1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # MARCXMLPlugout.pm -- the plugout module for MARC xml recored
|
---|
4 | # A component of the Greenstone digital library software
|
---|
5 | # from the New Zealand Digital Library Project at the
|
---|
6 | # University of Waikato, New Zealand.
|
---|
7 | #
|
---|
8 | # Copyright (C) 2006 New Zealand Digital Library Project
|
---|
9 | #
|
---|
10 | # This program is free software; you can redistribute it and/or modify
|
---|
11 | # it under the terms of the GNU General Public License as published by
|
---|
12 | # the Free Software Foundation; either version 2 of the License, or
|
---|
13 | # (at your option) any later version.
|
---|
14 | #
|
---|
15 | # This program is distributed in the hope that it will be useful,
|
---|
16 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
17 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
18 | # GNU General Public License for more details.
|
---|
19 | #
|
---|
20 | # You should have received a copy of the GNU General Public License
|
---|
21 | # along with this program; if not, write to the Free Software
|
---|
22 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
23 | #
|
---|
24 | ###########################################################################
|
---|
25 |
|
---|
26 | package MARCXMLPlugout;
|
---|
27 |
|
---|
28 | use strict;
|
---|
29 | no strict 'refs';
|
---|
30 | no strict 'subs'; # allow barewords (eg STDERR) as function arguments
|
---|
31 |
|
---|
32 | eval {require bytes};
|
---|
33 | use util;
|
---|
34 | use FileUtils;
|
---|
35 | use BasePlugout;
|
---|
36 | use docprint; # for escape_text
|
---|
37 |
|
---|
38 |
|
---|
39 | sub BEGIN {
|
---|
40 | @MARCXMLPlugout::ISA = ('BasePlugout');
|
---|
41 | }
|
---|
42 |
|
---|
43 | my $arguments = [
|
---|
44 | { 'name' => "group",
|
---|
45 | 'desc' => "{MARCXMLPlugout.group}",
|
---|
46 | 'type' => "flag",
|
---|
47 | 'reqd' => "no",
|
---|
48 | 'hiddengli' => "no"},
|
---|
49 | { 'name' => "mapping_file",
|
---|
50 | 'desc' => "{MARCXMLPlugout.mapping_file}",
|
---|
51 | 'type' => "string",
|
---|
52 | 'deft' => "dc2marc-mapping.xml",
|
---|
53 | 'reqd' => "no",
|
---|
54 | 'hiddengli' => "no"},
|
---|
55 | { 'name' => "xslt_file",
|
---|
56 | 'desc' => "{BasPlugout.xslt_file}",
|
---|
57 | 'type' => "string",
|
---|
58 | 'reqd' => "no",
|
---|
59 | 'deft' => "dc2marc.xsl",
|
---|
60 | 'hiddengli' => "no"}
|
---|
61 |
|
---|
62 | ];
|
---|
63 |
|
---|
64 | my $options = { 'name' => "MARCXMLPlugout",
|
---|
65 | 'desc' => "{MARCXMLPlugout.desc}",
|
---|
66 | 'abstract' => "no",
|
---|
67 | 'inherits' => "yes",
|
---|
68 | 'args' => $arguments
|
---|
69 | };
|
---|
70 |
|
---|
71 | sub new {
|
---|
72 | my ($class) = shift (@_);
|
---|
73 | my ($plugoutlist, $inputargs,$hashArgOptLists) = @_;
|
---|
74 | push(@$plugoutlist, $class);
|
---|
75 |
|
---|
76 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
77 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
78 |
|
---|
79 | my $self = new BasePlugout($plugoutlist,$inputargs,$hashArgOptLists);
|
---|
80 |
|
---|
81 | $self->{'buffered_output'} ="";
|
---|
82 |
|
---|
83 | # xslt_file is checked in BasePlugout
|
---|
84 | # check the mapping file here
|
---|
85 | my $full_path_to_mapping_file = &util::locate_config_file($self->{'mapping_file'});
|
---|
86 | if (! defined $full_path_to_mapping_file) {
|
---|
87 | print STDERR "Can not find mapping file $self->{'mapping_file'}, please make sure you have supplied the correct file path or put the file into the collection's etc or greenstone's etc folder\n";
|
---|
88 | die "\n";
|
---|
89 | }
|
---|
90 | $self->{'mapping_file'} = $full_path_to_mapping_file;
|
---|
91 |
|
---|
92 | return bless $self, $class;
|
---|
93 | }
|
---|
94 |
|
---|
95 | sub begin {
|
---|
96 |
|
---|
97 | my $self= shift (@_);
|
---|
98 | if ($self->{'group'}) {
|
---|
99 | # all output goes into this file
|
---|
100 | my $output_dir = $self->get_output_dir();
|
---|
101 | &FileUtils::makeAllDirectories ($output_dir) unless -e $output_dir;
|
---|
102 |
|
---|
103 | $self->{'short_doc_file'} = "marc.xml";
|
---|
104 | }
|
---|
105 | }
|
---|
106 | # override BasePlugout process
|
---|
107 | sub process {
|
---|
108 | my $self = shift (@_);
|
---|
109 | my ($doc_obj) = @_;
|
---|
110 |
|
---|
111 | my $output_info = $self->{'output_info'};
|
---|
112 | return if (!defined $output_info);
|
---|
113 |
|
---|
114 | $self->process_metafiles_metadata ($doc_obj);
|
---|
115 |
|
---|
116 | if ($self->{'group'}){
|
---|
117 | $self->{buffered_output} .= $self->get_top_metadata_list($doc_obj)."\n";
|
---|
118 | }
|
---|
119 | else {
|
---|
120 | # find out which directory to save to
|
---|
121 | my $doc_dir = $self->get_doc_dir($doc_obj);
|
---|
122 | my $output_file = &FileUtils::filenameConcatenate ($self->get_output_dir(), $doc_dir, "marc.xml");
|
---|
123 | $self->open_xslt_pipe($output_file,$self->{'xslt_file'});
|
---|
124 |
|
---|
125 | my $outhandler = $self->{'xslt_writer'};
|
---|
126 | $self->output_xml_header($outhandler, "MARCXML", 1);
|
---|
127 | print $outhandler $self->get_top_metadata_list($doc_obj);
|
---|
128 | $self->output_xml_footer($outhandler,"MARCXML");
|
---|
129 | $self->close_xslt_pipe();
|
---|
130 | $self->{'short_doc_file'} = &FileUtils::filenameConcatenate ($doc_dir, "marc.xml");
|
---|
131 | }
|
---|
132 |
|
---|
133 |
|
---|
134 | # write out data to archiveinf-doc.db
|
---|
135 | if ($self->{'generate_databases'}) {
|
---|
136 | $self->store_output_info_reference($doc_obj);
|
---|
137 | $self->archiveinf_db($doc_obj);
|
---|
138 | }
|
---|
139 | if ($self->{'group'}){
|
---|
140 | $self->{'gs_count'}++;
|
---|
141 | $self->{'group_position'}++;
|
---|
142 | }
|
---|
143 | }
|
---|
144 |
|
---|
145 |
|
---|
146 | # returns a xml element of the form <MetadataList><Metadata name="metadata-name">metadata_value</Metadata>...</MetadataList>
|
---|
147 |
|
---|
148 | sub get_top_metadata_list {
|
---|
149 |
|
---|
150 | my $self = shift (@_);
|
---|
151 | my ($doc_obj) = @_;
|
---|
152 |
|
---|
153 | my @topmetadata =$doc_obj->get_all_metadata($doc_obj->get_top_section());
|
---|
154 | my $metadatalist ='<MetadataList>';
|
---|
155 |
|
---|
156 | foreach my $i (@topmetadata){
|
---|
157 | foreach my $j (@$i){
|
---|
158 | my %metaMap = @$j;
|
---|
159 | foreach my $key (keys %metaMap){
|
---|
160 | $metadatalist .='<Metadata name='."\"$key\"".'>'.&docprint::escape_text($metaMap{$key}).'</Metadata>'."\n";
|
---|
161 | }
|
---|
162 | }
|
---|
163 | }
|
---|
164 |
|
---|
165 | $metadatalist .='</MetadataList>';
|
---|
166 | return $metadatalist;
|
---|
167 | }
|
---|
168 |
|
---|
169 |
|
---|
170 | sub close_group_output{
|
---|
171 | my $self = shift (@_);
|
---|
172 |
|
---|
173 | return unless $self->{'group'} and $self->{buffered_output};
|
---|
174 |
|
---|
175 | my $output_file = &FileUtils::filenameConcatenate($self->get_output_dir(), $self->{'short_doc_file'});
|
---|
176 |
|
---|
177 | $self->open_xslt_pipe($output_file,$self->{'xslt_file'});
|
---|
178 |
|
---|
179 | my $outhandler = $self->{'xslt_writer'};
|
---|
180 |
|
---|
181 | $self->output_xml_header($outhandler, "MARCXML", 1);
|
---|
182 | print $outhandler $self->{buffered_output};
|
---|
183 | $self->output_xml_footer($outhandler,"MARCXML");
|
---|
184 | $self->close_xslt_pipe();
|
---|
185 | }
|
---|
186 |
|
---|
187 |
|
---|
188 | sub is_group{
|
---|
189 | my $self = shift (@_);
|
---|
190 | return $self->{'group'};
|
---|
191 | }
|
---|
192 |
|
---|
193 |
|
---|
194 | 1;
|
---|