--- trunk/all2xml.pl 2002/12/01 22:51:29 4 +++ trunk2/all2all.pl 2004/09/11 08:36:38 422 @@ -1,209 +1,233 @@ #!/usr/bin/perl -w +=head1 NAME + +all2all.pl - basic script for all WebPAC needs + +=cut + use strict; -use OpenIsis; -use Getopt::Std; use Data::Dumper; -use XML::Simple; -use Text::Unaccent; -require Unicode::Map8; +use Carp; -my $config=XMLin(undef, forcearray => [ 'isis' ], forcecontent => 1); +use lib './lib'; +use WebPAC; +use WebPAC::jsFind; +use WebPAC::Index; -my %opts; +my $webpac = new WebPAC( + code_page => 'ISO-8859-2', +# limit_mfn => 100, +# debug => 1, + low_mem => 1, +) || die; -getopts('d:m:q', \%opts); +my $log = $webpac->_get_logger() || die "can't get logger"; -my $db_dir = $opts{d} || "ps"; # FIX +$log->debug("creating WebPAC::jsFind object"); -#die "usage: $0 -d [database_dir] -m [database1,database2] " if (! %opts); +my $index = new WebPAC::jsFind( + index_path => './out/index', + keys => 10, +) || die; -#print Dumper($config->{indexer}); -#print "-" x 70,"\n"; +my $thes; -# how to convert isis code page to UTF8? -my $isis_map = Unicode::Map8->new($config->{isis_codepage}) || die; +$|=1; -sub isis2xml { +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' }, + ], +); - my $row = shift @_; +$log->debug("isis file ",$webpac->{'isis_filename'}," opened"); - my $xml; +$log->info("rows: $maxmfn"); - 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 - my $all_sf = $row->{$isis_id}->[0]; - $all_sf =~ s/\^./ /g; nuke definirions - return $all_sf; - } elsif ($sf->{$subfield}) { - return $sf->{$subfield}; - } - } - } +$webpac->open_import_xml(type => 'isis_hidra_ths'); - foreach my $field (keys %{$config->{indexer}}) { +$log->debug("data: ",Dumper($webpac->{'data'})); - my $display_data = ""; - my $index_data = ""; +while (my $rec = $webpac->fetch_rec) { - foreach my $x (@{$config->{indexer}->{$field}->{isis}}) { + my @ds = $webpac->data_structure($rec); - 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); + if (0 && $log->is_debug) { + $log->debug("rec = ",Dumper($rec)); + $log->debug("ds = ",Dumper(\@ds)); + } -# $display_data .= $display_tmp if ($display_tmp ne ""); -# $index_data .= $index_tmp if ($index_tmp ne ""); - $display_data .= $display_tmp; - $index_data .= $index_tmp; + next if (! @ds); - } -#print "--display:$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 ); + my $filename = $webpac->{'current_filename'} || $log->logdie("no current_filename in webpac object"); + + if ($filename) { + $webpac->output_file( + file => $filename, + template => 'html.tt', + data => \@ds, + headline => $webpac->{'headline'}, + ); } else { - return; + print $webpac->output( + template => 'text.tt', + data => \@ds, + headline => $webpac->{'headline'}, + ); } -} -########################################################################## + my $headline = $webpac->{'headline'}; -my $last_tell=0; + my $f = $filename; + $f =~ s!out/!!; -my @isis_dirs = ( '.' ); # use dirname as database name + # save into index + foreach my $ds (@ds) { + next if (! $ds->{'swish'}); -if ($opts{m}) { - @isis_dirs = split(/,/,$opts{m}); -} + $index->insert( + index_name => $ds->{'tag'}, + path => $f, + headline => $headline, + words => join(" ",@{$ds->{'swish'}}) + ); + } -my @isis_dbs; + # save into sorted index (thesaurus) + foreach my $ds (@ds) { + next if (! $ds->{'index'}); -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"; + $thes->{$ds->{'tag'}} ||= new WebPAC::Index; + + foreach my $h (@{$ds->{'index'}}) { + $thes->{$ds->{'tag'}}->insert( + path => $f, + headline => $h, + ); + } } -} -print STDERR "FATAL: Can't find isis database.\nPerhaps isis_data (".$config->{isis_data}.") has wrong value?\n" if (! @isis_dbs); +# print Dumper(\@ds); -my $db; +} -foreach my $isis_db (@isis_dbs) { +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, + ); +} - my $db = OpenIsis::open( $isis_db ); - if (0) { -# # FIX -# if (! $db ) { - print STDERR "WARNING: can't open '$isis_db'\n"; - next ; +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})); } +} - my $max_rowid = OpenIsis::maxRowid( $db ); +$log->info("lookup hash: ",Dumper($webpac->{'lookup'})); - print STDERR "Reading database: $isis_db [$max_rowid rows]\n"; +# +# tree dump implementation which shouldn't be here :-) +# - my $last_p = 0; +my $file = 'out/browse.html'; +open(HTML, "> $file") || $log->logdie("can't open '$file'"); -# { 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 && $row->{mfn}) { +print HTML qq{ + + + Browse + + + + + + +}; + +$index->close; + +$log->info("elapsed time: ",$webpac->fmt_time(time() - $webpac->{'start_t'}));