source: gsdl/trunk/perllib/plugins/ISISPlugin.pm@ 17033

Last change on this file since 17033 was 16392, checked in by kjdon, 16 years ago

global block pass: read_block is no more, use can_process_this_file to see whether a file is for us or not. extra arg (block_hash) to read, read_into_doc_obj, metadata_read etc
  • Property svn:keywords set to Author Date Id Revision
File size: 13.8 KB
Line 
1###########################################################################
2#
3# ISISPlugin.pm -- A plugin for CDS/ISIS databases
4#
5# A component of the Greenstone digital library software
6# from the New Zealand Digital Library Project at the
7# University of Waikato, New Zealand.
8#
9# Copyright 1999-2004 New Zealand Digital Library Project
10#
11# This program is free software; you can redistribute it and/or modify
12# it under the terms of the GNU General Public License as published by
13# the Free Software Foundation; either version 2 of the License, or
14# (at your option) any later version.
15#
16# This program is distributed in the hope that it will be useful,
17# but WITHOUT ANY WARRANTY; without even the implied warranty of
18# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19# GNU General Public License for more details.
20#
21# You should have received a copy of the GNU General Public License
22# along with this program; if not, write to the Free Software
23# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
24#
25###########################################################################
26
27package ISISPlugin;
28
29
30use multiread;
31use SplitTextFile;
32
33use strict;
34no strict 'refs'; # allow filehandles to be variables and viceversa
35
36# ISISPlugin is a sub-class of SplitTextFile.
37sub BEGIN {
38 @ISISPlugin::ISA = ('SplitTextFile');
39}
40
41
42my $arguments =
43 [ { 'name' => "process_exp",
44 'desc' => "{BasePlugin.process_exp}",
45 'type' => "regexp",
46 'reqd' => "no",
47 'deft' => &get_default_process_exp() },
48 { 'name' => "block_exp",
49 'desc' => "{BasePlugin.block_exp}",
50 'type' => "regexp",
51 'reqd' => "no",
52 'deft' => &get_default_block_exp(),
53 'hiddengli' => "yes" },
54 { 'name' => "split_exp",
55 'desc' => "{SplitTextFile.split_exp}",
56 'type' => "regexp",
57 'reqd' => "no",
58 'deft' => &get_default_split_exp(),
59 'hiddengli' => "yes" },
60
61 # The interesting options
62 { 'name' => "entry_separator",
63 'desc' => "{ISISPlugin.entry_separator}",
64 'type' => "string",
65 'reqd' => "no",
66 'deft' => "<br>" },
67 { 'name' => "subfield_separator",
68 'desc' => "{ISISPlugin.subfield_separator}",
69 'type' => "string",
70 'reqd' => "no",
71 'deft' => ", " }
72 ];
73
74my $options = { 'name' => "ISISPlugin",
75 'desc' => "{ISISPlugin.desc}",
76 'abstract' => "no",
77 'inherits' => "yes",
78 'explodes' => "yes",
79 'args' => $arguments };
80
81
82# This plugin processes files with the suffix ".mst"
83sub get_default_process_exp {
84 return q^(?i)(\.mst)$^;
85}
86
87
88# This plugin blocks files with the suffix ".fdt" and ".xrf"
89sub get_default_block_exp {
90 #return q^(?i)(\.fdt|\.xrf)$^;
91 return "";
92}
93
94
95# This plugin splits the input text at the "----------" lines
96sub get_default_split_exp {
97 return q^\r?\n----------\r?\n^;
98}
99
100
101sub new
102{
103 my ($class) = shift (@_);
104 my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
105 push(@$pluginlist, $class);
106
107 push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
108 push(@{$hashArgOptLists->{"OptList"}},$options);
109
110 my $self = new SplitTextFile($pluginlist, $inputargs, $hashArgOptLists);
111
112 if ($self->{'info_only'}) {
113 # don't worry about any options etc
114 return bless $self, $class;
115 }
116
117 # isis plug doesn't care about encoding - it assumes ascii unless the user
118 # has specified an encoding
119 if ($self->{'input_encoding'} eq "auto") {
120 $self->{'input_encoding'} = "ascii";
121 }
122 return bless $self, $class;
123}
124
125# we block the corresponding fdt and xrf
126sub store_block_files {
127
128 my $self =shift (@_);
129 my ($filename_full_path, $block_hash) = @_;
130
131 $self->check_auxiliary_files($filename_full_path);
132 if (-e $self->{'fdt_file_path'}) {
133 my $fdt_file = $self->{'fdt_file_path'};
134 $block_hash->{'file_blocks'}->{$fdt_file} = 1;
135 }
136 if (-e $self->{'xrf_file_path'}) {
137 my $xrf_file = $self->{'xrf_file_path'};
138 $block_hash->{'file_blocks'}->{$xrf_file} = 1;
139 }
140
141
142}
143
144sub check_auxiliary_files {
145 my $self = shift (@_);
146 my ($filename) = @_;
147
148 my ($database_file_path_root) = ($filename =~ /(.*)\.mst$/i);
149 # Check the associated .fdt and .xrf files exist
150 $self->{'fdt_file_path'} = $database_file_path_root . ".FDT";
151 if (!-e $self->{'fdt_file_path'}) {
152 $self->{'fdt_file_path'} = $database_file_path_root . ".fdt";
153 }
154 $self->{'xrf_file_path'} = $database_file_path_root . ".XRF";
155 if (!-e $self->{'xrf_file_path'}) {
156 $self->{'xrf_file_path'} = $database_file_path_root . ".xrf";
157 }
158}
159
160
161sub read_file
162{
163 my $self = shift (@_);
164 my ($filename, $encoding, $language, $textref) = @_;
165 my $outhandle = $self->{'outhandle'};
166
167 my ($database_file_path_root) = ($filename =~ /(.*)\.mst$/i);
168 my $mst_file_path_relative = $filename;
169 $mst_file_path_relative =~ s/^.+import.(.*?)$/$1/;
170
171 # Check the associated .fdt and .xrf files exist
172 $self->check_auxiliary_files($filename);
173
174 if (!-e $self->{'fdt_file_path'}) {
175 print STDERR "<ProcessingError n='$mst_file_path_relative' r='Could not find ISIS FDT file $self->{'fdt_file_path'}'>\n" if ($self->{'gli'});
176 print $outhandle "Error: Could not find ISIS FDT file " . $self->{'fdt_file_path'} . ".\n";
177 return;
178 }
179 if (!-e $self->{'xrf_file_path'}) {
180 print STDERR "<ProcessingError n='$mst_file_path_relative' r='Could not find ISIS XRF file $self->{'xrf_file_path'}'>\n" if ($self->{'gli'});
181 print $outhandle "Error: Could not find ISIS XRF file " . $self->{'xrf_file_path'} . ".\n";
182 return;
183 }
184
185 # The text to split is exported from the database by the IsisGdl program
186 open(FILE, "IsisGdl \"$filename\" |");
187
188 my $reader = new multiread();
189 $reader->set_handle('ISISPlugin::FILE');
190 $reader->set_encoding($encoding);
191 $reader->read_file($textref);
192
193 close(FILE);
194
195 # Parse the associated ISIS database Field Definition Table file (.fdt)
196 my %fdt_mapping = &parse_field_definition_table($self->{'fdt_file_path'}, $encoding);
197 $self->{'fdt_mapping'} = \%fdt_mapping;
198
199 # Remove the line at the start, and any blank lines, so the data is split and processed properly
200 $$textref =~ s/^----------\n//;
201 $$textref =~ s/\n\n/\n/g;
202}
203
204
205sub process
206{
207 my $self = shift (@_);
208 my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
209 my $outhandle = $self->{'outhandle'};
210
211 my $section = $doc_obj->get_top_section();
212 my $fdt_mapping = $self->{'fdt_mapping'};
213 my $subfield_separator = $self->{'subfield_separator'};
214 my $entry_separator = $self->{'entry_separator'};
215 my $isis_record_html_metadata_value = "<table cellpadding=\"4\" cellspacing=\"0\">";
216
217 # Process each line of the ISIS record, one at a time
218 foreach my $line (split(/\n/, $$textref)) {
219 $line =~ s/(\s*)$//; # Remove any nasty whitespace (very important for Windows)
220 $line =~ /^tag=(.*) data=(.+)$/;
221 my $tag = $1;
222 my $tag_data = $2;
223 # print STDERR "\nTag: $tag, Data: $tag_data\n";
224
225 # Convert the tag number into a name, and remove any invalid characters
226 my $raw_metadata_name = $fdt_mapping->{$tag}{'name'} || "";
227 $raw_metadata_name =~ s/[,&\#\.\-\/]/ /g;
228 next if ($raw_metadata_name eq "");
229
230 # Metadata field names: title case, then remove spaces
231 my $metadata_name = "";
232 foreach my $word (split(/\s+/, $raw_metadata_name)) {
233 substr($word, 0, 1) =~ tr/a-z/A-Z/;
234 $metadata_name .= $word;
235 }
236
237 my $all_metadata_name = $metadata_name . "^all";
238 my $all_metadata_value = "";
239
240 # Handle repeatable fields
241 if ($fdt_mapping->{$tag}{'repeatable'}) {
242 # Multiple values are separated using the '%' character
243 foreach my $raw_metadata_value (split(/%/, $tag_data)) {
244 my $metadata_value = "";
245
246 # Handle subfields
247 while ($raw_metadata_value ne "") {
248 # If there is a subfield specifier, parse it off
249 my $sub_metadata_name = $metadata_name;
250 if ($raw_metadata_value =~ s/^\^// && $raw_metadata_value =~ s/^([a-z])//) {
251 $sub_metadata_name .= "^$1";
252 }
253
254 # Parse the value off and add it as metadata
255 $raw_metadata_value =~ s/^([^\^]*)//;
256 my $sub_metadata_value = &escape_metadata_value($1);
257
258 # print STDERR "Sub metadata name: $sub_metadata_name, value: $sub_metadata_value\n";
259 if ($sub_metadata_name ne $metadata_name) {
260 $doc_obj->add_utf8_metadata($section, $sub_metadata_name, $sub_metadata_value);
261 }
262
263 # If this tag has subfields and this is the first, use the value for the CDS/ISIS ^* field
264 if ($fdt_mapping->{$tag}{'subfields'} ne "" && $metadata_value eq "") {
265 $doc_obj->add_utf8_metadata($section, $metadata_name . "^*", $sub_metadata_value);
266 }
267
268 $metadata_value .= $subfield_separator unless ($metadata_value eq "");
269 $metadata_value .= $sub_metadata_value;
270 }
271
272 # Add the metadata value
273 # print STDERR "Metadata name: $metadata_name, value: $metadata_value\n";
274 $doc_obj->add_utf8_metadata($section, $metadata_name, $metadata_value);
275
276 $all_metadata_value .= $entry_separator unless ($all_metadata_value eq "");
277 $all_metadata_value .= $metadata_value;
278 }
279 }
280
281 # Handle non-repeatable fields
282 else {
283 my $raw_metadata_value = $tag_data;
284 my $metadata_value = "";
285
286 # Handle subfields
287 while ($raw_metadata_value ne "") {
288 # If there is a subfield specifier, parse it off
289 my $sub_metadata_name = $metadata_name;
290 if ($raw_metadata_value =~ s/^\^// && $raw_metadata_value =~ s/^([a-z])//) {
291 $sub_metadata_name .= "^$1";
292 }
293
294 # Parse the value off and add it as metadata
295 $raw_metadata_value =~ s/^([^\^]*)//;
296 my $sub_metadata_value = $1;
297
298 # Deal with the case when multiple values are specified using <...>
299 if ($sub_metadata_value =~ /\<(.+)\>/) {
300 my $sub_sub_metadata_name = $sub_metadata_name . "^sub";
301 my $tmp_sub_metadata_value = $sub_metadata_value;
302 while ($tmp_sub_metadata_value =~ s/\<(.+?)\>//) {
303 my $sub_sub_metadata_value = $1;
304 $doc_obj->add_utf8_metadata($section, $sub_sub_metadata_name, $sub_sub_metadata_value);
305 }
306 }
307 # Deal with the legacy case when multiple values are specified using /.../
308 elsif ($sub_metadata_value =~ /\/(.+)\//) {
309 my $sub_sub_metadata_name = $sub_metadata_name . "^sub";
310 my $tmp_sub_metadata_value = $sub_metadata_value;
311 while ($tmp_sub_metadata_value =~ s/\/(.+?)\///) {
312 my $sub_sub_metadata_value = $1;
313 $doc_obj->add_utf8_metadata($section, $sub_sub_metadata_name, $sub_sub_metadata_value);
314 }
315 }
316
317 # Escape the metadata value so it appears correctly in the final collection
318 $sub_metadata_value = &escape_metadata_value($sub_metadata_value);
319
320 # print STDERR "Sub metadata name: $sub_metadata_name, value: $sub_metadata_value\n";
321 if ($sub_metadata_name ne $metadata_name) {
322 $doc_obj->add_utf8_metadata($section, $sub_metadata_name, $sub_metadata_value);
323 }
324
325 # If this tag has subfields and this is the first, use the value for the CDS/ISIS ^* field
326 if ($fdt_mapping->{$tag}{'subfields'} ne "" && $metadata_value eq "") {
327 $doc_obj->add_utf8_metadata($section, $metadata_name . "^*", $sub_metadata_value);
328 }
329
330 $metadata_value .= $subfield_separator unless ($metadata_value eq "");
331 $metadata_value .= $sub_metadata_value;
332 }
333
334 # Add the metadata value
335 # print STDERR "Metadata name: $metadata_name, value: $metadata_value\n";
336 $doc_obj->add_utf8_metadata($section, $metadata_name, $metadata_value);
337
338 $all_metadata_value .= $entry_separator unless ($all_metadata_value eq "");
339 $all_metadata_value .= $metadata_value;
340 }
341
342 # Add the "^all" metadata value
343 # print STDERR "All metadata name: $all_metadata_name, value: $all_metadata_value\n";
344 $doc_obj->add_utf8_metadata($section, $all_metadata_name, $all_metadata_value);
345
346 $isis_record_html_metadata_value .= "<tr><td valign=top><nobr><b>" . $fdt_mapping->{$tag}{'name'} . "</b></nobr></td><td valign=top>" . $all_metadata_value . "</td></tr>";
347 }
348
349 # Add a reasonably formatted HTML table view of the record as the document text
350 $isis_record_html_metadata_value .= "</table>";
351 $doc_obj->add_utf8_text($section, $isis_record_html_metadata_value);
352
353 # Add the full raw record as metadata
354 my $isis_raw_record_metadata_value = &escape_metadata_value($$textref);
355 $doc_obj->add_utf8_metadata($section, "ISISRawRecord", $isis_raw_record_metadata_value);
356
357 # Add FileFormat metadata
358 $doc_obj->add_utf8_metadata($section, "FileFormat", "CDS/ISIS");
359
360 # Record was processed successfully
361 return 1;
362}
363
364
365sub parse_field_definition_table
366{
367 my $fdtfilename = shift(@_);
368 my $encoding = shift(@_);
369
370 my %fdtmapping = ();
371
372 open(FDT_FILE, "<$fdtfilename") || die "Error: Could not open file $fdtfilename.\n";
373
374 my $fdtfiletext = "";
375 my $reader = new multiread();
376 $reader->set_handle('ISISPlugin::FDT_FILE');
377 $reader->set_encoding($encoding);
378 $reader->read_file($fdtfiletext);
379
380 my $amongstdefinitions = 0;
381 foreach my $fdtfileline (split(/\n/, $$fdtfiletext)) {
382 $fdtfileline =~ s/(\s*)$//; # Remove any nasty spaces at the end of the lines
383
384 if ($amongstdefinitions) {
385 my $fieldname = &unicode::substr($fdtfileline, 0, 30);
386 my $fieldsubfields = &unicode::substr($fdtfileline, 30, 20);
387 my $fieldspecs = &unicode::substr($fdtfileline, 50, 50);
388
389 # Remove extra spaces
390 $fieldname =~ s/(\s*)$//;
391 $fieldsubfields =~ s/(\s*)$//;
392 $fieldspecs =~ s/(\s*)$//;
393
394 # Map from tag number to metadata field title, subfields, and repeatability
395 my $fieldtag = (split(/ /, $fieldspecs))[0];
396 my $fieldrepeatable = (split(/ /, $fieldspecs))[3];
397 $fdtmapping{$fieldtag} = { 'name' => $fieldname,
398 'subfields' => $fieldsubfields,
399 'repeatable' => $fieldrepeatable };
400 }
401 elsif ($fdtfileline eq "***") {
402 $amongstdefinitions = 1;
403 }
404 }
405
406 close(FDT_FILE);
407
408 return %fdtmapping;
409}
410
411
412sub escape_metadata_value
413{
414 my $value = shift(@_);
415 $value =~ s/\</&lt;/g;
416 $value =~ s/\>/&gt;/g;
417 $value =~ s/\\/\\\\/g;
418 return $value;
419}
420
421
422sub clean_up_after_exploding
423{
424 my $self = shift(@_);
425
426 # Delete the FDT and XRF files too
427 &util::rm($self->{'fdt_file_path'});
428 &util::rm($self->{'xrf_file_path'});
429}
430
431
4321;
Note: See TracBrowser for help on using the repository browser.