--- trunk/all2xml.pl 2002/11/24 20:52:11 1
+++ trunk2/all2all.pl 2004/11/06 21:00:14 587
@@ -1,66 +1,391 @@
#!/usr/bin/perl -w
-use strict;
-use OpenIsis;
-use Getopt::Std;
-use Data::Dumper;
-use XML::Simple;
+=head1 NAME
+
+all2all.pl - basic script for all WebPAC needs
-my $config=XMLin();
+=cut
-print Dumper($config);
+use strict;
+use locale;
+use YAML;
+use Carp;
+use Getopt::Long;
+use Text::Unaccent 1.02;
+
+use lib './lib';
+use WebPAC;
+use WebPAC::jsFind;
+use WebPAC::Index;
+use WebPAC::Tree;
+
+# options which can be changed via command line
+#
+my $code_page = 'ISO-8859-2';
+my ($limit_mfn, $start_mfn, $debug, $low_mem);
+my $index_path = './out/index';
+
+my $result = GetOptions(
+ "code_page=s" => \$code_page,
+ "limit_mfn=i" => \$limit_mfn,
+ "start_mfn=i" => \$start_mfn,
+ "debug!" => \$debug,
+ "low_mem!" => \$low_mem,
+);
+
+my $filter = {
+ 'CROVOC' => sub {
+ my $tmp = shift || return;
+ $tmp =~ s/\s*HIDRA.*//;
+ return undef unless ($tmp =~ s/\s*CROVOC.*$/ #C#/);
+ # remove repeating stars
+ # FIXME this should be fixed at right place, not here!
+ $tmp =~ s/(\s*#C#)+/ #C#/g;
+ return $tmp;
+ },
+ 'CROVOC_tree' => sub {
+ my $tmp = shift || return;
+ $tmp =~ s/\s*CROVOC.*$/ /;
+ $tmp =~ s/\s*EUROVOC.*//;
+ $tmp =~ s/\s*HIDRA.*//;
+ return $tmp;
+ },
+ # TT filter
+ 'CROVOC_img' => sub {
+ my $tmp = shift;
+ $tmp =~ s/\s*#C#\s*/ /gis;
+ $tmp =~ s/"img\/crovoc.png"/"..\/img\/crovoc.png"/gis;
+ return $tmp;
+ }
+};
+
+## remove accented characters
+#
+sub unac {
+ my $string = shift || return;
+ $string = unac_string($code_page,$string);
+ $string =~ tr/ðÐ/dD/;
+ $string = unac_entities($string);
+ return $string;
+}
+sub unac_2 {
+ my $string = shift || return;
+ if (length($string) > 2) {
+ my $pr = substr($string,0,2);
+ $string = unac_string($code_page,substr($string,2));
+ $string =~ tr/ðÐ/dD/;
+ $string = $pr . $string;
+ }
+ $string = unac_entities($string);
+ return lc($string);
+}
-my %opts;
+sub unac_entities {
+ my $ent = shift || return;
-getopts('d:m:q', \%opts);
+ $ent =~ s/&(\w)(acute|cedil|circ|grave|ring|slash|tilde|uml);/$1/gi;
+ $ent =~ s/ð/d/g;
+ $ent =~ s/&E[tT][hH];/D/g;
+ $ent =~ s/&(\w\w)lig;/$1/gi;
-my $db_dir = $opts{d};
+ return $ent;
+}
-die "usage: $0 -d [database_dir] -m [database1,database2] " if (! %opts);
+# create WebPAC object
+#
+my $webpac = new WebPAC(
+ code_page => $code_page,
+ limit_mfn => $limit_mfn,
+ start_mfn => $start_mfn,
+ debug => $debug,
+ low_mem => $low_mem,
+ filter => $filter,
+) || die;
+
+my $log = $webpac->_get_logger() || die "can't get logger";
+
+$log->debug("creating WebPAC::jsFind object");
+
+my $index = new WebPAC::jsFind(
+ index_path => $index_path,
+ keys => 62,
+) || die;
+
+my $thes;
+
+$|=1;
+
+my $maxmfn = $webpac->open_isis(
+ filename => shift @ARGV || '/data/hidra/THS/THS',
+ lookup => [
+ { 'key' => 'd:v900', 'val' => 'filter{CROVOC_tree}v250^a v800' },
+# { 'eval' => '"v901^a" eq "Podruèje"', 'key' => 'pa:v561^4:v562^4:v461^1', 'val' => 'v900' },
+# { 'eval '=> '"v901^a" eq "Mikrotezaurus"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },
+# { 'eval' => '"v901^a" eq "Deskriptor"', 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },
+ { 'key' => 'a:v561^4:v562^4:v461^1', 'val' => 'v900' },
+ { 'key' => '900_mfn:v900', 'val' => 'v000' },
+ # tree structure
+ { 'eval' => 'length("v251") == 2 && "v800" =~ m/EUROVOC/ || "v800" =~ m/CROVOC/ && "v251" =~ m/^(H|HD|L|Z|P)$/', 'key' => 'root:v251', 'val' => 'v900' },
+ { 'eval' => '"v251"', 'key' => 'code:v900', 'val' => 'v561^4:v251' },
+ { 'eval' => '"v561^4" && "v562^4"', 'key' => 'code:v900', 'val' => 'v561^4:v562^4' },
+ ],
+);
+
+$log->debug("isis file ",$webpac->{'isis_filename'}," opened");
+
+$log->info("rows: $maxmfn");
+
+$webpac->open_import_xml(type => 'isis_hidra_ths');
+
+if(1) { # XXX
+
+while (my $rec = $webpac->fetch_rec) {
+
+ my @ds = $webpac->data_structure($rec);
+
+ if (0 && $log->is_debug) {
+ $log->debug("rec = ",Dump($rec));
+ $log->debug("ds = ",Dump(\@ds));
+ }
-#--------------------------------------------------------------------
+ next if (! @ds);
-my $last_tell=0;
+ my $filename = $webpac->{'current_filename'} || $log->logdie("no current_filename in webpac object");
-my @isis_dirs = ( '.' ); # use dirname as database name
+ if ($filename) {
+ $webpac->output_file(
+ file => $filename,
+ template => 'html.tt',
+ data => \@ds,
+ headline => $webpac->{'headline'},
+ );
+ } else {
+ print $webpac->output(
+ template => 'text.tt',
+ data => \@ds,
+ headline => $webpac->{'headline'},
+ );
+ }
-if ($opts{m}) {
- @isis_dirs = split(/,/,$opts{m});
-}
+ my $headline = $webpac->{'headline'};
-my @isis_dbs;
+ my $f = $filename;
+ $f =~ s!out/!!;
-foreach (@isis_dirs) {
- if (-e "$common::isis_data/$db_dir/$_/LIBRI") {
- push @isis_dbs,"$common::isis_data/$db_dir/$_/LIBRI/LIBRI";
+ # save into index
+ foreach my $ds (@ds) {
+ next if (! $ds->{'swish'});
+
+ # strip all non word characters from beginning or end
+ # of word
+ my $words = unac(join(" ",@{$ds->{'swish'}}));
+ $words =~ s/^\W+//;
+ $words =~ s/\W*\s+\W*/ /g;
+ $words =~ s/\W+$//;
+
+ # first try to generate headline for this entry from index
+ my $h = $ds->{'index'}->[0];
+ # then, from display
+ $h ||= $ds->{'display'}->[0];
+ # and as last resport, fallback to headline
+ $h ||= $headline;
+
+ $index->insert(
+ index_name => $ds->{'tag'},
+ #path => $f,
+ path => $webpac->mfn,
+ headline => $h,
+ words => $words,
+ );
}
- if (-e "$common::isis_data/$db_dir/$_/PERI") {
- push @isis_dbs,"$common::isis_data/$db_dir/$_/PERI/PERI";
+
+ # save into sorted index (thesaurus)
+ foreach my $ds (@ds) {
+ next if (! $ds->{'index'});
+
+ $thes->{$ds->{'tag'}} ||= new WebPAC::Index( name => $ds->{'tag'} );
+
+ foreach my $h (@{$ds->{'index'}}) {
+ $thes->{$ds->{'tag'}}->insert(
+ sort_by => unac_2($h),
+ mfn => $webpac->mfn,
+ headline => $h,
+ );
+ }
}
- if (-e "$common::isis_data/$db_dir/$_/AMS") {
- push @isis_dbs,"$common::isis_data/$db_dir/$_/AMS/AMS";
+
+# print Dump(\@ds);
+
+}
+
+foreach my $t (keys %{$thes}) {
+
+ my @e = $thes->{$t}->elements;
+ if (! @e) {
+ $log->logwarn("no elements in sorted index $t?");
+ next;
}
- if (-e "$common::isis_data/$db_dir/$_/ARTI") {
-# push @isis_dbs,"$common::isis_data/$db_dir/$_/ARTI/ARTI";
+
+ 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,
+ );
+}
+
+if (0 && $log->is_debug) {
+ $log->debug("lookup hash: ",Dump($webpac->{'lookup'}));
+ $log->debug("data hash: ",Dump($webpac->{'data'}));
+ foreach my $t (keys %{$thes}) {
+ $log->debug("thesaurus $t hash: ",Dump($thes->{$t}));
}
}
-foreach my $isis_db (@isis_dbs) {
+} # XXX if(0)
- my $db = OpenIsis::open( "$isis_db" ) || warn "can't open '$isis_db'";
+#$log->debug("lookup hash: ",Dump($webpac->{'lookup'}));
- my $max_rowid = OpenIsis::maxRowid( $db );
+$log->info("creating tree");
- my $last_pcnt = 0;
+#
+# define tree structure
+#
+
+my $l = $webpac->{'lookup'} || $log->logconfess("can't find lookup");
+
+my @tree = ({
+ # level 0
+ code_arr => sub { sort keys %{$l} },
+ filter_code => sub {
+ my $t = shift;
+ return $t if ($t =~ s/root://);
+ },
+ lookup_v900 => sub { shift @{$l->{"root:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[1]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[1]}} },
+ have_children => sub { return $l->{"a:".$_[0]."::"} },
+ have_children_at_level => sub {
+ return unless (defined($l->{"code:".$_[1]}));
+ my $code = shift @{$l->{"code:".$_[1]}};
+ print STDERR "## $_[1] -> $code\n";
+ return undef unless($code);
+ return(9, $l->{"a:$code:"} ) if (defined($l->{"a:$code:"}));
+ },
+ style => 'display: none',
+ iframe => 1,
+ },{
+ # 1
+ code_arr => sub { @{$_[0]} },
+ filter_code => sub { shift }, # nop
+ lookup_v900 => sub { shift @{$l->{"code:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[0]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[0]}} },
+ have_children => sub { return $l->{"a:".$_[1].":"} },
+ style => 'display: none',
+ },{
+ # 2
+ code_arr => sub { @{$_[0]} },
+ filter_code => sub { shift },
+ lookup_v900 => sub { shift @{$l->{"code:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[0]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[0]}} },
+ have_children => sub { return $l->{"a:".$_[1].":".$_[0]} },
+ style => 'display: none',
+ },{
+ # 3 u¾i pojam
+ code_arr => sub { @{$_[0]} },
+ filter_code => sub { shift },
+ lookup_v900 => sub { shift @{$l->{"code:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[0]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[0]}} },
+ have_children => sub { return $l->{"a:".$_[1].":".$_[0]} },
+ },{
+ # 4
+ code_arr => sub { @{$_[0]} },
+ filter_code => sub { shift },
+ lookup_v900 => sub { shift @{$l->{"code:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[0]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[0]}} },
+ have_children => sub { return $l->{"a:".$_[1].":".$_[0]} },
+ },{
+ # 5
+ code_arr => sub { @{$_[0]} },
+ filter_code => sub { shift },
+ lookup_v900 => sub { shift @{$l->{"code:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[0]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[0]}} },
+ have_children => sub { return $l->{"a:".$_[1].":".$_[0]} },
+ },{
+ # 6
+ code_arr => sub { @{$_[0]} },
+ filter_code => sub { shift },
+ lookup_v900 => sub { shift @{$l->{"code:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[0]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[0]}} },
+ have_children => sub { return $l->{"a:".$_[1].":".$_[0]} },
+ },{
+ # 7
+ code_arr => sub { @{$_[0]} },
+ filter_code => sub { shift },
+ lookup_v900 => sub { shift @{$l->{"code:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[0]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[0]}} },
+ have_children => sub { return $l->{"a:".$_[1].":".$_[0]} },
+ },{
+ # 8
+ code_arr => sub { @{$_[0]} },
+ filter_code => sub { shift },
+ lookup_v900 => sub { shift @{$l->{"code:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[0]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[0]}} },
+# have_children => sub { return $l->{"a:".$_[1].":".$_[0]} },
+ have_children => sub { 0 },
+ },{
+ # 9 - level which is never reached except explicitly
+ code_arr => sub { @{$_[0]} },
+ filter_code => sub { shift },
+ lookup_v900 => sub { shift @{$l->{"code:".$_[0]}} },
+ lookup_term => sub { shift @{$l->{"d:".$_[0]}} },
+ lookup_mfn => sub { shift @{$l->{"900_mfn:".$_[0]}} },
+ have_children => sub { 0 },
+ have_children_at_level => sub { defined($l->{"a:".$_[1].":".$_[0]}) && return (9,$l->{"a:".$_[1].":".$_[0]}) },
+ },{
+});
+
+my $tree = new WebPAC::Tree(
+ tree => \@tree,
+ detail_url => sub {
+ my $mfn = shift;
+ my $url = "thes/${mfn}.html";
+ return $url if (-e "./out/$url");
+ },
+ nodes_dir => 'nodes',
+);
+
+$tree->output(
+ dir => './out',
+ html => 'browse.html',
+ template_dir => './output_template/',
+ template_tree => 'tree.tt',
+ template_node => 'node.tt',
+ js => 'tree-ids.js',
+);
+
+$tree->output(
+ dir => './eurovoc',
+ html => 'hijerarhija.html',
+ template_dir => './output_template/',
+ template_tree => 'hijerarhija.tt',
+ template_node => 'hijerarhija_node.tt',
+ js => 'tree-ids.js',
+);
- for (my $row_id = 1; $row_id <= $max_rowid; $row_id++ ) {
- my $row = OpenIsis::read( $db, $row_id );
- # output current process indicator
- my $pcnt = int($row->{mfn} * 100 / $max_rowid);
- if ($pcnt != $last_pcnt) {
- printf STDERR ("%5d / %5d -- %-2d %%\n",$row->{mfn},$max_rowid,$pcnt) if (! $opts{q});
- $last_pcnt = $pcnt;
- }
- }
-}
+$log->info("closing index");
+$index->close;
+
+$log->info("elapsed time: ",$webpac->fmt_time(time() - $webpac->{'start_t'}));