/[webpac]/branches/tehnika/all2xml.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 /branches/tehnika/all2xml.pl

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

revision 291 by dpavlin, Sun Mar 14 20:34:24 2004 UTC revision 458 by dpavlin, Tue Sep 21 16:53:44 2004 UTC
# Line 5  use OpenIsis; Line 5  use OpenIsis;
5  use Getopt::Std;  use Getopt::Std;
6  use Data::Dumper;  use Data::Dumper;
7  use XML::Simple;  use XML::Simple;
 use Text::Unaccent 1.02;        # 1.01 won't compile on my platform,  
8  use Text::Iconv;  use Text::Iconv;
9  use Config::IniFiles;  use Config::IniFiles;
10  use Encode;  use Encode;
# Line 17  $|=1; Line 16  $|=1;
16    
17  my $config_file = $0;  my $config_file = $0;
18  $config_file =~ s/\.pl$/.conf/;  $config_file =~ s/\.pl$/.conf/;
19    $config_file = $ARGV[0] if ($ARGV[0] && -f $ARGV[0]);
20  die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file);  die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file);
21    
22  my $config;  my $config;
# Line 525  sub data2xml { Line 525  sub data2xml {
525                                          $swish_data =~ s/ +/ /g;                                          $swish_data =~ s/ +/ /g;
526                                          $swish_data =~ s/ +$//g;                                          $swish_data =~ s/ +$//g;
527    
528                                          $xml .= xmlify($field."_swish", unac_string($codepage,$swish_data));                                          $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
529                                  }                                  }
530    
531                                  my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page];                                  my $swish_exact_data = $cache->{swish_exact_data}->{$field}->[$page];
# Line 535  sub data2xml { Line 535  sub data2xml {
535    
536                                          # add delimiters before and after word.                                          # add delimiters before and after word.
537                                          # That is required to produce exact match                                          # That is required to produce exact match
538                                          $xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data));                                          $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
539                                  }                                  }
540                                                                    
541                                  my $idel = $cache->{index_delimiter}->{$field};                                  my $idel = $cache->{index_delimiter}->{$field};
# Line 568  sub data2xml { Line 568  sub data2xml {
568                                  $swish_data =~ s/ +/ /g;                                  $swish_data =~ s/ +/ /g;
569                                  $swish_data =~ s/ +$//g;                                  $swish_data =~ s/ +$//g;
570    
571                                  $xml .= xmlify($field."_swish", unac_string($codepage,$swish_data));                                  $xml .= xmlify($field."_swish", my_unac_string($codepage,$swish_data));
572                          }                          }
573    
574                          if ($swish_exact_data) {                          if ($swish_exact_data) {
# Line 577  sub data2xml { Line 577  sub data2xml {
577    
578                                  # add delimiters before and after word.                                  # add delimiters before and after word.
579                                  # That is required to produce exact match                                  # That is required to produce exact match
580                                  $xml .= xmlify($field."_swish_exact", unac_string($codepage,$swish_exact_data));                                  $xml .= xmlify($field."_swish_exact", my_unac_string($codepage,$swish_exact_data));
581                          }                          }
582                  }                  }
583          }          }
# Line 614  $index = new index_DBI( Line 614  $index = new index_DBI(
614    
615  my $show_progress = $cfg_global->val('global', 'show_progress');  my $show_progress = $cfg_global->val('global', 'show_progress');
616    
617  my $unac_filter = $cfg_global->val('global', 'unac_filter');  my $my_unac_filter = $cfg_global->val('global', 'my_unac_filter');
618  if ($unac_filter) {  if ($my_unac_filter) {
619          require $unac_filter;          print STDERR "using $my_unac_filter to filter characters for search\n";
620            require $my_unac_filter;
621    } else {
622            print STDERR "### fallback to default my_unac_string!\n";
623            eval q{
624            sub main::my_unac_string($$) {
625                    my ($charset, $string) = (@_);
626                    return $string;
627            }
628            };
629  }  }
630    
631  foreach my $database ($cfg->Sections) {  foreach my $database ($cfg->Sections) {
# Line 628  foreach my $database ($cfg->Sections) { Line 637  foreach my $database ($cfg->Sections) {
637          my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional          my $lookup_file = $cfg -> val($database, 'lookup_newfile'); # optional
638          if ($lookup_file) {          if ($lookup_file) {
639                  #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644;                  #tie %lhash, 'GDBM_File', $lookup_file, &GDBM_NEWDB, 0644;
640                    if (! -e $lookup_file) {
641                            open(LOOKUP, "> $lookup_file") || die "can't create $lookup_file': $!";
642                            close(LOOKUP);
643                    }
644                  tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644;                  tie %lhash, 'TDB_File', $lookup_file, TDB_CLEAR_IF_FIRST, O_RDWR, 0644;
645                  print STDERR "creating lookup file '$lookup_file'\n";                  print STDERR "creating lookup file '$lookup_file'\n";
646                  # delete memory cache for lookup file                  # delete memory cache for lookup file
# Line 918  __END__ Line 931  __END__
931    
932  all2xml.pl - read various file formats and dump XML for SWISH-E  all2xml.pl - read various file formats and dump XML for SWISH-E
933    
934    =head1 SYNOPSYS
935    
936     $ all2xml.pl [test.conf]
937    
938  =head1 DESCRIPTION  =head1 DESCRIPTION
939    
940  This command will read ISIS data file using OpenIsis perl module, MARC  This command will read ISIS data file using OpenIsis perl module, MARC
# Line 926  create one XML file for usage with I<SWI Line 943  create one XML file for usage with I<SWI
943  this script B<isn't general xml generator> from isis files (isis allready  this script B<isn't general xml generator> from isis files (isis allready
944  has something like that). Output of this script is tailor-made for SWISH-E.  has something like that). Output of this script is tailor-made for SWISH-E.
945    
946    If no configuration file is specified, it will use default one called
947    C<all2xml.conf>.
948    
949  =head1 BUGS  =head1 BUGS
950    
951  Documentation is really lacking. However, in true Open Source spirit, source  Documentation is really lacking. However, in true Open Source spirit, source

Legend:
Removed from v.291  
changed lines
  Added in v.458

  ViewVC Help
Powered by ViewVC 1.1.26