1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # DSpacePlugout.pm -- the plugout module for DSpace archives
|
---|
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 DSpacePlugout;
|
---|
27 |
|
---|
28 | use strict;
|
---|
29 | no strict 'refs';
|
---|
30 |
|
---|
31 | eval {require bytes};
|
---|
32 | use util;
|
---|
33 | use BasPlugout;
|
---|
34 |
|
---|
35 | sub BEGIN {
|
---|
36 | @DSpacePlugout::ISA = ('BasPlugout');
|
---|
37 | }
|
---|
38 |
|
---|
39 | my $arguments = [];
|
---|
40 |
|
---|
41 | my $options = { 'name' => "DSpacePlugout",
|
---|
42 | 'desc' => "{DSpacePlugout.desc}",
|
---|
43 | 'abstract' => "no",
|
---|
44 | 'inherits' => "yes" };
|
---|
45 |
|
---|
46 | sub new {
|
---|
47 | my ($class) = shift (@_);
|
---|
48 | my ($plugoutlist, $inputargs,$hashArgOptLists) = @_;
|
---|
49 | push(@$plugoutlist, $class);
|
---|
50 |
|
---|
51 | if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
|
---|
52 | if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
|
---|
53 |
|
---|
54 | my $self = (defined $hashArgOptLists)? new BasPlugout($plugoutlist,$inputargs,$hashArgOptLists): new BasPlugout($plugoutlist,$inputargs);
|
---|
55 |
|
---|
56 |
|
---|
57 | return bless $self, $class;
|
---|
58 | }
|
---|
59 |
|
---|
60 | sub saveas {
|
---|
61 | my $self = shift (@_);
|
---|
62 | my ($doc_obj,$doc_dir) = @_;
|
---|
63 |
|
---|
64 | my $output_dir = $self->get_output_dir();
|
---|
65 | &util::mk_all_dir ($output_dir) unless -e $output_dir;
|
---|
66 |
|
---|
67 | my $working_dir = &util::filename_cat ($output_dir, $doc_dir);
|
---|
68 | &util::mk_all_dir ($working_dir, $doc_dir);
|
---|
69 |
|
---|
70 | #########################
|
---|
71 | # save the handle file
|
---|
72 | #########################
|
---|
73 | my $outhandle = $self->{'output_handle'};
|
---|
74 |
|
---|
75 | # Genereate handle file
|
---|
76 | # (Note: this section of code would benefit from being restructured)
|
---|
77 | my $doc_handle_file = &util::filename_cat ($working_dir, "handle");
|
---|
78 |
|
---|
79 | my $env_hp = $ENV{'DSPACE_HANDLE_PREFIX'};
|
---|
80 | my $handle_prefix = (defined $env_hp) ? $env_hp : "123456789";
|
---|
81 |
|
---|
82 | my $outhandler = $self->get_output_handler($doc_handle_file);
|
---|
83 |
|
---|
84 | my ($handle) = ($doc_dir =~ m/^(.*)\.dir$/);
|
---|
85 |
|
---|
86 | print $outhandler "$handle_prefix/$handle\n";
|
---|
87 |
|
---|
88 | close ($outhandler);
|
---|
89 |
|
---|
90 | #########################
|
---|
91 | # save the content file
|
---|
92 | #########################
|
---|
93 | my $doc_contents_file = &util::filename_cat ($working_dir, "contents");
|
---|
94 |
|
---|
95 | $outhandler = $self->get_output_handler($doc_contents_file);
|
---|
96 |
|
---|
97 | $self->process_assoc_files ($doc_obj, $doc_dir, $outhandler);
|
---|
98 |
|
---|
99 | close($outhandler);
|
---|
100 |
|
---|
101 | #############################
|
---|
102 | # save the dublin_core.xml file
|
---|
103 | ###############################
|
---|
104 | my $doc_dc_file = &util::filename_cat ($working_dir, "dublin_core.xml");
|
---|
105 | $self->open_xslt_pipe($doc_dc_file,$self->{'xslt_file'});
|
---|
106 |
|
---|
107 | if (defined $self->{'xslt_writer'}){
|
---|
108 | $outhandler = $self->{'xslt_writer'};
|
---|
109 | }
|
---|
110 | else{
|
---|
111 | $outhandler = $self->get_output_handler($doc_dc_file);
|
---|
112 | }
|
---|
113 |
|
---|
114 | $self->output_xml_header($outhandler, "dublin_core",1);
|
---|
115 |
|
---|
116 | my $all_text = $self->get_dc_metadata($doc_obj, $doc_obj->get_top_section());
|
---|
117 | print $outhandler $all_text;
|
---|
118 |
|
---|
119 | $self->output_xml_footer($outhandler,"dublin_core");
|
---|
120 |
|
---|
121 | if (defined $self->{'xslt_writer'}){
|
---|
122 | $self->close_xslt_pipe();
|
---|
123 | }
|
---|
124 | else{
|
---|
125 | close($outhandler);
|
---|
126 | }
|
---|
127 |
|
---|
128 | $self->{'short_doc_file'} = &util::filename_cat ($doc_dir, "dublin_core.xml");
|
---|
129 | $self->store_output_info_reference($doc_obj);
|
---|
130 | }
|
---|
131 |
|
---|
132 | sub process_assoc_files {
|
---|
133 | my $self = shift (@_);
|
---|
134 | my ($doc_obj, $doc_dir, $handle) = @_;
|
---|
135 |
|
---|
136 | my $outhandler = $self->{'output_handle'};
|
---|
137 |
|
---|
138 | my $output_dir = $self->get_output_dir();
|
---|
139 | return if (!defined $output_dir);
|
---|
140 |
|
---|
141 | my $working_dir = &util::filename_cat($output_dir, $doc_dir);
|
---|
142 |
|
---|
143 | my @assoc_files = ();
|
---|
144 | my $filename;;
|
---|
145 |
|
---|
146 | my $source_filename = $doc_obj->get_source_filename();
|
---|
147 |
|
---|
148 | my $collect_dir = $ENV{'GSDLCOLLECTDIR'};
|
---|
149 |
|
---|
150 | if (defined $collect_dir) {
|
---|
151 | my $dirsep_regexp = &util::get_os_dirsep();
|
---|
152 |
|
---|
153 | if ($collect_dir !~ /$dirsep_regexp$/) {
|
---|
154 | $collect_dir .= &util::get_dirsep(); # ensure there is a slash at the end
|
---|
155 | }
|
---|
156 |
|
---|
157 | # This test is never going to fail on Windows -- is this a problem?
|
---|
158 | if ($source_filename !~ /^$dirsep_regexp/) {
|
---|
159 | $source_filename = &util::filename_cat($collect_dir, $source_filename);
|
---|
160 | }
|
---|
161 | }
|
---|
162 |
|
---|
163 | my ($tail_filename) = ($source_filename =~ m/\/([^\/\\]*)$/);
|
---|
164 |
|
---|
165 | print $handle "$tail_filename\n";
|
---|
166 |
|
---|
167 | $filename = &util::filename_cat($working_dir, $tail_filename);
|
---|
168 | &util::hard_link ($source_filename, $filename);
|
---|
169 |
|
---|
170 | # set the assocfile path (even if we have no assoc files - need this for lucene)
|
---|
171 | $doc_obj->set_utf8_metadata_element ($doc_obj->get_top_section(),
|
---|
172 | "assocfilepath",
|
---|
173 | "$doc_dir");
|
---|
174 | foreach my $assoc_file_rec (@{$doc_obj->get_assoc_files()}) {
|
---|
175 | my ($dir, $afile) = $assoc_file_rec->[1] =~ /^(.*?)([^\/\\]+)$/;
|
---|
176 | $dir = "" unless defined $dir;
|
---|
177 |
|
---|
178 |
|
---|
179 | my $real_filename = $assoc_file_rec->[0];
|
---|
180 | # for some reasons the image associate file has / before the full path
|
---|
181 | $real_filename =~ s/^\\(.*)/$1/i;
|
---|
182 | if (-e $real_filename) {
|
---|
183 |
|
---|
184 | if ($real_filename =~ m/$source_filename$/) {
|
---|
185 | next;
|
---|
186 | }
|
---|
187 | else {
|
---|
188 | my $bundle = "bundle:ORIGINAL";
|
---|
189 |
|
---|
190 | if ($afile =~ m/^thumbnail\./) {
|
---|
191 | $bundle = "bundle:THUMBNAIL";
|
---|
192 | }
|
---|
193 |
|
---|
194 | # Store the associated file to the "contents" file
|
---|
195 | print $handle "$assoc_file_rec->[1]\t$bundle\n";
|
---|
196 | }
|
---|
197 |
|
---|
198 |
|
---|
199 | $filename = &util::filename_cat($working_dir, $afile);
|
---|
200 |
|
---|
201 |
|
---|
202 | &util::hard_link ($real_filename, $filename);
|
---|
203 |
|
---|
204 | $doc_obj->add_utf8_metadata ($doc_obj->get_top_section(),
|
---|
205 | "gsdlassocfile",
|
---|
206 | "$afile:$assoc_file_rec->[2]:$dir");
|
---|
207 | } elsif ($self->{'verbosity'} > 2) {
|
---|
208 | print $outhandler "DSpacePlugout::process couldn't copy the associated file " .
|
---|
209 | "$real_filename to $afile\n";
|
---|
210 | }
|
---|
211 | }
|
---|
212 | }
|
---|
213 |
|
---|
214 |
|
---|
215 | sub get_new_doc_dir{
|
---|
216 | my $self = shift (@_);
|
---|
217 | my($working_info,$working_dir,$OID) = @_;
|
---|
218 |
|
---|
219 | return $OID;
|
---|
220 |
|
---|
221 | }
|
---|