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, "$ENV{'GSDLHOME'}/perllib/plugins");
|
---|
38 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/classify");
|
---|
39 | }
|
---|
40 |
|
---|
41 | use arcinfo;
|
---|
42 | use colcfg;
|
---|
43 | use plugin;
|
---|
44 | use docprint;
|
---|
45 | use util;
|
---|
46 | use parsargv;
|
---|
47 | use FileHandle;
|
---|
48 |
|
---|
49 | sub print_usage {
|
---|
50 | print STDERR "\n usage: $0 [options] collection-name\n\n";
|
---|
51 | print STDERR " options:\n";
|
---|
52 | print STDERR " -verbosity number 0=none, 3=lots\n";
|
---|
53 | print STDERR " -importdir directory Where the original material lives\n";
|
---|
54 | print STDERR " -archivedir directory Where the converted material ends up\n";
|
---|
55 | print STDERR " -keepold Will not destroy the current contents of the\n";
|
---|
56 | print STDERR " archives directory (the default)\n";
|
---|
57 | print STDERR " -removeold Will remove the old contents of the archives\n";
|
---|
58 | print STDERR " directory -- use with care\n";
|
---|
59 | print STDERR " -gzip Use gzip to compress resulting gml documents\n";
|
---|
60 | print STDERR " (don't forget to include ZIPPlug in your plugin\n";
|
---|
61 | print STDERR " list when building from compressed documents)\n";
|
---|
62 | print STDERR " -maxdocs number Maximum number of documents to import\n";
|
---|
63 | print STDERR " -groupsize number Number of GML documents to group into one file\n";
|
---|
64 | print STDERR " -sortmeta metadata Sort documents alphabetically by metadata for\n";
|
---|
65 | print STDERR " building. This will be disabled if groupsize > 1\n";
|
---|
66 | print STDERR " -debug Print imported text to STDOUT\n";
|
---|
67 | print STDERR " -out Filename or handle to print output status to.\n";
|
---|
68 | print STDERR " The default is STDERR\n\n";
|
---|
69 | }
|
---|
70 |
|
---|
71 | &main();
|
---|
72 |
|
---|
73 | sub main {
|
---|
74 | my ($verbosity, $importdir, $archivedir, $keepold,
|
---|
75 | $removeold, $gzip, $groupsize, $debug, $maxdocs, $collection,
|
---|
76 | $configfilename, $collectcfg, $pluginfo, $sortmeta,
|
---|
77 | $archive_info_filename, $archive_info, $processor, $out);
|
---|
78 | if (!parsargv::parse(\@ARGV,
|
---|
79 | 'verbosity/\d+/2', \$verbosity,
|
---|
80 | 'importdir/.*/', \$importdir,
|
---|
81 | 'archivedir/.*/', \$archivedir,
|
---|
82 | 'keepold', \$keepold,
|
---|
83 | 'removeold', \$removeold,
|
---|
84 | 'gzip', \$gzip,
|
---|
85 | 'groupsize/\d+/1', \$groupsize,
|
---|
86 | 'sortmeta/.*/', \$sortmeta,
|
---|
87 | 'debug', \$debug,
|
---|
88 | 'maxdocs/^\-?\d+/-1', \$maxdocs,
|
---|
89 | 'out/.*/STDERR', \$out)) {
|
---|
90 | &print_usage();
|
---|
91 | die "\n";
|
---|
92 | }
|
---|
93 |
|
---|
94 | my $close_out = 0;
|
---|
95 | if ($out !~ /^(STDERR|STDOUT)$/i) {
|
---|
96 | open (OUT, ">$out") || die "Couldn't open output file $out\n";
|
---|
97 | $out = 'import::OUT';
|
---|
98 | $close_out = 1;
|
---|
99 | }
|
---|
100 | $out->autoflush(1);
|
---|
101 |
|
---|
102 | # set removeold to false if it has been defined
|
---|
103 | $removeold = 0 if ($keepold);
|
---|
104 |
|
---|
105 | # get and check the collection name
|
---|
106 | if (($collection = &util::use_collection(@ARGV)) eq "") {
|
---|
107 | &print_usage();
|
---|
108 | die "\n";
|
---|
109 | }
|
---|
110 |
|
---|
111 | # check sortmeta
|
---|
112 | $sortmeta = undef unless defined $sortmeta && $sortmeta =~ /\S/;
|
---|
113 | if (defined $sortmeta && $groupsize > 1) {
|
---|
114 | print $out "WARNING: import.pl cannot sort documents when groupsize > 1\n";
|
---|
115 | print $out " sortmeta option will be ignored\n\n";
|
---|
116 | $sortmeta = undef;
|
---|
117 | }
|
---|
118 |
|
---|
119 | # dynamically load 'docsave' module so it can pick up on a collection
|
---|
120 | # specific docsave.pm is specified.
|
---|
121 |
|
---|
122 | unshift (@INC, "$ENV{'GSDLCOLLECTDIR'}/perllib");
|
---|
123 | require docsave;
|
---|
124 |
|
---|
125 |
|
---|
126 | # get the list of plugins for this collection
|
---|
127 | my $plugins = [];
|
---|
128 | $configfilename = &util::filename_cat ($ENV{'GSDLCOLLECTDIR'}, "etc/collect.cfg");
|
---|
129 | if (-e $configfilename) {
|
---|
130 | $collectcfg = &colcfg::read_collect_cfg ($configfilename);
|
---|
131 | if (defined $collectcfg->{'plugin'}) {
|
---|
132 | $plugins = $collectcfg->{'plugin'};
|
---|
133 | }
|
---|
134 | if (defined $collectcfg->{'importdir'} && $importdir eq "") {
|
---|
135 | $importdir = $collectcfg->{'importdir'};
|
---|
136 | }
|
---|
137 | if (defined $collectcfg->{'archivedir'} && $archivedir eq "") {
|
---|
138 | $archivedir = $collectcfg->{'archivedir'};
|
---|
139 | }
|
---|
140 | if (defined $collectcfg->{'removeold'}) {
|
---|
141 | if ($collectcfg->{'removeold'} =~ /^true$/i && !$keepold) {
|
---|
142 | $removeold = 1;
|
---|
143 | }
|
---|
144 | if ($collectcfg->{'removeold'} =~ /^false$/i && !$removeold) {
|
---|
145 | $removeold = 0;
|
---|
146 | }
|
---|
147 | }
|
---|
148 | } else {
|
---|
149 | die "Couldn't find the configuration file $configfilename\n";
|
---|
150 | }
|
---|
151 |
|
---|
152 | # fill in the default import and archives directories if none
|
---|
153 | # were supplied, turn all \ into / and remove trailing /
|
---|
154 | $importdir = "$ENV{'GSDLCOLLECTDIR'}/import" if $importdir eq "";
|
---|
155 | $importdir =~ s/[\\\/]+/\//g;
|
---|
156 | $importdir =~ s/\/$//;
|
---|
157 | $archivedir = "$ENV{'GSDLCOLLECTDIR'}/archives" if $archivedir eq "";
|
---|
158 | $archivedir =~ s/[\\\/]+/\//g;
|
---|
159 | $archivedir =~ s/\/$//;
|
---|
160 |
|
---|
161 | # load all the plugins
|
---|
162 | $pluginfo = &plugin::load_plugins ($plugins, $verbosity, $out);
|
---|
163 | if (scalar(@$pluginfo) == 0) {
|
---|
164 | print $out "No plugins were loaded.\n";
|
---|
165 | die "\n";
|
---|
166 | }
|
---|
167 |
|
---|
168 | # remove the old contents of the archives directory if needed
|
---|
169 | if ($removeold && -e $archivedir) {
|
---|
170 | print $out "Warning - removing current contents of the archives directory\n";
|
---|
171 | print $out " in preparation for the import\n";
|
---|
172 | sleep(5); # just in case...
|
---|
173 | &util::rm_r ($archivedir);
|
---|
174 | }
|
---|
175 |
|
---|
176 | # read the archive information file
|
---|
177 | if (!$debug) {
|
---|
178 | $archive_info_filename = &util::filename_cat ($archivedir, "archives.inf");
|
---|
179 | $archive_info = new arcinfo ();
|
---|
180 | $archive_info->load_info ($archive_info_filename);
|
---|
181 |
|
---|
182 | # create a docsave object to process the documents
|
---|
183 | $processor = new docsave ($collection, $archive_info, $verbosity, $gzip, $groupsize, $out);
|
---|
184 | $processor->setarchivedir ($archivedir);
|
---|
185 | $processor->set_sortmeta ($sortmeta) if defined $sortmeta;
|
---|
186 | } else {
|
---|
187 | $processor = new docprint ();
|
---|
188 | }
|
---|
189 |
|
---|
190 | &plugin::begin($pluginfo, $importdir, $processor, $maxdocs);
|
---|
191 |
|
---|
192 | # process the import directory
|
---|
193 | &plugin::read ($pluginfo, $importdir, "", {}, $processor, $maxdocs);
|
---|
194 |
|
---|
195 | &plugin::end($pluginfo);
|
---|
196 |
|
---|
197 | # write out the archive information file
|
---|
198 | if (!$debug) {
|
---|
199 | $processor->close_file_output() if $groupsize > 1;
|
---|
200 | $archive_info->save_info($archive_info_filename);
|
---|
201 | }
|
---|
202 | close OUT if $close_out;
|
---|
203 | }
|
---|