1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # mgbuildproc.pm --
|
---|
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) 1999 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 | # This document processor outputs a document
|
---|
27 | # for mg to process
|
---|
28 |
|
---|
29 |
|
---|
30 | package mgbuildproc;
|
---|
31 |
|
---|
32 | use classify;
|
---|
33 | use doc;
|
---|
34 | use docproc;
|
---|
35 | use util;
|
---|
36 |
|
---|
37 | BEGIN {
|
---|
38 | @ISA = ('docproc');
|
---|
39 | }
|
---|
40 |
|
---|
41 |
|
---|
42 | sub new {
|
---|
43 | my ($class, $collection, $source_dir, $build_dir,
|
---|
44 | $verbosity, $outhandle) = @_;
|
---|
45 | my $self = new docproc ();
|
---|
46 |
|
---|
47 | # outhandle is where all the debugging info goes
|
---|
48 | # output_handle is where the output of the plugins is piped
|
---|
49 | # to (i.e. mg, gdbm etc.)
|
---|
50 | $outhandle = STDERR unless defined $outhandle;
|
---|
51 |
|
---|
52 | $self->{'collection'} = $collection;
|
---|
53 | $self->{'source_dir'} = $source_dir;
|
---|
54 | $self->{'build_dir'} = $build_dir;
|
---|
55 | $self->{'verbosity'} = $verbosity;
|
---|
56 | $self->{'classifiers'} = [];
|
---|
57 | $self->{'mode'} = "text";
|
---|
58 | $self->{'assocdir'} = $build_dir;
|
---|
59 | $self->{'dontgdbm'} = {};
|
---|
60 | $self->{'index'} = "section:text";
|
---|
61 | $self->{'indexexparr'} = [];
|
---|
62 | $self->{'output_handle'} = "STDOUT";
|
---|
63 | $self->{'num_docs'} = 0;
|
---|
64 | $self->{'num_sections'} = 0;
|
---|
65 | $self->{'num_bytes'} = 0;
|
---|
66 | $self->{'num_processed_bytes'} = 0;
|
---|
67 | $self->{'outhandle'} = $outhandle;
|
---|
68 |
|
---|
69 | $self->{'indexing_text'} = 0;
|
---|
70 |
|
---|
71 | return bless $self, $class;
|
---|
72 | }
|
---|
73 |
|
---|
74 | sub reset {
|
---|
75 | my $self = shift (@_);
|
---|
76 |
|
---|
77 | $self->{'num_docs'} = 0;
|
---|
78 | $self->{'num_sections'} = 0;
|
---|
79 | $self->{'num_processed_bytes'} = 0;
|
---|
80 | $self->{'num_bytes'} = 0;
|
---|
81 | }
|
---|
82 |
|
---|
83 | sub get_num_docs {
|
---|
84 | my $self = shift (@_);
|
---|
85 |
|
---|
86 | return $self->{'num_docs'};
|
---|
87 | }
|
---|
88 |
|
---|
89 | sub get_num_sections {
|
---|
90 | my $self = shift (@_);
|
---|
91 |
|
---|
92 | return $self->{'num_sections'};
|
---|
93 | }
|
---|
94 |
|
---|
95 | # num_bytes is the actual number of bytes in the collection
|
---|
96 | # this is normally the same as what's processed during text compression
|
---|
97 | sub get_num_bytes {
|
---|
98 | my $self = shift (@_);
|
---|
99 |
|
---|
100 | return $self->{'num_bytes'};
|
---|
101 | }
|
---|
102 |
|
---|
103 | # num_processed_bytes is the number of bytes actually passed
|
---|
104 | # to mg for the current index
|
---|
105 | sub get_num_processed_bytes {
|
---|
106 | my $self = shift (@_);
|
---|
107 |
|
---|
108 | return $self->{'num_processed_bytes'};
|
---|
109 | }
|
---|
110 |
|
---|
111 | sub set_output_handle {
|
---|
112 | my $self = shift (@_);
|
---|
113 | my ($handle) = @_;
|
---|
114 |
|
---|
115 | $self->{'output_handle'} = $handle;
|
---|
116 | }
|
---|
117 |
|
---|
118 | sub set_mode {
|
---|
119 | my $self = shift (@_);
|
---|
120 | my ($mode) = @_;
|
---|
121 |
|
---|
122 | $self->{'mode'} = $mode;
|
---|
123 | }
|
---|
124 |
|
---|
125 | sub set_assocdir {
|
---|
126 | my $self = shift (@_);
|
---|
127 | my ($assocdir) = @_;
|
---|
128 |
|
---|
129 | $self->{'assocdir'} = $assocdir;
|
---|
130 | }
|
---|
131 |
|
---|
132 | sub set_dontgdbm {
|
---|
133 | my $self = shift (@_);
|
---|
134 | my ($dontgdbm) = @_;
|
---|
135 |
|
---|
136 | $self->{'dontgdbm'} = $dontgdbm;
|
---|
137 | }
|
---|
138 |
|
---|
139 | sub set_index {
|
---|
140 | my $self = shift (@_);
|
---|
141 | my ($index, $indexexparr) = @_;
|
---|
142 |
|
---|
143 | $self->{'index'} = $index;
|
---|
144 | $self->{'indexexparr'} = $indexexparr if defined $indexexparr;
|
---|
145 | }
|
---|
146 |
|
---|
147 | sub get_index {
|
---|
148 | my $self = shift (@_);
|
---|
149 |
|
---|
150 | return $self->{'index'};
|
---|
151 | }
|
---|
152 |
|
---|
153 | sub set_classifiers {
|
---|
154 | my $self = shift (@_);
|
---|
155 | my ($classifiers) = @_;
|
---|
156 |
|
---|
157 | $self->{'classifiers'} = $classifiers;
|
---|
158 | }
|
---|
159 |
|
---|
160 | sub set_indexing_text {
|
---|
161 | my $self = shift (@_);
|
---|
162 | my ($indexing_text) = @_;
|
---|
163 |
|
---|
164 | $self->{'indexing_text'} = $indexing_text;
|
---|
165 | }
|
---|
166 |
|
---|
167 | sub get_indexing_text {
|
---|
168 | my $self = shift (@_);
|
---|
169 |
|
---|
170 | return $self->{'indexing_text'};
|
---|
171 | }
|
---|
172 |
|
---|
173 | sub process {
|
---|
174 | my $self = shift (@_);
|
---|
175 | my $method = $self->{'mode'};
|
---|
176 |
|
---|
177 | $self->$method(@_);
|
---|
178 | }
|
---|
179 |
|
---|
180 | # use 'Paged' if document has no more than 2 levels
|
---|
181 | # and each section at second level has a number for
|
---|
182 | # Title metadata
|
---|
183 | sub get_document_type {
|
---|
184 | my $self = shift (@_);
|
---|
185 | my ($doc_obj) = @_;
|
---|
186 |
|
---|
187 | my $thistype = "VList";
|
---|
188 | my $childtype = "VList";
|
---|
189 | my $title;
|
---|
190 | my @tmp = ();
|
---|
191 |
|
---|
192 | my $section = $doc_obj->get_top_section ();
|
---|
193 | my $first = 1;
|
---|
194 | while (defined $section) {
|
---|
195 | @tmp = split /\./, $section;
|
---|
196 | if (scalar(@tmp) > 1) {
|
---|
197 | return ($thistype, $childtype);
|
---|
198 | }
|
---|
199 | if (!$first) {
|
---|
200 | $title = $doc_obj->get_metadata_element ($section, "Title");
|
---|
201 | if (!defined $title || $title !~ /^\d+$/) {
|
---|
202 | return ($thistype, $childtype);
|
---|
203 | }
|
---|
204 | }
|
---|
205 | $first = 0;
|
---|
206 | $section = $doc_obj->get_next_section($section);
|
---|
207 | }
|
---|
208 | if ($doc_obj->get_text_length ($doc_obj->get_top_section())) {
|
---|
209 | $thistype = "Paged";
|
---|
210 | } else {
|
---|
211 | $thistype = "Invisible";
|
---|
212 | }
|
---|
213 | $childtype = "Paged";
|
---|
214 | return ($thistype, $childtype);
|
---|
215 | }
|
---|
216 |
|
---|
217 | sub assoc_files {
|
---|
218 | my $self = shift (@_);
|
---|
219 | my ($doc_obj, $archivedir) = @_;
|
---|
220 | my ($afile);
|
---|
221 |
|
---|
222 | foreach my $assoc_file (@{$doc_obj->get_assoc_files()}) {
|
---|
223 | # if assoc file contains directory structure of
|
---|
224 | # its own use it, otherwise use HASH... directory
|
---|
225 | if ($assoc_file->[1] =~ /[\/\\]/) {
|
---|
226 | $afile = &util::filename_cat($self->{'assocdir'}, $assoc_file->[1]);
|
---|
227 | } else {
|
---|
228 | $afile = &util::filename_cat($self->{'assocdir'}, $archivedir, $assoc_file->[1]);
|
---|
229 | }
|
---|
230 | &util::hard_link ($assoc_file->[0], $afile);
|
---|
231 | }
|
---|
232 | }
|
---|
233 |
|
---|
234 | sub infodb {
|
---|
235 | my $self = shift (@_);
|
---|
236 | my ($doc_obj, $filename) = @_;
|
---|
237 | my $handle = $self->{'output_handle'};
|
---|
238 | # $handle = "main::STDOUT";
|
---|
239 |
|
---|
240 | my $doctype = $doc_obj->get_doc_type();
|
---|
241 |
|
---|
242 | # only output this document if it is one to be indexed
|
---|
243 | return if ($doctype ne "indexed_doc");
|
---|
244 |
|
---|
245 | my ($archivedir) = $filename =~ /^(.*?)(?:\/|\\)[^\/\\]*$/;
|
---|
246 | $archivedir = "" unless defined $archivedir;
|
---|
247 | $archivedir =~ s/\\/\//g;
|
---|
248 | $archivedir =~ s/^\/+//;
|
---|
249 | $archivedir =~ s/\/+$//;
|
---|
250 |
|
---|
251 | $self->assoc_files ($doc_obj, $archivedir);
|
---|
252 |
|
---|
253 | # this is another document
|
---|
254 | $self->{'num_docs'} += 1 unless ($doctype eq "classification");
|
---|
255 |
|
---|
256 | # is this a paged or a hierarchical document
|
---|
257 | my ($thistype, $childtype) = $self->get_document_type ($doc_obj);
|
---|
258 |
|
---|
259 | my $section = $doc_obj->get_top_section ();
|
---|
260 | my $doc_OID = $doc_obj->get_OID();
|
---|
261 | my $first = 1;
|
---|
262 | my $url = "";
|
---|
263 | while (defined $section) {
|
---|
264 | # update a few statistics
|
---|
265 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
---|
266 | $self->{'num_sections'} += 1 unless ($doctype eq "classification");
|
---|
267 |
|
---|
268 | # output the section name
|
---|
269 | if ($section eq "") { print $handle "[$doc_OID]\n"; }
|
---|
270 | else { print $handle "[$doc_OID.$section]\n"; }
|
---|
271 |
|
---|
272 | # output the fact that this document is a document (unless doctype
|
---|
273 | # has been set to something else from within a plugin
|
---|
274 | my $dtype = $doc_obj->get_metadata_element ($section, "doctype");
|
---|
275 | if (!defined $dtype || $dtype !~ /\w/) {
|
---|
276 | print $handle "<doctype>doc\n";
|
---|
277 | }
|
---|
278 |
|
---|
279 | # output whether this node contains text
|
---|
280 | if ($doc_obj->get_text_length($section) > 0) {
|
---|
281 | print $handle "<hastxt>1\n";
|
---|
282 | } else {
|
---|
283 | print $handle "<hastxt>0\n";
|
---|
284 | }
|
---|
285 |
|
---|
286 | # output all the section metadata
|
---|
287 | my $metadata = $doc_obj->get_all_metadata ($section);
|
---|
288 | foreach my $pair (@$metadata) {
|
---|
289 | my ($field, $value) = (@$pair);
|
---|
290 |
|
---|
291 | if ($field ne "Identifier" && $field !~ /^gsdl/ &&
|
---|
292 | defined $value && $value ne "") {
|
---|
293 |
|
---|
294 | # escape problematic stuff
|
---|
295 | $value =~ s/\\/\\\\/g;
|
---|
296 | $value =~ s/\n/\\n/g;
|
---|
297 | $value =~ s/\r/\\r/g;
|
---|
298 |
|
---|
299 | # special case for URL metadata
|
---|
300 | if ($field =~ /^URL$/i) {
|
---|
301 | $url .= "[$value]\n";
|
---|
302 | if ($section eq "") {$url .= "<section>$doc_OID\n";}
|
---|
303 | else {$url .= "<section>$doc_OID.$section\n";}
|
---|
304 | $url .= '-' x 70 . "\n";
|
---|
305 | }
|
---|
306 |
|
---|
307 | if (!defined $self->{'dontgdbm'}->{$field}) {
|
---|
308 | print $handle "<$field>$value\n";
|
---|
309 | }
|
---|
310 | }
|
---|
311 | }
|
---|
312 |
|
---|
313 | # output archivedir if at top level
|
---|
314 | if ($section eq $doc_obj->get_top_section()) {
|
---|
315 | print $handle "<archivedir>$archivedir\n";
|
---|
316 | }
|
---|
317 |
|
---|
318 | # output document display type
|
---|
319 | if ($first) {
|
---|
320 | print $handle "<thistype>$thistype\n";
|
---|
321 | }
|
---|
322 |
|
---|
323 | # output a list of children
|
---|
324 | my $children = $doc_obj->get_children ($section);
|
---|
325 | if (scalar(@$children) > 0) {
|
---|
326 | print $handle "<childtype>$childtype\n";
|
---|
327 | print $handle "<contains>";
|
---|
328 | my $firstchild = 1;
|
---|
329 | foreach my $child (@$children) {
|
---|
330 | print $handle ";" unless $firstchild;
|
---|
331 | $firstchild = 0;
|
---|
332 | if ($child =~ /^.*?\.(\d+)$/) {
|
---|
333 | print $handle "\".$1";
|
---|
334 | } else {
|
---|
335 | print $handle "\".$child";
|
---|
336 | }
|
---|
337 | # if ($child eq "") { print $handle "$doc_OID"; }
|
---|
338 | # elsif ($section eq "") { print $handle "$doc_OID.$child"; }
|
---|
339 | # else { print $handle "$doc_OID.$section.$child"; }
|
---|
340 | }
|
---|
341 | print $handle "\n";
|
---|
342 | }
|
---|
343 |
|
---|
344 | # output the matching document number
|
---|
345 | print $handle "<docnum>$self->{'num_sections'}\n";
|
---|
346 |
|
---|
347 | print $handle '-' x 70, "\n";
|
---|
348 |
|
---|
349 |
|
---|
350 | # output a database entry for the document number
|
---|
351 | print $handle "[$self->{'num_sections'}]\n";
|
---|
352 | if ($section eq "") { print $handle "<section>$doc_OID\n"; }
|
---|
353 | else { print $handle "<section>$doc_OID.$section\n"; }
|
---|
354 | print $handle '-' x 70, "\n";
|
---|
355 |
|
---|
356 | # output entry for url
|
---|
357 | if ($url ne "") {
|
---|
358 | print $handle $url;
|
---|
359 | }
|
---|
360 |
|
---|
361 | $first = 0;
|
---|
362 | $section = $doc_obj->get_next_section($section);
|
---|
363 | }
|
---|
364 |
|
---|
365 | # classify this document
|
---|
366 | &classify::classify_doc ($self->{'classifiers'}, $doc_obj);
|
---|
367 |
|
---|
368 | }
|
---|
369 |
|
---|
370 | sub find_paragraphs {
|
---|
371 | $_[1] =~ s/(<p\b)/\cC$1/gi;
|
---|
372 | }
|
---|
373 |
|
---|
374 | sub filter_text {
|
---|
375 | # $self->filter_text ($field, $new_text);
|
---|
376 | # don't want to do anything for this version, however,
|
---|
377 | # in a particular collection you might want to override
|
---|
378 | # this method to post-process certain fields depending on
|
---|
379 | # the field, or whether we are outputting it for indexing
|
---|
380 | }
|
---|
381 |
|
---|
382 | sub text {
|
---|
383 | my $self = shift (@_);
|
---|
384 | my ($doc_obj) = @_;
|
---|
385 | my $handle = $self->{'output_handle'};
|
---|
386 | my $indexed_doc = 1;
|
---|
387 |
|
---|
388 | # only output this document if it is one to be indexed
|
---|
389 | return if ($doc_obj->get_doc_type() ne "indexed_doc");
|
---|
390 |
|
---|
391 | # see if this document belongs to this subcollection
|
---|
392 | foreach my $indexexp (@{$self->{'indexexparr'}}) {
|
---|
393 | $indexed_doc = 0;
|
---|
394 | my ($field, $exp, $options) = split /\//, $indexexp;
|
---|
395 | if (defined ($field) && defined ($exp)) {
|
---|
396 | my ($bool) = $field =~ /^(.)/;
|
---|
397 | $field =~ s/^.// if $bool eq '!';
|
---|
398 | if ($field =~ /^filename$/i) {
|
---|
399 | $field = $doc_obj->get_source_filename();
|
---|
400 | } else {
|
---|
401 | $field = $doc_obj->get_metadata_element($doc_obj->get_top_section(), $field);
|
---|
402 | }
|
---|
403 | next unless defined $field;
|
---|
404 | if ($bool eq '!') {
|
---|
405 | if ($options =~ /^i$/i) {
|
---|
406 | if ($field !~ /$exp/i) {$indexed_doc = 1; last;}
|
---|
407 | } else {
|
---|
408 | if ($field !~ /$exp/) {$indexed_doc = 1; last;}
|
---|
409 | }
|
---|
410 | } else {
|
---|
411 | if ($options =~ /^i$/i) {
|
---|
412 | if ($field =~ /$exp/i) {$indexed_doc = 1; last;}
|
---|
413 | } else {
|
---|
414 | if ($field =~ /$exp/) {$indexed_doc = 1; last;}
|
---|
415 | }
|
---|
416 | }
|
---|
417 | }
|
---|
418 | }
|
---|
419 |
|
---|
420 | # this is another document
|
---|
421 | $self->{'num_docs'} += 1;
|
---|
422 |
|
---|
423 | # get the parameters for the output
|
---|
424 | my ($level, $fields) = split (/:/, $self->{'index'});
|
---|
425 | $fields =~ s/\ball\b/Title,Creator,text/;
|
---|
426 | $fields =~ s/\btopall\b/topTitle,topCreator,toptext/;
|
---|
427 |
|
---|
428 | my $doc_section = 0; # just for this document
|
---|
429 | my $text = "";
|
---|
430 | my $text_extra = "";
|
---|
431 |
|
---|
432 | # get the text for this document
|
---|
433 | my $section = $doc_obj->get_top_section();
|
---|
434 | while (defined $section) {
|
---|
435 | # update a few statistics
|
---|
436 | $doc_section++;
|
---|
437 | $self->{'num_sections'} += 1;
|
---|
438 |
|
---|
439 | if ($indexed_doc) {
|
---|
440 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
---|
441 | foreach my $field (split (/,/, $fields)) {
|
---|
442 | # only deal with this field if it doesn't start with top or
|
---|
443 | # this is the first section
|
---|
444 | my $real_field = $field;
|
---|
445 | if (!($real_field =~ s/^top//) || ($doc_section == 1)) {
|
---|
446 | my $new_text = "";
|
---|
447 | if ($real_field eq "text") {
|
---|
448 | $new_text = $doc_obj->get_text ($section);
|
---|
449 | $self->{'num_processed_bytes'} += length ($new_text);
|
---|
450 | $new_text =~ s/[\cB\cC]//g;
|
---|
451 | $self->find_paragraphs($new_text);
|
---|
452 |
|
---|
453 | } else {
|
---|
454 | my $first = 1;
|
---|
455 | foreach $meta (@{$doc_obj->get_metadata ($section, $real_field)}) {
|
---|
456 | $meta =~ s/[\cB\cC]//g;
|
---|
457 | $self->{'num_processed_bytes'} += length ($meta);
|
---|
458 | $new_text .= "\cC" unless $first;
|
---|
459 | $new_text .= $meta;
|
---|
460 | $first = 0;
|
---|
461 | }
|
---|
462 | }
|
---|
463 |
|
---|
464 | # filter the text
|
---|
465 | $self->filter_text ($field, $new_text);
|
---|
466 |
|
---|
467 | $text .= "$new_text\cC";
|
---|
468 | }
|
---|
469 | }
|
---|
470 | }
|
---|
471 |
|
---|
472 | if ($level eq "document") { $text_extra .= "\cB"; }
|
---|
473 | else { $text .= "\cB"; }
|
---|
474 |
|
---|
475 | $section = $doc_obj->get_next_section($section);
|
---|
476 | }
|
---|
477 |
|
---|
478 | print $handle "$text$text_extra";
|
---|
479 | }
|
---|
480 |
|
---|
481 | 1;
|
---|
482 |
|
---|