/[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 352 by dpavlin, Tue Jun 15 22:40:07 2004 UTC revision 410 by dpavlin, Sun Sep 5 21:40:57 2004 UTC
# Line 2  Line 2 
2    
3  =head1 NAME  =head1 NAME
4    
5  all2all.pl - basic script for all WebPac needs  all2all.pl - basic script for all WebPAC needs
6    
7  =cut  =cut
8    
9  use strict;  use strict;
 use OpenIsis;  
 use Text::Iconv;  
10  use Data::Dumper;  use Data::Dumper;
11  use Carp;  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,
21    #       debug => 1,
22  ) || die;  ) || die;
23    
24    my $log = $webpac->_get_logger();
25    
26    my $index = new WebPAC::jsFind(
27            index_path => './out/index',
28            keys => 80,
29    ) || die;
30    
31    my $thes;
32    
33  $|=1;  $|=1;
34    
 print "reading database\n";  
35  my $maxmfn = $webpac->open_isis(  my $maxmfn = $webpac->open_isis(
36          filename => shift @ARGV || '/data/hidra/THS-500/THS',          filename => shift @ARGV || '/data/hidra/THS/THS',
37          lookup => [          lookup => [
38          { 'key' => 'd:v900', 'val' => 'v250^a' },          { 'key' => 'd:v900', 'val' => 'v250^a' },
39  #       { 'eval' => '"v901^a" eq "Područje"', 'key' => 'pa:v561^4:v562^4:v461^1', 'val' => 'v900' },  #       { 'eval' => '"v901^a" eq "Područje"', 'key' => 'pa:v561^4:v562^4:v461^1', 'val' => 'v900' },
40  #       { 'eval '=> '"v901^a" eq "Mikrotezaurus"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },  #       { 'eval '=> '"v901^a" eq "Mikrotezaurus"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },
41  #       { 'eval' => '"v901^a" eq "Deskriptor"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },  #       { 'eval' => '"v901^a" eq "Deskriptor"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },
42          { 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },          { 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },
43            { 'key' => '900_mfn:v900', 'val' => 'v000' },
44          ],          ],
45  );  );
46    
47  print "rows: $maxmfn\n\n";  $log->info("rows: $maxmfn");
48    
49  for (my $mfn = 1; $mfn <= $maxmfn; $mfn++) {  $webpac->open_import_xml(type => 'isis_hidra_ths');
         my $rec = $webpac->{'data'}->{$mfn} || die "no record with mfn $mfn";  
50    
51          print "-- ",$webpac->fill_in($rec,'v250^a (v901^a)'),"\n";  while (my $rec = $webpac->fetch_rec) {
52    
53          my @t = $webpac->fill_in($rec,'v553^1;;v553^a');          my @ds = $webpac->data_structure($rec);
         print " Uži pojam: ",join("\t\n",@t),"\n" if (@t);  
54    
55          @t = $webpac->fill_in($rec,'[a:v251::];;[d:[a:v251::]]');          if ($log->is_debug) {
56          @t = map { $webpac->lookup($_) => $_ } @t;                  $log->debug("rec = ",Dumper($rec));
57          print " Područje: ",join("\t\n",@t),"\n" if (@t);                  $log->debug("ds = ",Dumper(\@ds));
58            }
59    
60            next if (! @ds);
61    
62            my $filename = $webpac->{'current_filename'};
63    
64            if ($filename) {
65                    open(OUT,"> $filename") || $log->logdie("can't open output '$filename': $!");
66                    print OUT $webpac->output(
67                            template => 'html.tt',
68                            data => \@ds,
69                            headline => $webpac->{'headline'},
70                    );
71                    close(OUT);
72            } else {
73                    print $webpac->output(
74                            template => 'text.tt',
75                            data => \@ds,
76                            headline => $webpac->{'headline'},
77                    );
78            }
79    
80            my $headline = $webpac->{'headline'};
81    
82            my $f = $filename;
83            $f =~ s!out/!!;
84    
85            # save into index
86            foreach my $ds (@ds) {
87                    next if (! $ds->{'swish'});
88    
89                    $index->insert(
90                            index_name => $ds->{'tag'},
91                            path => $f,
92                            headline => $headline,
93                            words => join(" ",@{$ds->{'swish'}})
94                    );
95            }
96    
97            # save into sorted index (thesaurus)
98            foreach my $ds (@ds) {
99                    next if (! $ds->{'index'});
100    
101                    $thes->{$ds->{'tag'}} ||= new WebPAC::Index;
102    
103                    $thes->{$ds->{'tag'}}->insert(
104                            path => $f,
105                            headline => $headline,
106                    );
107            }
108    
109          @t = $webpac->fill_in($rec,'[a:v561^4:v251:];;[d:[a:v561^4:v251:]]');  #       print Dumper(\@ds);
         print " Mikrotezaurus: ",join("\t\n",@t),"\n" if (@t);  
110    
         @t = $webpac->fill_in($rec,'[a:v561^4:v562^4:v900];;[d:[a:v561^4:v562^4:v900]]');  
         print " Deskriptor: ",join("\t\n",@t),"\n" if (@t);  
111  }  }
112    
113  print "## lookup ",Dumper($webpac->{'lookup'});  if ($log->is_debug) {
114  print "## data ",Dumper($webpac->{'data'});          $log->debug("lookup hash: ",Dumper($webpac->{'lookup'}));
115            $log->debug("data hash: ",Dumper($webpac->{'data'}));
116            foreach my $t (keys %{$thes}) {
117                    $log->debug("thesaurus $t hash: ",Dumper($thes->{$t}));
118            }
119    }
120    
121    $index->close;
122    

Legend:
Removed from v.352  
changed lines
  Added in v.410

  ViewVC Help
Powered by ViewVC 1.1.26