1 | #!/usr/bin/perl -w
|
---|
2 |
|
---|
3 | ###########################################################################
|
---|
4 | #
|
---|
5 | # import.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) 1999 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 import a number of files into a particular collection
|
---|
30 |
|
---|
31 | package import;
|
---|
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/plugins");
|
---|
39 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugouts");
|
---|
40 |
|
---|
41 | if (defined $ENV{'GSDLEXTS'}) {
|
---|
42 | my @extensions = split(/:/,$ENV{'GSDLEXTS'});
|
---|
43 | foreach my $e (@extensions) {
|
---|
44 | my $ext_prefix = "$ENV{'GSDLHOME'}/ext/$e";
|
---|
45 |
|
---|
46 | unshift (@INC, "$ext_prefix/perllib");
|
---|
47 | unshift (@INC, "$ext_prefix/perllib/cpan");
|
---|
48 | unshift (@INC, "$ext_prefix/perllib/plugins");
|
---|
49 | unshift (@INC, "$ext_prefix/perllib/plugouts");
|
---|
50 | }
|
---|
51 | }
|
---|
52 | if (defined $ENV{'GSDL3EXTS'}) {
|
---|
53 | my @extensions = split(/:/,$ENV{'GSDL3EXTS'});
|
---|
54 | foreach my $e (@extensions) {
|
---|
55 | my $ext_prefix = "$ENV{'GSDL3SRCHOME'}/ext/$e";
|
---|
56 |
|
---|
57 | unshift (@INC, "$ext_prefix/perllib");
|
---|
58 | unshift (@INC, "$ext_prefix/perllib/cpan");
|
---|
59 | unshift (@INC, "$ext_prefix/perllib/plugins");
|
---|
60 | unshift (@INC, "$ext_prefix/perllib/plugouts");
|
---|
61 | }
|
---|
62 | }
|
---|
63 | }
|
---|
64 |
|
---|
65 | use strict;
|
---|
66 | use inexport;
|
---|
67 |
|
---|
68 | my $oidtype_list =
|
---|
69 | [ { 'name' => "hash",
|
---|
70 | 'desc' => "{import.OIDtype.hash}" },
|
---|
71 | { 'name' => "assigned",
|
---|
72 | 'desc' => "{import.OIDtype.assigned}" },
|
---|
73 | { 'name' => "incremental",
|
---|
74 | 'desc' => "{import.OIDtype.incremental}" },
|
---|
75 | { 'name' => "dirname",
|
---|
76 | 'desc' => "{import.OIDtype.dirname}" } ];
|
---|
77 |
|
---|
78 |
|
---|
79 | # used to control output file format
|
---|
80 | my $saveas_list =
|
---|
81 | [ { 'name' => "GreenstoneXML",
|
---|
82 | 'desc' => "{export.saveas.GreenstoneXML}"},
|
---|
83 | { 'name' => "GreenstoneMETS",
|
---|
84 | 'desc' => "{export.saveas.GreenstoneMETS}"},
|
---|
85 | ];
|
---|
86 |
|
---|
87 |
|
---|
88 | # Possible attributes for each argument
|
---|
89 | # name: The name of the argument
|
---|
90 | # desc: A description (or more likely a reference to a description) for this argument
|
---|
91 | # type: The type of control used to represent the argument. Options include: string, int, flag, regexp, metadata, language, enum etc
|
---|
92 | # reqd: Is this argument required?
|
---|
93 | # hiddengli: Is this argument hidden in GLI?
|
---|
94 | # modegli: The lowest detail mode this argument is visible at in GLI
|
---|
95 |
|
---|
96 | my $saveas_argument
|
---|
97 | = { 'name' => "saveas",
|
---|
98 | 'desc' => "{import.saveas}",
|
---|
99 | 'type' => "enum",
|
---|
100 | 'list' => $saveas_list,
|
---|
101 | 'deft' => "GreenstoneXML",
|
---|
102 | 'reqd' => "no",
|
---|
103 | 'modegli' => "3" };
|
---|
104 |
|
---|
105 |
|
---|
106 | my $arguments =
|
---|
107 | [
|
---|
108 | $saveas_argument,
|
---|
109 | { 'name' => "archivedir",
|
---|
110 | 'desc' => "{import.archivedir}",
|
---|
111 | 'type' => "string",
|
---|
112 | 'reqd' => "no",
|
---|
113 | 'hiddengli' => "yes" },
|
---|
114 | { 'name' => "importdir",
|
---|
115 | 'desc' => "{import.importdir}",
|
---|
116 | 'type' => "string",
|
---|
117 | 'reqd' => "no",
|
---|
118 | 'hiddengli' => "yes" },
|
---|
119 | { 'name' => "collectdir",
|
---|
120 | 'desc' => "{import.collectdir}",
|
---|
121 | 'type' => "string",
|
---|
122 | # parsearg left "" as default
|
---|
123 | #'deft' => &util::filename_cat ($ENV{'GSDLHOME'}, "collect"),
|
---|
124 | 'deft' => "",
|
---|
125 | 'reqd' => "no",
|
---|
126 | 'hiddengli' => "yes" },
|
---|
127 | { 'name' => "site",
|
---|
128 | 'desc' => "{import.site}",
|
---|
129 | 'type' => "string",
|
---|
130 | 'deft' => "",
|
---|
131 | 'reqd' => "no",
|
---|
132 | 'hiddengli' => "yes" },
|
---|
133 | { 'name' => "manifest",
|
---|
134 | 'desc' => "{import.manifest}",
|
---|
135 | 'type' => "string",
|
---|
136 | 'deft' => "",
|
---|
137 | 'reqd' => "no",
|
---|
138 | 'hiddengli' => "yes" },
|
---|
139 | { 'name' => "debug",
|
---|
140 | 'desc' => "{import.debug}",
|
---|
141 | 'type' => "flag",
|
---|
142 | 'reqd' => "no",
|
---|
143 | 'hiddengli' => "yes" },
|
---|
144 | { 'name' => "faillog",
|
---|
145 | 'desc' => "{import.faillog}",
|
---|
146 | 'type' => "string",
|
---|
147 | # parsearg left "" as default
|
---|
148 | #'deft' => &util::filename_cat("<collectdir>", "colname", "etc", "fail.log"),
|
---|
149 | 'deft' => "",
|
---|
150 | 'reqd' => "no",
|
---|
151 | 'modegli' => "3" },
|
---|
152 | { 'name' => "incremental",
|
---|
153 | 'desc' => "{import.incremental}",
|
---|
154 | 'type' => "flag",
|
---|
155 | 'hiddengli' => "yes" },
|
---|
156 | { 'name' => "keepold",
|
---|
157 | 'desc' => "{import.keepold}",
|
---|
158 | 'type' => "flag",
|
---|
159 | 'reqd' => "no",
|
---|
160 | 'hiddengli' => "yes" },
|
---|
161 | { 'name' => "removeold",
|
---|
162 | 'desc' => "{import.removeold}",
|
---|
163 | 'type' => "flag",
|
---|
164 | 'reqd' => "no",
|
---|
165 | 'hiddengli' => "yes" },
|
---|
166 | { 'name' => "language",
|
---|
167 | 'desc' => "{scripts.language}",
|
---|
168 | 'type' => "string",
|
---|
169 | 'reqd' => "no",
|
---|
170 | 'hiddengli' => "yes" },
|
---|
171 | { 'name' => "maxdocs",
|
---|
172 | 'desc' => "{import.maxdocs}",
|
---|
173 | 'type' => "int",
|
---|
174 | 'reqd' => "no",
|
---|
175 | # parsearg left "" as default
|
---|
176 | #'deft' => "-1",
|
---|
177 | 'range' => "1,",
|
---|
178 | 'modegli' => "1" },
|
---|
179 | # don't set the default to hash - want to allow this to come from
|
---|
180 | # entry in collect.cfg but want to override it here
|
---|
181 | { 'name' => "OIDtype",
|
---|
182 | 'desc' => "{import.OIDtype}",
|
---|
183 | 'type' => "enum",
|
---|
184 | 'list' => $oidtype_list,
|
---|
185 | # parsearg left "" as default
|
---|
186 | #'deft' => "hash",
|
---|
187 | 'reqd' => "no",
|
---|
188 | 'modegli' => "2" },
|
---|
189 | { 'name' => "OIDmetadata",
|
---|
190 | 'desc' => "{import.OIDmetadata}",
|
---|
191 | 'type' => "string",
|
---|
192 | #'type' => "metadata", #doesn't work properly in GLI
|
---|
193 | # parsearg left "" as default
|
---|
194 | #'deft' => "dc.Identifier",
|
---|
195 | 'reqd' => "no",
|
---|
196 | 'modegli' => "2" },
|
---|
197 | { 'name' => "out",
|
---|
198 | 'desc' => "{import.out}",
|
---|
199 | 'type' => "string",
|
---|
200 | 'deft' => "STDERR",
|
---|
201 | 'reqd' => "no",
|
---|
202 | 'hiddengli' => "yes" },
|
---|
203 | { 'name' => "sortmeta",
|
---|
204 | 'desc' => "{import.sortmeta}",
|
---|
205 | 'type' => "string",
|
---|
206 | #'type' => "metadata", #doesn't work properly in GLI
|
---|
207 | 'reqd' => "no",
|
---|
208 | 'modegli' => "2" },
|
---|
209 | { 'name' => "removeprefix",
|
---|
210 | 'desc' => "{BasClas.removeprefix}",
|
---|
211 | 'type' => "regexp",
|
---|
212 | 'deft' => "",
|
---|
213 | 'reqd' => "no",
|
---|
214 | 'modegli' => "3" },
|
---|
215 | { 'name' => "removesuffix",
|
---|
216 | 'desc' => "{BasClas.removesuffix}",
|
---|
217 | 'type' => "regexp",
|
---|
218 | 'deft' => "",
|
---|
219 | 'reqd' => "no",
|
---|
220 | 'modegli' => "3" },
|
---|
221 | { 'name' => "groupsize",
|
---|
222 | 'desc' => "{import.groupsize}",
|
---|
223 | 'type' => "int",
|
---|
224 | 'deft' => "1",
|
---|
225 | 'reqd' => "no",
|
---|
226 | 'modegli' => "2" },
|
---|
227 | { 'name' => "gzip",
|
---|
228 | 'desc' => "{import.gzip}",
|
---|
229 | 'type' => "flag",
|
---|
230 | 'reqd' => "no",
|
---|
231 | 'modegli' => "3" },
|
---|
232 | { 'name' => "statsfile",
|
---|
233 | 'desc' => "{import.statsfile}",
|
---|
234 | 'type' => "string",
|
---|
235 | 'deft' => "STDERR",
|
---|
236 | 'reqd' => "no",
|
---|
237 | 'hiddengli' => "yes" },
|
---|
238 | { 'name' => "verbosity",
|
---|
239 | 'desc' => "{import.verbosity}",
|
---|
240 | 'type' => "int",
|
---|
241 | 'range' => "0,",
|
---|
242 | # parsearg left "" as default
|
---|
243 | # 'deft' => "2",
|
---|
244 | 'reqd' => "no",
|
---|
245 | 'modegli' => "3" },
|
---|
246 | { 'name' => "gli",
|
---|
247 | 'desc' => "{scripts.gli}",
|
---|
248 | 'type' => "flag",
|
---|
249 | 'reqd' => "no",
|
---|
250 | 'hiddengli' => "yes" },
|
---|
251 | { 'name' => "xml",
|
---|
252 | 'desc' => "{scripts.xml}",
|
---|
253 | 'type' => "flag",
|
---|
254 | 'reqd' => "no",
|
---|
255 | 'hiddengli' => "yes" }];
|
---|
256 |
|
---|
257 | my $options = { 'name' => "import.pl",
|
---|
258 | 'desc' => "{import.desc}",
|
---|
259 | 'args' => $arguments };
|
---|
260 |
|
---|
261 |
|
---|
262 |
|
---|
263 | sub main
|
---|
264 | {
|
---|
265 | my $inexport = new inexport("import",\@ARGV,$options);
|
---|
266 |
|
---|
267 | my $collection = $inexport->get_collection();
|
---|
268 |
|
---|
269 | if (defined $collection) {
|
---|
270 | my ($config_filename,$collect_cfg)
|
---|
271 | = $inexport->read_collection_cfg($collection,$options);
|
---|
272 |
|
---|
273 | $inexport->set_collection_options($collect_cfg);
|
---|
274 |
|
---|
275 | my $pluginfo = $inexport->process_files($config_filename,$collect_cfg);
|
---|
276 |
|
---|
277 | $inexport->generate_statistics($pluginfo);
|
---|
278 | }
|
---|
279 | }
|
---|
280 |
|
---|
281 |
|
---|
282 | &main();
|
---|