->_get_logger();
25    
26    my $index = new WebPAC::jsFind(
27            index_path => './out/index',
28            keys => 10,
29    ) || die;
30    
31    my $thes;
32    
33  $|=1;  $|=1;
34    
35  my $maxmfn = $webpac->open_isis(  my $maxmfn = $webpac->open_isis(
# Line 44  while (my $rec = $webpac->fetch_rec) { Line 52  while (my $rec = $webpac->fetch_rec) {
52    
53          my @ds = $webpac->data_structure($rec);          my @ds = $webpac->data_structure($rec);
54    
55          if ($log->is_debug) {          if (0 && $log->is_debug) {
56                  $log->debug("rec = ",Dumper($rec));                  $log->debug("rec = ",Dumper($rec));
57                  $log->debug("ds = ",Dumper(\@ds));                  $log->debug("ds = ",Dumper(\@ds));
58          }          }
# Line 54  while (my $rec = $webpac->fetch_rec) { Line 62  while (my $rec = $webpac->fetch_rec) {
62          my $filename = $webpac->{'current_filename'};          my $filename = $webpac->{'current_filename'};
63    
64          if ($filename) {          if ($filename) {
65                  open(OUT,"> $filename") || $log->logdie("can't open output '$filename': $!");                  $webpac->output_file(
66                  print OUT $webpac->output(                          file => $filename,
67                          template => 'html.tt',                          template => 'html.tt',
68                          data => \@ds,                          data => \@ds,
69                            headline => $webpac->{'headline'},
70                  );                  );
                 close(OUT);  
71          } else {          } else {
72                  print $webpac->output(                  print $webpac->output(
73                          template => 'text.tt',                          template => 'text.tt',
74                          data => \@ds,                          data => \@ds,
75                            headline => $webpac->{'headline'},
76                    );
77            }
78    
79            my $headline = $webpac->{'headline'};
80    
81            my $f = $filename;
82            $f =~ s!out/!!;
83    
84            # save into index
85            foreach my $ds (@ds) {
86                    next if (! $ds->{'swish'});
87    
88                    $index->insert(
89                            index_name => $ds->{'tag'},
90                            path => $f,
91                            headline => $headline,
92                            words => join(" ",@{$ds->{'swish'}})
93                    );
94            }
95    
96            # save into sorted index (thesaurus)
97            foreach my $ds (@ds) {
98                    next if (! $ds->{'index'});
99    
100                    $thes->{$ds->{'tag'}} ||= new WebPAC::Index;
101    
102                    $thes->{$ds->{'tag'}}->insert(
103                            path => $f,
104                            headline => join(" ",@{$ds->{'index'}}),
105                  );                  );
106          }          }
107    
108    #       print Dumper(\@ds);
109    
110  }  }
111    
112  if ($log->is_debug) {  foreach my $t (keys %{$thes}) {
113    
114            my @e = $thes->{$t}->elements;
115            if (! @e) {
116                    $log->logwarn("no elements in sorted index $t?");
117                    next;
118            }
119    
120            $log->info("saving sorted index $t [".scalar(@e)." elements]");
121    
122            $webpac->output_file(
123                    file => "./out/bfilter/$t.txt",
124                    template => 'index.tt',
125                    data => \@e,
126                    index_name => $t,
127            );
128    }
129    
130    if (0 && $log->is_debug) {
131          $log->debug("lookup hash: ",Dumper($webpac->{'lookup'}));          $log->debug("lookup hash: ",Dumper($webpac->{'lookup'}));
132          $log->debug("data hash: ",Dumper($webpac->{'data'}));          $log->debug("data hash: ",Dumper($webpac->{'data'}));
133            foreach my $t (keys %{$thes}) {
134                    $log->debug("thesaurus $t hash: ",Dumper($thes->{$t}));
135            }
136  }  }
137    
138    $index->close;
139    
Legend:
Removed from v.389  
changed lines
  Added in v.413
  ViewVC Help
Powered by ViewVC 1.1.26