--- trunk/all2xml.pl 2003/06/23 20:20:32 54
+++ trunk2/all2all.pl 2004/09/09 18:09:21 419
@@ -1,380 +1,215 @@
#!/usr/bin/perl -w
-use strict;
-use OpenIsis;
-use Getopt::Std;
-use Data::Dumper;
-use XML::Simple;
-use Text::Unaccent 1.02; # 1.01 won't compile on my platform,
-use Text::Iconv;
-use Config::IniFiles;
-use Encode;
-
-$|=1;
-
-my $config_file = $0;
-$config_file =~ s/\.pl$/.conf/;
-die "FATAL: can't find configuration file '$config_file'" if (! -e $config_file);
-
-my $config;
-
-use index_DBI; # there is no other, right now ;-)
-my $index;
-
-my %opts;
-
-# usage:
-# -d directory name
-# -m multiple directories
-# -q quiet
-# -s run swish
-
-getopts('d:m:qs', \%opts);
+=head1 NAME
-my $path; # this is name of database
+all2all.pl - basic script for all WebPAC needs
-Text::Iconv->raise_error(1); # Conversion errors raise exceptions
+=cut
-# this is encoding of all files on disk, including import_xml/*.xml file and
-# filter/*.pm files! It will be used to store strings in perl internally!
-my $codepage = 'ISO-8859-2';
+use strict;
+use Data::Dumper;
+use Carp;
-my $utf2cp = Text::Iconv->new('UTF-8',$codepage);
-# this function will convert data from XML files to local encoding
-sub x {
- return $utf2cp->convert($_[0]);
-}
+use lib './lib';
+use WebPAC;
+use WebPAC::jsFind;
+use WebPAC::Index;
+
+my $webpac = new WebPAC(
+ code_page => 'ISO-8859-2',
+# limit_mfn => 500,
+# debug => 1,
+) || die;
+
+my $log = $webpac->_get_logger();
+
+my $index = new WebPAC::jsFind(
+ index_path => './out/index',
+ keys => 10,
+) || die;
-# decode isis/excel or other import codepage
-my $import2cp;
+my $thes;
-# outgoing xml must be in UTF-8
-my $cp2utf = Text::Iconv->new($codepage,'UTF-8');
+$|=1;
-# mapping between data type and tag which specify
-# format in XML file
-my %type2tag = (
- 'isis' => 'isis',
- 'excel' => 'column'
+my $maxmfn = $webpac->open_isis(
+ filename => shift @ARGV || '/data/hidra/THS/THS',
+ lookup => [
+ { 'key' => 'd:v900', 'val' => 'v250^a' },
+# { 'eval' => '"v901^a" eq "Područje"', 'key' => 'pa:v561^4:v562^4:v461^1', 'val' => 'v900' },
+# { 'eval '=> '"v901^a" eq "Mikrotezaurus"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },
+# { 'eval' => '"v901^a" eq "Deskriptor"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },
+ { 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },
+ { 'key' => '900_mfn:v900', 'val' => 'v000' },
+ # tree structure
+ { 'eval' => 'length("v251") == 2', 'key' => 'root:v251', 'val' => 'v900' },
+ { 'eval' => '"v251"', 'key' => 'code:v900', 'val' => 'v561^4:v251' },
+ ],
);
-sub data2xml {
+$log->info("rows: $maxmfn");
- use xmlify;
+$webpac->open_import_xml(type => 'isis_hidra_ths');
- my $type = shift @_;
- my $row = shift @_;
- my $add_xml = shift @_;
+while (my $rec = $webpac->fetch_rec) {
- my $xml;
+ my @ds = $webpac->data_structure($rec);
- use parse_format;
-
- my $html = ""; # html formatted display output
-
- my %field_usage; # counter for usage of each field
-
- # sort subrouting using order="" attribute
- sub by_order {
- return 0 if (! $config->{indexer}->{$a}->{order});
- return 0 if (! $config->{indexer}->{$b}->{order});
-
- return $config->{indexer}->{$a}->{order} <=>
- $config->{indexer}->{$b}->{order} ;
+ if (0 && $log->is_debug) {
+ $log->debug("rec = ",Dumper($rec));
+ $log->debug("ds = ",Dumper(\@ds));
}
- foreach my $field (sort by_order keys %{$config->{indexer}}) {
+ next if (! @ds);
- $field=x($field);
+ my $filename = $webpac->{'current_filename'};
- $field_usage{$field}++;
-
- my $swish_data = "";
- my $display_data = "";
- my $line_delimiter;
-
- my ($swish,$display);
+ if ($filename) {
+ $webpac->output_file(
+ file => $filename,
+ template => 'html.tt',
+ data => \@ds,
+ headline => $webpac->{'headline'},
+ );
+ } else {
+ print $webpac->output(
+ template => 'text.tt',
+ data => \@ds,
+ headline => $webpac->{'headline'},
+ );
+ }
- my $tag = $type2tag{$type} || die "can't find which tag to use for type $type";
- foreach my $x (@{$config->{indexer}->{$field}->{$tag}}) {
+ my $headline = $webpac->{'headline'};
- my $format = x($x->{content});
- my $delimiter = x($x->{delimiter}) || ' ';
+ my $f = $filename;
+ $f =~ s!out/!!;
- my $repeat_off = 0; # repeatable offset
+ # save into index
+ foreach my $ds (@ds) {
+ next if (! $ds->{'swish'});
- my ($s,$d,$i) = (1,1,0); # swish, display default
- $s = 0 if (lc($x->{type}) eq "display");
- $d = 0 if (lc($x->{type}) eq "swish");
- ($s,$d,$i) = (0,0,1) if (lc($x->{type}) eq "index");
+ $index->insert(
+ index_name => $ds->{'tag'},
+ path => $f,
+ headline => $headline,
+ words => join(" ",@{$ds->{'swish'}})
+ );
+ }
- # what will separate last line from this one?
- if ($display_data && $x->{append} && $x->{append} eq "1") {
- $line_delimiter = ' ';
- } elsif ($display_data) {
- $line_delimiter = '
';
- }
+ # save into sorted index (thesaurus)
+ foreach my $ds (@ds) {
+ next if (! $ds->{'index'});
- # init vars so that we go into while...
- ($swish,$display) = (1,1);
+ $thes->{$ds->{'tag'}} ||= new WebPAC::Index;
- if ($swish || $display) {
- ($swish,$display) = parse_format($type, $format,$row,$repeat_off++,$import2cp);
- # filter="name" ; filter this field through
- # filter/[name].pm
- my $filter = $x->{filter};
- if ($filter) {
- require "filter/".$filter.".pm";
- }
- # type="swish" ; field for swish
- if ($s && $swish) {
- if ($filter) {
- no strict 'refs';
- $swish_data .= join(" ",&$filter($swish));
- } else {
- $swish_data .= $swish;
- }
- }
-
- # type="display" ; field for display
- if ($d && $display) {
- if ($line_delimiter && $display_data) {
- $display_data .= $line_delimiter;
- undef $line_delimiter;
- }
- if ($filter) {
- no strict 'refs';
- $display_data .= join($delimiter,&$filter($display));
- } else {
- if ($display_data) {
- $display_data .= $delimiter.$display;
- } else {
- $display_data .= $display;
- }
- }
- }
-
- # type="index" ; insert into index
- if ($i && $display) {
- my $index_data = $display;
- if ($filter) {
- no strict 'refs';
- foreach my $d (&$filter($index_data)) {
- $index->insert($field, $d, $path);
- }
- } else {
- $index->insert($field, $index_data, $path);
- }
- }
- }
+ foreach my $h (@{$ds->{'index'}}) {
+ $thes->{$ds->{'tag'}}->insert(
+ path => $f,
+ headline => $h,
+ );
}
+ }
+# print Dumper(\@ds);
- if ($display_data) {
-
- if ($field eq "headline") {
- $xml .= xmlify("headline", $display_data);
- } else {
-
- # find field name (signular, plural)
- my $field_name = "";
- if ($config->{indexer}->{$field}->{name_singular} && $field_usage{$field} == 1) {
- $field_name = $config->{indexer}->{$field}->{name_singular}."#-#";
- } elsif ($config->{indexer}->{$field}->{name_plural}) {
- $field_name = $config->{indexer}->{$field}->{name_plural}."#-#";
- } elsif ($config->{indexer}->{$field}->{name}) {
- $field_name = $config->{indexer}->{$field}->{name}."#-#";
- } else {
- print STDERR "WARNING: field '$field' doesn't have 'name' attribute!";
- }
- if ($field_name) {
- $html .= x($field_name);
- }
- $html .= $display_data."###\n";
- }
- }
- if ($swish_data) {
- # remove extra spaces
- $swish_data =~ s/ +/ /g;
- $swish_data =~ s/ +$//g;
-
- $xml .= xmlify($field."_swish", unac_string($codepage,$swish_data));
- }
+}
+foreach my $t (keys %{$thes}) {
- }
+ my @e = $thes->{$t}->elements;
+ if (! @e) {
+ $log->logwarn("no elements in sorted index $t?");
+ next;
+ }
+
+ my $file = "./out/bfilter/$t.txt";
+ $log->info("saving sorted index $t to '$file' [".scalar(@e)." elements]");
+
+ $webpac->output_file(
+ file => $file,
+ template => 'index.tt',
+ data => \@e,
+ index_name => $t,
+ );
+}
- # dump formatted output in
- if ($html) {
- $xml .= xmlify("html",$html);
- }
-
- if ($xml) {
- $xml .= $add_xml if ($add_xml);
- return "