1 | #!/usr/bin/perl -w
|
---|
2 |
|
---|
3 | ###########################################################################
|
---|
4 | #
|
---|
5 | # export.pl --
|
---|
6 | # A component of the Greenstone digital library software
|
---|
7 | # from the New Zealand Digital Library Project at the
|
---|
8 | # University of Waikato, New Zealand.
|
---|
9 | #
|
---|
10 | # Copyright (C) 2004 New Zealand Digital Library Project
|
---|
11 | #
|
---|
12 | # This program is free software; you can redistribute it and/or modify
|
---|
13 | # it under the terms of the GNU General Public License as published by
|
---|
14 | # the Free Software Foundation; either version 2 of the License, or
|
---|
15 | # (at your option) any later version.
|
---|
16 | #
|
---|
17 | # This program is distributed in the hope that it will be useful,
|
---|
18 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
19 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
20 | # GNU General Public License for more details.
|
---|
21 | #
|
---|
22 | # You should have received a copy of the GNU General Public License
|
---|
23 | # along with this program; if not, write to the Free Software
|
---|
24 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
25 | #
|
---|
26 | ###########################################################################
|
---|
27 |
|
---|
28 |
|
---|
29 | # This program will export a particular collection into a specific Format (e.g. METS or DSpace) by importing then saving as a different format.
|
---|
30 |
|
---|
31 | package export;
|
---|
32 |
|
---|
33 | BEGIN {
|
---|
34 | die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
|
---|
35 | die "GSDLOS not set\n" unless defined $ENV{'GSDLOS'};
|
---|
36 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib");
|
---|
37 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/cpan");
|
---|
38 | # unshift (@INC, "$ENV{'GSDLHOME'}/perllib/cpan/perl-5.8");
|
---|
39 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugins");
|
---|
40 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugouts");
|
---|
41 |
|
---|
42 | if (defined $ENV{'GSDLEXTS'}) {
|
---|
43 | my @extensions = split(/:/,$ENV{'GSDLEXTS'});
|
---|
44 | foreach my $e (@extensions) {
|
---|
45 | my $ext_prefix = "$ENV{'GSDLHOME'}/ext/$e";
|
---|
46 |
|
---|
47 | unshift (@INC, "$ext_prefix/perllib");
|
---|
48 | unshift (@INC, "$ext_prefix/perllib/cpan");
|
---|
49 | unshift (@INC, "$ext_prefix/perllib/plugins");
|
---|
50 | unshift (@INC, "$ext_prefix/perllib/plugouts");
|
---|
51 | }
|
---|
52 | }
|
---|
53 | if (defined $ENV{'GSDL3EXTS'}) {
|
---|
54 | my @extensions = split(/:/,$ENV{'GSDL3EXTS'});
|
---|
55 | foreach my $e (@extensions) {
|
---|
56 | my $ext_prefix = "$ENV{'GSDL3SRCHOME'}/ext/$e";
|
---|
57 |
|
---|
58 | unshift (@INC, "$ext_prefix/perllib");
|
---|
59 | unshift (@INC, "$ext_prefix/perllib/cpan");
|
---|
60 | unshift (@INC, "$ext_prefix/perllib/plugins");
|
---|
61 | unshift (@INC, "$ext_prefix/perllib/plugouts");
|
---|
62 | }
|
---|
63 | }
|
---|
64 |
|
---|
65 | }
|
---|
66 |
|
---|
67 | use strict;
|
---|
68 | #no strict 'refs'; # allow filehandles to be variables and vice versa
|
---|
69 | #no strict 'subs'; # allow barewords (eg STDERR) as function arguments
|
---|
70 | use inexport;
|
---|
71 |
|
---|
72 | my $oidtype_list =
|
---|
73 | [ { 'name' => "hash",
|
---|
74 | 'desc' => "{import.OIDtype.hash}" },
|
---|
75 | { 'name' => "assigned",
|
---|
76 | 'desc' => "{import.OIDtype.assigned}" },
|
---|
77 | { 'name' => "incremental",
|
---|
78 | 'desc' => "{import.OIDtype.incremental}" },
|
---|
79 | { 'name' => "dirname",
|
---|
80 | 'desc' => "{import.OIDtype.dirname}" } ];
|
---|
81 |
|
---|
82 | # what format to export as
|
---|
83 | my $saveas_list =
|
---|
84 | [ { 'name' => "GreenstoneMETS",
|
---|
85 | 'desc' => "{export.saveas.GreenstoneMETS}"},
|
---|
86 | { 'name' => "FedoraMETS",
|
---|
87 | 'desc' => "{export.saveas.FedoraMETS}"},
|
---|
88 | { 'name' => "MARCXML",
|
---|
89 | 'desc' => "{export.saveas.MARCXML}"},
|
---|
90 | { 'name' => "DSpace",
|
---|
91 | 'desc' => "{export.saveas.DSpace}" }
|
---|
92 | ];
|
---|
93 |
|
---|
94 |
|
---|
95 | # Possible attributes for each argument
|
---|
96 | # name: The name of the argument
|
---|
97 | # desc: A description (or more likely a reference to a description) for this argument
|
---|
98 | # type: The type of control used to represent the argument. Options include: string, int, flag, regexp, metadata, language, enum etc
|
---|
99 | # reqd: Is this argument required?
|
---|
100 | # hiddengli: Is this argument hidden in GLI?
|
---|
101 | # modegli: The lowest detail mode this argument is visible at in GLI
|
---|
102 |
|
---|
103 | my $saveas_argument =
|
---|
104 | { 'name' => "saveas",
|
---|
105 | 'desc' => "{export.saveas}",
|
---|
106 | 'type' => "enum",
|
---|
107 | 'list' => $saveas_list,
|
---|
108 | 'deft' => "GreenstoneMETS",
|
---|
109 | 'reqd' => "no",
|
---|
110 | 'modegli' => "3" };
|
---|
111 |
|
---|
112 |
|
---|
113 | my $arguments =
|
---|
114 | [
|
---|
115 | $saveas_argument,
|
---|
116 | { 'name' => "exportdir",
|
---|
117 | 'desc' => "{export.archivedir}",
|
---|
118 | 'type' => "string",
|
---|
119 | 'reqd' => "no",
|
---|
120 | 'hiddengli' => "yes" },
|
---|
121 | { 'name' => "importdir",
|
---|
122 | 'desc' => "{import.importdir}",
|
---|
123 | 'type' => "string",
|
---|
124 | 'reqd' => "no",
|
---|
125 | 'hiddengli' => "yes" },
|
---|
126 | { 'name' => "collectdir",
|
---|
127 | 'desc' => "{export.collectdir}",
|
---|
128 | 'type' => "string",
|
---|
129 | # parsearg left "" as default
|
---|
130 | #'deft' => &util::filename_cat ($ENV{'GSDLHOME'}, "collect"),
|
---|
131 | 'deft' => "",
|
---|
132 | 'reqd' => "no",
|
---|
133 | 'hiddengli' => "yes" },
|
---|
134 | { 'name' => "site",
|
---|
135 | 'desc' => "{import.site}",
|
---|
136 | 'type' => "string",
|
---|
137 | 'deft' => "",
|
---|
138 | 'reqd' => "no",
|
---|
139 | 'hiddengli' => "yes" },
|
---|
140 | { 'name' => "manifest",
|
---|
141 | 'desc' => "{import.manifest}",
|
---|
142 | 'type' => "string",
|
---|
143 | 'deft' => "",
|
---|
144 | 'reqd' => "no",
|
---|
145 | 'hiddengli' => "yes" },
|
---|
146 | { 'name' => "debug",
|
---|
147 | 'desc' => "{export.debug}",
|
---|
148 | 'type' => "flag",
|
---|
149 | 'reqd' => "no",
|
---|
150 | 'hiddengli' => "yes" },
|
---|
151 | { 'name' => "faillog",
|
---|
152 | 'desc' => "{export.faillog}",
|
---|
153 | 'type' => "string",
|
---|
154 | 'deft' => "",
|
---|
155 | 'reqd' => "no",
|
---|
156 | 'modegli' => "3" },
|
---|
157 | # does this make sense?
|
---|
158 | { 'name' => "incremental",
|
---|
159 | 'desc' => "{import.incremental}",
|
---|
160 | 'type' => "flag",
|
---|
161 | 'hiddengli' => "yes" },
|
---|
162 | { 'name' => "keepold",
|
---|
163 | 'desc' => "{export.keepold}",
|
---|
164 | 'type' => "flag",
|
---|
165 | 'reqd' => "no",
|
---|
166 | 'hiddengli' => "yes" },
|
---|
167 | { 'name' => "removeold",
|
---|
168 | 'desc' => "{export.removeold}",
|
---|
169 | 'type' => "flag",
|
---|
170 | 'reqd' => "no",
|
---|
171 | 'hiddengli' => "yes" },
|
---|
172 | { 'name' => "language",
|
---|
173 | 'desc' => "{scripts.language}",
|
---|
174 | 'type' => "string",
|
---|
175 | 'reqd' => "no",
|
---|
176 | 'hiddengli' => "yes" },
|
---|
177 | { 'name' => "maxdocs",
|
---|
178 | 'desc' => "{export.maxdocs}",
|
---|
179 | 'type' => "int",
|
---|
180 | 'reqd' => "no",
|
---|
181 | 'range' => "1,",
|
---|
182 | 'modegli' => "1" },
|
---|
183 | { 'name' => "OIDtype",
|
---|
184 | 'desc' => "{import.OIDtype}",
|
---|
185 | 'type' => "enum",
|
---|
186 | 'list' => $oidtype_list,
|
---|
187 | # parsearg left "" as default
|
---|
188 | #'deft' => "hash",
|
---|
189 | 'reqd' => "no",
|
---|
190 | 'modegli' => "2" },
|
---|
191 | { 'name' => "OIDmetadata",
|
---|
192 | 'desc' => "{import.OIDmetadata}",
|
---|
193 | 'type' => "string",
|
---|
194 | #'type' => "metadata", #doesn't work properly in GLI
|
---|
195 | 'deft' => "dc.Identifier",
|
---|
196 | 'reqd' => "no",
|
---|
197 | 'modegli' => "2" },
|
---|
198 | { 'name' => "out",
|
---|
199 | 'desc' => "{export.out}",
|
---|
200 | 'type' => "string",
|
---|
201 | 'deft' => "STDERR",
|
---|
202 | 'reqd' => "no",
|
---|
203 | 'hiddengli' => "yes" },
|
---|
204 | { 'name' => "statsfile",
|
---|
205 | 'desc' => "{export.statsfile}",
|
---|
206 | 'type' => "string",
|
---|
207 | 'deft' => "STDERR",
|
---|
208 | 'reqd' => "no",
|
---|
209 | 'hiddengli' => "yes" },
|
---|
210 | { 'name' => "xsltfile",
|
---|
211 | 'desc' => "{BasPlugout.xslt_file}",
|
---|
212 | 'type' => "string",
|
---|
213 | 'reqd' => "no",
|
---|
214 | 'hiddengli' => "yes" },
|
---|
215 | { 'name' => "xslt_txt",
|
---|
216 | 'desc' => "{METSPlugout.xslt_txt}",
|
---|
217 | 'type' => "string",
|
---|
218 | 'reqd' => "no",
|
---|
219 | 'hiddengli' => "no" },
|
---|
220 | { 'name' => "xslt_mets",
|
---|
221 | 'desc' => "{METSPlugout.xslt_mets}",
|
---|
222 | 'type' => "string",
|
---|
223 | 'reqd' => "no",
|
---|
224 | 'hiddengli' => "no" },
|
---|
225 | { 'name' => "fedora_namespace",
|
---|
226 | 'desc' => "{FedoraMETSPlugout.fedora_namespace} (-saveas FedoraMETS)",
|
---|
227 | 'type' => "string",
|
---|
228 | 'deft' => "greenstone",
|
---|
229 | 'reqd' => "no",
|
---|
230 | 'hiddengli' => "no" },
|
---|
231 | { 'name' => "mapping_file",
|
---|
232 | 'desc' => "{MARCXMLPlugout.mapping_file} (-saveas MARCXML)",
|
---|
233 | 'type' => "string",
|
---|
234 | 'reqd' => "no",
|
---|
235 | 'hiddengli' => "no" },
|
---|
236 | { 'name' => "group_marc",
|
---|
237 | 'desc' => "{MARCXMLPlugout.group} (-saveas MARCXML)",
|
---|
238 | 'type' => "flag",
|
---|
239 | 'reqd' => "no",
|
---|
240 | 'hiddengli' => "no" },
|
---|
241 | { 'name' => "verbosity",
|
---|
242 | 'desc' => "{export.verbosity}",
|
---|
243 | 'type' => "int",
|
---|
244 | 'range' => "0,3",
|
---|
245 | 'deft' => "2",
|
---|
246 | 'reqd' => "no",
|
---|
247 | 'modegli' => "3" },
|
---|
248 | { 'name' => "gli",
|
---|
249 | 'desc' => "{scripts.gli}",
|
---|
250 | 'type' => "flag",
|
---|
251 | 'reqd' => "no",
|
---|
252 | 'hiddengli' => "yes" },
|
---|
253 | { 'name' => "listall",
|
---|
254 | 'desc' => "{export.listall}",
|
---|
255 | 'type' => "flag",
|
---|
256 | 'reqd' => "no" },
|
---|
257 | { 'name' => "xml",
|
---|
258 | 'desc' => "{scripts.xml}",
|
---|
259 | 'type' => "flag",
|
---|
260 | 'reqd' => "no",
|
---|
261 | 'hiddengli' => "yes" }
|
---|
262 | ];
|
---|
263 |
|
---|
264 | my $options = { 'name' => "export.pl",
|
---|
265 | 'desc' => "{export.desc}",
|
---|
266 | 'args' => $arguments };
|
---|
267 |
|
---|
268 | my $listall_options = { 'name' => "export.pl",
|
---|
269 | 'desc' => "{export.desc}",
|
---|
270 | 'args' => [ $saveas_argument ] };
|
---|
271 |
|
---|
272 |
|
---|
273 |
|
---|
274 | sub main
|
---|
275 | {
|
---|
276 | my $inexport = new inexport("export",\@ARGV,$options,$listall_options);
|
---|
277 |
|
---|
278 | my $collection = $inexport->get_collection();
|
---|
279 |
|
---|
280 | if (defined $collection) {
|
---|
281 | my ($config_filename,$collect_cfg)
|
---|
282 | = $inexport->read_collection_cfg($collection,$options);
|
---|
283 |
|
---|
284 | $inexport->set_collection_options($collect_cfg);
|
---|
285 |
|
---|
286 | my $pluginfo = $inexport->process_files($config_filename,$collect_cfg);
|
---|
287 |
|
---|
288 | $inexport->generate_statistics($pluginfo);
|
---|
289 | }
|
---|
290 | }
|
---|
291 |
|
---|
292 |
|
---|
293 | &main();
|
---|
294 |
|
---|
295 |
|
---|