--- trunk/all2xml.pl 2002/11/30 00:36:34 3 +++ trunk2/all2all.pl 2004/09/05 22:22:37 411 @@ -1,207 +1,139 @@ #!/usr/bin/perl -w -use strict; -use OpenIsis; -use Getopt::Std; -use Data::Dumper; -use XML::Simple; -use Text::Unaccent; -require Unicode::Map8; - -my $config=XMLin(undef, forcearray => [ 'isis' ], forcecontent => 1); - -my %opts; - -getopts('d:m:q', \%opts); - -my $db_dir = $opts{d} || "ps"; # FIX - -#die "usage: $0 -d [database_dir] -m [database1,database2] " if (! %opts); - -#print Dumper($config->{indexer}); -#print "-" x 70,"\n"; - -# how to convert isis code page to UTF8? -my $isis_map = Unicode::Map8->new($config->{isis_codepage}) || die; - -sub isis2xml { - - my $row = shift @_; - - my $xml; - - sub isis_sf { - my $row = shift @_; - my $isis_id = shift @_; - my $subfield = shift @_; - if ($row->{$isis_id}->[0]) { - my $sf = OpenIsis::subfields($row->{$isis_id}->[0]); - if (! defined $subfield || length($subfield) == 0) { - # subfield list undef, empty or no defined subfields for this record - return $row->{$isis_id}->[0]; - } elsif ($sf->{$subfield}) { - return $sf->{$subfield}; - } - } - } - - foreach my $field (keys %{$config->{indexer}}) { - - my $display_data = ""; - my $index_data = ""; - - foreach my $x (@{$config->{indexer}->{$field}->{isis}}) { - - my $display_tmp = ""; - my $index_tmp = ""; - - my $format = $x->{content}; - my $i = 1; # index only - my $d = 1; # display only - $i = 0 if (lc($x->{type}) eq "display"); - $d = 0 if (lc($x->{type}) eq "index"); -#print "## i: $i d: $d ## $format ##"; - # parse format - my $prefix = ""; - if ($format =~ s/^([^\d]+)//) { - $prefix = $1; - } - while ($format) { - if ($format =~ s/^(\d\d\d)(\w?)//) { - my $isis_tmp = isis_sf($row,$1,$2); - if ($isis_tmp) { -# $display_tmp .= $prefix . "/$1/$2/".$isis_tmp if ($d); - $display_tmp .= $prefix . $isis_tmp if ($d); - $index_tmp .= $isis_tmp." " if ($i); -#print " $isis_tmp <--\n"; - } - $prefix = ""; - } elsif ($format =~ s/^([^\d]+)//) { - $prefix = $1; - } else { - print STDERR "WARNING: unparsed format '$format'\n"; - last; - }; - } - # add suffix - $display_tmp .= $prefix if ($display_tmp); - -# $display_data .= $display_tmp if ($display_tmp ne ""); -# $index_data .= $index_tmp if ($index_tmp ne ""); - $display_data .= $display_tmp; - $index_data .= $index_tmp; - - } -#print "--display:$display_data\n--index:$index_data\n"; - $xml->{$field}->{display} .= $isis_map->tou($display_data)->utf8 if ($display_data); - $xml->{$field}->{index} .= unac_string($config->{isis_codepage},$index_data) if ($index_data); - - } - if ($xml) { - return XMLout($xml, rootname => 'xml', noattr => 1 ); - } else { - return; - } -} +=head1 NAME -########################################################################## +all2all.pl - basic script for all WebPAC needs -my $last_tell=0; +=cut -my @isis_dirs = ( '.' ); # use dirname as database name +use strict; +use Data::Dumper; +use Carp; -if ($opts{m}) { - @isis_dirs = split(/,/,$opts{m}); -} +use lib './lib'; +use WebPAC; +use WebPAC::jsFind; +use WebPAC::Index; -my @isis_dbs; +my $webpac = new WebPAC( + code_page => 'ISO-8859-2', + limit_mfn => 500, +# debug => 1, +) || die; -foreach (@isis_dirs) { - if (-e $config->{isis_data}."/$db_dir/$_/LIBRI") { - push @isis_dbs,$config->{isis_data}."/$db_dir/$_/LIBRI/LIBRI"; - } - if (-e $config->{isis_data}."/$db_dir/$_/PERI") { - push @isis_dbs,$config->{isis_data}."/$db_dir/$_/PERI/PERI"; - } - if (-e $config->{isis_data}."/$db_dir/$_/AMS") { - push @isis_dbs,$config->{isis_data}."/$db_dir/$_/AMS/AMS"; - } - if (-e $config->{isis_data}."/$db_dir/$_/ARTI") { -# push @isis_dbs,$config->{isis_data}."/$db_dir/$_/ARTI/ARTI"; - } -} +my $log = $webpac->_get_logger(); -print STDERR "FATAL: Can't find isis database.\nPerhaps isis_data (".$config->{isis_data}.") has wrong value?\n" if (! @isis_dbs); +my $index = new WebPAC::jsFind( + index_path => './out/index', + keys => 80, +) || die; -my $db; +my $thes; -foreach my $isis_db (@isis_dbs) { +$|=1; +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' }, + ], +); - my $db = OpenIsis::open( $isis_db ); - if (0) { -# # FIX -# if (! $db ) { - print STDERR "WARNING: can't open '$isis_db'\n"; - next ; - } +$log->info("rows: $maxmfn"); - my $max_rowid = OpenIsis::maxRowid( $db ); +$webpac->open_import_xml(type => 'isis_hidra_ths'); - print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; +while (my $rec = $webpac->fetch_rec) { - my $last_p = 0; + my @ds = $webpac->data_structure($rec); + + if (0 && $log->is_debug) { + $log->debug("rec = ",Dumper($rec)); + $log->debug("ds = ",Dumper(\@ds)); + } -# { my $row_id = 1; -# FIX - for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) { - my $row = OpenIsis::read( $db, $row_id ); - if ($row && $row->{mfn}) { + next if (! @ds); - # output current process indicator - my $p = int($row->{mfn} * 100 / $max_rowid); - if ($p != $last_p) { - printf STDERR ("%5d / %5d [%-51s] %-2d %% \r",$row->{mfn},$max_rowid,"=" x ($p/2).">", $p ) if (! $opts{q}); - $last_p = $p; - } + my $filename = $webpac->{'current_filename'}; - if (my $xml = isis2xml($row)) { - print "Path-Name: $isis_db#".$row->{mfn}."\n"; - print "Content-Length: ".(length($xml)+1)."\n"; - print "Document-Type: XML\n\n$xml\n"; - } - } + 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'}, + ); } - print STDERR "\n"; -} + my $headline = $webpac->{'headline'}; -1; -__END__ -########################################################################## + my $f = $filename; + $f =~ s!out/!!; -=head1 NAME + # save into index + foreach my $ds (@ds) { + next if (! $ds->{'swish'}); -isis2xml.pl - read isis file and dump XML + $index->insert( + index_name => $ds->{'tag'}, + path => $f, + headline => $headline, + words => join(" ",@{$ds->{'swish'}}) + ); + } -=head1 DESCRIPTION + # save into sorted index (thesaurus) + foreach my $ds (@ds) { + next if (! $ds->{'index'}); -This command will read ISIS data file using OpenIsis perl module and -create XML file for usage with I -indexer. Dispite it's name, this script B -from isis files (isis allready has something like that). Output of this -script is tailor-made for SWISH-E. + $thes->{$ds->{'tag'}} ||= new WebPAC::Index; -=head1 AUTHOR + $thes->{$ds->{'tag'}}->insert( + path => $f, + headline => $headline, + ); + } -Dobrica Pavlinusic +# print Dumper(\@ds); -=head1 COPYRIGHT +} -GNU Public License (GPL) v2 or later +foreach my $t (keys %{$thes}) { -=head1 SEE ALSO + my @e = $thes->{$t}->elements; + if (! @e) { + $log->logwarn("no elements in sorted index $t?"); + next; + } + + $log->debug("saving sorted index $t [".scalar(@e)." elements]"); + + $webpac->output_file( + file => "./out/thes_$t.html", + template => 'index.tt', + data => \@e, + index_name => $t, + ); +} -SWISH-E web site at http://www.swish-e.org +if (0 && $log->is_debug) { + $log->debug("lookup hash: ",Dumper($webpac->{'lookup'})); + $log->debug("data hash: ",Dumper($webpac->{'data'})); + foreach my $t (keys %{$thes}) { + $log->debug("thesaurus $t hash: ",Dumper($thes->{$t})); + } +} + +$index->close; -=cut