3 # Copyright 2011 BibLibre
5 # This file is part of Koha.
7 # Koha is free software; you can redistribute it and/or modify it
8 # under the terms of the GNU General Public License as published by
9 # the Free Software Foundation; either version 3 of the License, or
10 # (at your option) any later version.
12 # Koha is distributed in the hope that it will be useful, but
13 # WITHOUT ANY WARRANTY; without even the implied warranty of
14 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 # GNU General Public License for more details.
17 # You should have received a copy of the GNU General Public License
18 # along with Koha; if not, see <http://www.gnu.org/licenses>.
22 This script build OAI-PMH sets (to be used by opac/oai.pl) according to sets
23 and mappings defined in Koha. It reads informations from oai_sets and
24 oai_sets_mappings, and then fill table oai_sets_biblios with builded infos.
28 build_oai_sets.pl [-h] [-v] [-r] [-i] [-l LENGTH [-o OFFSET]]
29 -h Print help message;
31 -r Truncate table oai_sets_biblios before inserting new rows
32 -i Embed items informations, mandatory if you defined mappings
34 -l LENGTH Process LENGTH biblios
35 -o OFFSET If LENGTH is defined, start processing from OFFSET
42 use List::MoreUtils qw( uniq );
43 use Getopt::Std qw( getopts );
47 use C4::Charset qw( StripNonXmlChars );
60 $Getopt::Std::STANDARD_HELP_VERSION = 1;
61 my $go = getopts('vo:l:ihr', \%opts);
68 my $verbose = $opts{v};
69 my $offset = $opts{o};
70 my $length = $opts{l};
71 my $embed_items = $opts{i};
74 my $dbh = C4::Context->dbh;
76 # Get OAI sets mappings
77 my $mappings = GetOAISetsMappings;
79 # Get all biblionumbers and marcxml
80 print "Retrieving biblios... " if $verbose;
82 SELECT biblionumber, metadata
84 WHERE format='marcxml'
87 SELECT biblionumber, metadata
88 FROM deletedbiblio_metadata
89 WHERE format='marcxml'
93 $query .= "LIMIT $length";
95 $query .= " OFFSET $offset";
98 my $sth = $dbh->prepare($query);
99 $sth->execute( C4::Context->preference('marcflavour'), C4::Context->preference('marcflavour'));
100 my $results = $sth->fetchall_arrayref({});
101 print "done.\n" if $verbose;
103 # Build lists of parents sets
104 my $sets = GetOAISets;
106 foreach my $set (@$sets) {
107 my $setSpec = $set->{'spec'};
108 while($setSpec =~ /^(.+):(.+)$/) {
110 my $parent_set = GetOAISetBySpec($parent);
112 push @{ $parentsets->{$set->{'id'}} }, $parent_set->{'id'};
120 my $num_biblios = scalar @$results;
122 my $sets_biblios = {};
123 foreach my $res (@$results) {
124 my $biblionumber = $res->{'biblionumber'};
125 my $marcxml = $res->{'metadata'};
126 if($verbose and $i % 1000 == 0) {
127 my $percent = ($i * 100) / $num_biblios;
128 $percent = sprintf("%.2f", $percent);
129 say "Progression: $i/$num_biblios ($percent %)";
131 # The following lines are copied from GetMarcBiblio
132 # We don't call GetMarcBiblio to avoid a sql query to be executed each time
133 $marcxml = StripNonXmlChars($marcxml);
134 MARC::File::XML->default_record_format(C4::Context->preference('marcflavour'));
137 $record = MARC::Record::new_from_xml($marcxml, "UTF-8", C4::Context->preference('marcflavour'));
140 warn "(biblio $biblionumber) Error while creating record from marcxml: $@";
144 C4::Biblio::EmbedItemsInMarcBiblio({
145 marc_record => $record,
146 biblionumber => $biblionumber });
149 my @biblio_sets = CalcOAISetsBiblio($record, $mappings);
150 foreach my $set_id (@biblio_sets) {
151 push @{ $sets_biblios->{$set_id} }, $biblionumber;
152 foreach my $parent_set_id ( @{ $parentsets->{$set_id} } ) {
153 push @{ $sets_biblios->{$parent_set_id} }, $biblionumber;
158 say "Progression: done." if $verbose;
161 foreach my $set_id (keys %$sets_biblios) {
162 $sets_biblios->{$set_id} = [ uniq @{ $sets_biblios->{$set_id} } ];
163 my $set = GetOAISet($set_id);
164 my $setSpec = $set->{'spec'};
165 say "Set '$setSpec': ". scalar(@{$sets_biblios->{$set_id}}) ." biblios";
168 print "Updating database... ";
170 ModOAISetsBiblios( {} );
172 AddOAISetsBiblios($sets_biblios);
176 print "build_oai_sets.pl: Build OAI-PMH sets, according to mappings defined in Koha\n";
177 print "Usage: build_oai_sets.pl [-h] [-v] [-i] [-l LENGTH [-o OFFSET]]\n\n";
178 print "\t-h\t\tPrint this help and exit\n";
179 print "\t-v\t\tBe verbose\n";
180 print "\t-i\t\tEmbed items informations, mandatory if you defined mappings on item fields\n";
181 print "\t-l LENGTH\tProcess LENGTH biblios\n";
182 print "\t-o OFFSET\tIf LENGTH is defined, start processing from OFFSET\n\n";