1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # basebuildproc.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 for indexing (should be
|
---|
27 | # implemented by subclass) and storing in gdbm database
|
---|
28 |
|
---|
29 | package basebuildproc;
|
---|
30 |
|
---|
31 | eval {require bytes};
|
---|
32 |
|
---|
33 | use classify;
|
---|
34 | use doc;
|
---|
35 | use docproc;
|
---|
36 | use util;
|
---|
37 |
|
---|
38 | BEGIN {
|
---|
39 | @basebuildproc::ISA = ('docproc');
|
---|
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->{'outhandle'} = $outhandle;
|
---|
57 |
|
---|
58 | $self->{'classifiers'} = [];
|
---|
59 | $self->{'mode'} = "text";
|
---|
60 | $self->{'assocdir'} = $build_dir;
|
---|
61 | $self->{'dontgdbm'} = {};
|
---|
62 |
|
---|
63 | $self->{'index'} = "section:text";
|
---|
64 | $self->{'indexexparr'} = [];
|
---|
65 |
|
---|
66 | $self->{'output_handle'} = "STDOUT";
|
---|
67 | $self->{'num_docs'} = 0;
|
---|
68 | $self->{'num_sections'} = 0;
|
---|
69 | $self->{'num_bytes'} = 0;
|
---|
70 | $self->{'num_processed_bytes'} = 0;
|
---|
71 | $self->{'store_text'} = 1;
|
---|
72 |
|
---|
73 | # what level (section/document) the gdbm database - indexer intersection is
|
---|
74 | $self->{'gdbm_level'} = "section";
|
---|
75 | #used by browse interface
|
---|
76 | $self->{'doclist'} = [];
|
---|
77 |
|
---|
78 | $self->{'indexing_text'} = 0;
|
---|
79 |
|
---|
80 | return bless $self, $class;
|
---|
81 |
|
---|
82 | }
|
---|
83 |
|
---|
84 | sub reset {
|
---|
85 | my $self = shift (@_);
|
---|
86 |
|
---|
87 | $self->{'num_docs'} = 0;
|
---|
88 | $self->{'num_sections'} = 0;
|
---|
89 | $self->{'num_processed_bytes'} = 0;
|
---|
90 | $self->{'num_bytes'} = 0;
|
---|
91 | }
|
---|
92 |
|
---|
93 | sub get_num_docs {
|
---|
94 | my $self = shift (@_);
|
---|
95 |
|
---|
96 | return $self->{'num_docs'};
|
---|
97 | }
|
---|
98 |
|
---|
99 | sub get_num_sections {
|
---|
100 | my $self = shift (@_);
|
---|
101 |
|
---|
102 | return $self->{'num_sections'};
|
---|
103 | }
|
---|
104 |
|
---|
105 | # num_bytes is the actual number of bytes in the collection
|
---|
106 | # this is normally the same as what's processed during text compression
|
---|
107 | sub get_num_bytes {
|
---|
108 | my $self = shift (@_);
|
---|
109 |
|
---|
110 | return $self->{'num_bytes'};
|
---|
111 | }
|
---|
112 |
|
---|
113 | # num_processed_bytes is the number of bytes actually passed
|
---|
114 | # to mg for the current index
|
---|
115 | sub get_num_processed_bytes {
|
---|
116 | my $self = shift (@_);
|
---|
117 |
|
---|
118 | return $self->{'num_processed_bytes'};
|
---|
119 | }
|
---|
120 |
|
---|
121 | sub set_output_handle {
|
---|
122 | my $self = shift (@_);
|
---|
123 | my ($handle) = @_;
|
---|
124 |
|
---|
125 | $self->{'output_handle'} = $handle;
|
---|
126 | }
|
---|
127 |
|
---|
128 |
|
---|
129 | sub set_mode {
|
---|
130 | my $self = shift (@_);
|
---|
131 | my ($mode) = @_;
|
---|
132 |
|
---|
133 | $self->{'mode'} = $mode;
|
---|
134 | }
|
---|
135 |
|
---|
136 | sub set_assocdir {
|
---|
137 | my $self = shift (@_);
|
---|
138 | my ($assocdir) = @_;
|
---|
139 |
|
---|
140 | $self->{'assocdir'} = $assocdir;
|
---|
141 | }
|
---|
142 |
|
---|
143 | sub set_dontgdbm {
|
---|
144 | my $self = shift (@_);
|
---|
145 | my ($dontgdbm) = @_;
|
---|
146 |
|
---|
147 | $self->{'dontgdbm'} = $dontgdbm;
|
---|
148 | }
|
---|
149 |
|
---|
150 | sub set_index {
|
---|
151 | my $self = shift (@_);
|
---|
152 | my ($index, $indexexparr) = @_;
|
---|
153 |
|
---|
154 | $self->{'index'} = $index;
|
---|
155 | $self->{'indexexparr'} = $indexexparr if defined $indexexparr;
|
---|
156 | }
|
---|
157 |
|
---|
158 | sub set_index_languages {
|
---|
159 | my $self = shift (@_);
|
---|
160 | my ($lang_meta, $langarr) = @_;
|
---|
161 | $self->{'lang_meta'} = $lang_meta;
|
---|
162 | $self->{'langarr'} = $langarr;
|
---|
163 | }
|
---|
164 |
|
---|
165 | sub get_index {
|
---|
166 | my $self = shift (@_);
|
---|
167 |
|
---|
168 | return $self->{'index'};
|
---|
169 | }
|
---|
170 |
|
---|
171 | sub set_classifiers {
|
---|
172 | my $self = shift (@_);
|
---|
173 | my ($classifiers) = @_;
|
---|
174 |
|
---|
175 | $self->{'classifiers'} = $classifiers;
|
---|
176 | }
|
---|
177 |
|
---|
178 | sub set_indexing_text {
|
---|
179 | my $self = shift (@_);
|
---|
180 | my ($indexing_text) = @_;
|
---|
181 |
|
---|
182 | $self->{'indexing_text'} = $indexing_text;
|
---|
183 | }
|
---|
184 |
|
---|
185 | sub get_indexing_text {
|
---|
186 | my $self = shift (@_);
|
---|
187 |
|
---|
188 | return $self->{'indexing_text'};
|
---|
189 | }
|
---|
190 |
|
---|
191 | sub set_store_text {
|
---|
192 | my $self = shift (@_);
|
---|
193 | my ($store_text) = @_;
|
---|
194 |
|
---|
195 | $self->{'store_text'} = $store_text;
|
---|
196 | }
|
---|
197 | sub get_doc_list {
|
---|
198 | my $self = shift(@_);
|
---|
199 |
|
---|
200 | return @{$self->{'doclist'}};
|
---|
201 | }
|
---|
202 |
|
---|
203 | # the standard gdbm level is section, but you may want to change it to document
|
---|
204 | sub set_gdbm_level {
|
---|
205 | my $self= shift (@_);
|
---|
206 | my ($gdbm_level) = @_;
|
---|
207 |
|
---|
208 | $self->{'gdbm_level'} = $gdbm_level;
|
---|
209 | }
|
---|
210 |
|
---|
211 | sub process {
|
---|
212 | my $self = shift (@_);
|
---|
213 | my $method = $self->{'mode'};
|
---|
214 |
|
---|
215 | $self->$method(@_);
|
---|
216 | }
|
---|
217 |
|
---|
218 | sub infodb {
|
---|
219 | my $self = shift (@_);
|
---|
220 | my ($doc_obj, $filename) = @_;
|
---|
221 | my $handle = $self->{'output_handle'};
|
---|
222 |
|
---|
223 | my $doctype = $doc_obj->get_doc_type();
|
---|
224 |
|
---|
225 | # only output this document if it is one to be indexed
|
---|
226 | return if ($doctype ne "indexed_doc");
|
---|
227 |
|
---|
228 | my ($archivedir) = $filename =~ /^(.*?)(?:\/|\\)[^\/\\]*$/;
|
---|
229 | $archivedir = "" unless defined $archivedir;
|
---|
230 | $archivedir =~ s/\\/\//g;
|
---|
231 | $archivedir =~ s/^\/+//;
|
---|
232 | $archivedir =~ s/\/+$//;
|
---|
233 |
|
---|
234 | # resolve the final filenames of the files associated with this document
|
---|
235 | $self->assoc_files ($doc_obj, $archivedir);
|
---|
236 |
|
---|
237 | #GRB: moved 1/06/2004 from GRB01062004
|
---|
238 | #add this document to the browse structure
|
---|
239 | push(@{$self->{'doclist'}},$doc_obj->get_OID())
|
---|
240 | unless ($doctype eq "classification");
|
---|
241 |
|
---|
242 | # classify this document
|
---|
243 | &classify::classify_doc ($self->{'classifiers'}, $doc_obj);
|
---|
244 | #GRB: end of moved block
|
---|
245 |
|
---|
246 | # this is another document
|
---|
247 | $self->{'num_docs'} += 1 unless ($doctype eq "classification");
|
---|
248 |
|
---|
249 | # is this a paged or a hierarchical document
|
---|
250 | my ($thistype, $childtype) = $self->get_document_type ($doc_obj);
|
---|
251 |
|
---|
252 | my $section = $doc_obj->get_top_section ();
|
---|
253 | my $doc_OID = $doc_obj->get_OID();
|
---|
254 | my $first = 1;
|
---|
255 | my $url = "";
|
---|
256 | while (defined $section) {
|
---|
257 | # update a few statistics
|
---|
258 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
---|
259 | $self->{'num_sections'} += 1 unless ($doctype eq "classification");
|
---|
260 |
|
---|
261 | # output the section name
|
---|
262 | if ($section eq "") { print $handle "[$doc_OID]\n"; }
|
---|
263 | else { print $handle "[$doc_OID.$section]\n"; }
|
---|
264 |
|
---|
265 | # output the fact that this document is a document (unless doctype
|
---|
266 | # has been set to something else from within a plugin
|
---|
267 | my $dtype = $doc_obj->get_metadata_element ($section, "doctype");
|
---|
268 | if (!defined $dtype || $dtype !~ /\w/) {
|
---|
269 | print $handle "<doctype>doc\n";
|
---|
270 | }
|
---|
271 |
|
---|
272 | # output whether this node contains text
|
---|
273 | if ($doc_obj->get_text_length($section) > 0) {
|
---|
274 | print $handle "<hastxt>1\n";
|
---|
275 | } else {
|
---|
276 | print $handle "<hastxt>0\n";
|
---|
277 | }
|
---|
278 |
|
---|
279 | # output all the section metadata
|
---|
280 | my $metadata = $doc_obj->get_all_metadata ($section);
|
---|
281 | foreach my $pair (@$metadata) {
|
---|
282 | my ($field, $value) = (@$pair);
|
---|
283 |
|
---|
284 | if ($field ne "Identifier" && $field !~ /^gsdl/ &&
|
---|
285 | defined $value && $value ne "") {
|
---|
286 |
|
---|
287 | # escape problematic stuff
|
---|
288 | $value =~ s/\\/\\\\/g;
|
---|
289 | $value =~ s/\n/\\n/g;
|
---|
290 | $value =~ s/\r/\\r/g;
|
---|
291 | if ($value =~ /-{70,}/) {
|
---|
292 | # if value contains 70 or more hyphens in a row we need
|
---|
293 | # to escape them to prevent txt2db from treating them
|
---|
294 | # as a separator
|
---|
295 | $value =~ s/-/&\#045;/gi;
|
---|
296 | }
|
---|
297 |
|
---|
298 | # special case for URL metadata
|
---|
299 | if ($field =~ /^URL$/i) {
|
---|
300 | $url .= "[$value]\n";
|
---|
301 | if ($section eq "") {$url .= "<section>$doc_OID\n";}
|
---|
302 | else {$url .= "<section>$doc_OID.$section\n";}
|
---|
303 | $url .= '-' x 70 . "\n";
|
---|
304 | }
|
---|
305 |
|
---|
306 | if (!defined $self->{'dontgdbm'}->{$field}) {
|
---|
307 | print $handle "<$field>$value\n";
|
---|
308 | }
|
---|
309 | }
|
---|
310 | }
|
---|
311 |
|
---|
312 | # output archivedir if at top level
|
---|
313 | if ($section eq $doc_obj->get_top_section()) {
|
---|
314 | print $handle "<archivedir>$archivedir\n";
|
---|
315 | }
|
---|
316 |
|
---|
317 | # output document display type
|
---|
318 | if ($first) {
|
---|
319 | print $handle "<thistype>$thistype\n";
|
---|
320 | }
|
---|
321 |
|
---|
322 | if ($self->{'gdbm_level'} eq "document") {
|
---|
323 | # doc num is num_docs not num_sections
|
---|
324 | # output the matching document number
|
---|
325 | print $handle "<docnum>$self->{'num_docs'}\n";
|
---|
326 | } else {
|
---|
327 | # output a list of children
|
---|
328 | my $children = $doc_obj->get_children ($section);
|
---|
329 | if (scalar(@$children) > 0) {
|
---|
330 | print $handle "<childtype>$childtype\n";
|
---|
331 | print $handle "<contains>";
|
---|
332 | my $firstchild = 1;
|
---|
333 | foreach my $child (@$children) {
|
---|
334 | print $handle ";" unless $firstchild;
|
---|
335 | $firstchild = 0;
|
---|
336 | if ($child =~ /^.*?\.(\d+)$/) {
|
---|
337 | print $handle "\".$1";
|
---|
338 | } else {
|
---|
339 | print $handle "\".$child";
|
---|
340 | }
|
---|
341 | # if ($child eq "") { print $handle "$doc_OID"; }
|
---|
342 | # elsif ($section eq "") { print $handle "$doc_OID.$child"; }
|
---|
343 | # else { print $handle "$doc_OID.$section.$child"; }
|
---|
344 | }
|
---|
345 | print $handle "\n";
|
---|
346 | }
|
---|
347 | #output the matching doc number
|
---|
348 | print $handle "<docnum>$self->{'num_sections'}\n";
|
---|
349 |
|
---|
350 | }
|
---|
351 |
|
---|
352 | print $handle '-' x 70, "\n";
|
---|
353 |
|
---|
354 |
|
---|
355 | # output a database entry for the document number
|
---|
356 | if ($self->{'gdbm_level'} eq "document") {
|
---|
357 | print $handle "[$self->{'num_docs'}]\n";
|
---|
358 | print $handle "<section>$doc_OID\n";
|
---|
359 | }
|
---|
360 | else {
|
---|
361 | print $handle "[$self->{'num_sections'}]\n";
|
---|
362 | if ($section eq "") { print $handle "<section>$doc_OID\n"; }
|
---|
363 | else { print $handle "<section>$doc_OID.$section\n"; }
|
---|
364 | }
|
---|
365 | print $handle '-' x 70, "\n";
|
---|
366 |
|
---|
367 | # output entry for url
|
---|
368 | if ($url ne "") {
|
---|
369 | print $handle $url;
|
---|
370 | }
|
---|
371 |
|
---|
372 | $first = 0;
|
---|
373 | $section = $doc_obj->get_next_section($section);
|
---|
374 | last if ($self->{'gdbm_level'} eq "document"); # if no sections wanted, only gdbm the docs
|
---|
375 | }
|
---|
376 |
|
---|
377 | #GRB01062004: see code above moved from here
|
---|
378 | }
|
---|
379 |
|
---|
380 |
|
---|
381 | sub text {
|
---|
382 | my $self = shift (@_);
|
---|
383 | my ($doc_obj) = @_;
|
---|
384 |
|
---|
385 | my $handle = $self->{'outhandle'};
|
---|
386 | print $handle "basebuildproc::text function must be implemented in sub classes\n";
|
---|
387 | die "\n";
|
---|
388 | }
|
---|
389 |
|
---|
390 | # should the document be indexed - according to the subcollection and language
|
---|
391 | # specification.
|
---|
392 | sub is_subcollection_doc {
|
---|
393 | my $self = shift (@_);
|
---|
394 | my ($doc_obj) = @_;
|
---|
395 |
|
---|
396 | my $indexed_doc = 1;
|
---|
397 | foreach my $indexexp (@{$self->{'indexexparr'}}) {
|
---|
398 | $indexed_doc = 0;
|
---|
399 | my ($field, $exp, $options) = split /\//, $indexexp;
|
---|
400 | if (defined ($field) && defined ($exp)) {
|
---|
401 | my ($bool) = $field =~ /^(.)/;
|
---|
402 | $field =~ s/^.// if $bool eq '!';
|
---|
403 | if ($field =~ /^filename$/i) {
|
---|
404 | $field = $doc_obj->get_source_filename();
|
---|
405 | } else {
|
---|
406 | $field = $doc_obj->get_metadata_element($doc_obj->get_top_section(), $field);
|
---|
407 | }
|
---|
408 | next unless defined $field;
|
---|
409 | if ($bool eq '!') {
|
---|
410 | if ($options =~ /^i$/i) {
|
---|
411 | if ($field !~ /$exp/i) {$indexed_doc = 1; last;}
|
---|
412 | } else {
|
---|
413 | if ($field !~ /$exp/) {$indexed_doc = 1; last;}
|
---|
414 | }
|
---|
415 | } else {
|
---|
416 | if ($options =~ /^i$/i) {
|
---|
417 | if ($field =~ /$exp/i) {$indexed_doc = 1; last;}
|
---|
418 | } else {
|
---|
419 | if ($field =~ /$exp/) {$indexed_doc = 1; last;}
|
---|
420 | }
|
---|
421 | }
|
---|
422 | }
|
---|
423 | }
|
---|
424 |
|
---|
425 | # if this doc is so far in the sub collection, and we have lang info,
|
---|
426 | # now we check the languages to see if it matches
|
---|
427 | if($indexed_doc && defined $self->{'lang_meta'}) {
|
---|
428 | $indexed_doc = 0;
|
---|
429 | my $field = $doc_obj->get_metadata_element($doc_obj->get_top_section(), $self->{'lang_meta'});
|
---|
430 | if (defined $field) {
|
---|
431 | foreach my $lang (@{$self->{'langarr'}}) {
|
---|
432 | my ($bool) = $lang =~ /^(.)/;
|
---|
433 | if ($bool eq '!') {
|
---|
434 | $lang =~ s/^.//;
|
---|
435 | if ($field !~ /$lang/) {
|
---|
436 | $indexed_doc = 1; last;
|
---|
437 | }
|
---|
438 | } else {
|
---|
439 | if ($field =~ /$lang/) {
|
---|
440 | $indexed_doc = 1; last;
|
---|
441 | }
|
---|
442 | }
|
---|
443 | }
|
---|
444 | }
|
---|
445 | }
|
---|
446 | return $indexed_doc;
|
---|
447 |
|
---|
448 | }
|
---|
449 |
|
---|
450 | # use 'Paged' if document has no more than 2 levels
|
---|
451 | # and each section at second level has a number for
|
---|
452 | # Title metadata
|
---|
453 | # also use Paged if gsdlthistype metadata is set to Paged
|
---|
454 | sub get_document_type {
|
---|
455 | my $self = shift (@_);
|
---|
456 | my ($doc_obj) = @_;
|
---|
457 |
|
---|
458 | my $thistype = "VList";
|
---|
459 | my $childtype = "VList";
|
---|
460 | my $title;
|
---|
461 | my @tmp = ();
|
---|
462 |
|
---|
463 | my $section = $doc_obj->get_top_section ();
|
---|
464 |
|
---|
465 | my $gsdlthistype = $doc_obj->get_metadata_element ($section, "gsdlthistype");
|
---|
466 | if (defined $gsdlthistype) {
|
---|
467 | if ($gsdlthistype eq "Paged") {
|
---|
468 | $childtype = "Paged";
|
---|
469 | if ($doc_obj->get_text_length ($doc_obj->get_top_section())) {
|
---|
470 | $thistype = "Paged";
|
---|
471 | } else {
|
---|
472 | $thistype = "Invisible";
|
---|
473 | }
|
---|
474 |
|
---|
475 | return ($thistype, $childtype);
|
---|
476 | } elsif ($gsdlthistype eq "Hierarchy") {
|
---|
477 | return ($thistype, $childtype); # use VList, VList
|
---|
478 | }
|
---|
479 | }
|
---|
480 | my $first = 1;
|
---|
481 | while (defined $section) {
|
---|
482 | @tmp = split /\./, $section;
|
---|
483 | if (scalar(@tmp) > 1) {
|
---|
484 | return ($thistype, $childtype);
|
---|
485 | }
|
---|
486 | if (!$first) {
|
---|
487 | $title = $doc_obj->get_metadata_element ($section, "Title");
|
---|
488 | if (!defined $title || $title !~ /^\d+$/) {
|
---|
489 | return ($thistype, $childtype);
|
---|
490 | }
|
---|
491 | }
|
---|
492 | $first = 0;
|
---|
493 | $section = $doc_obj->get_next_section($section);
|
---|
494 | }
|
---|
495 | if ($doc_obj->get_text_length ($doc_obj->get_top_section())) {
|
---|
496 | $thistype = "Paged";
|
---|
497 | } else {
|
---|
498 | $thistype = "Invisible";
|
---|
499 | }
|
---|
500 | $childtype = "Paged";
|
---|
501 | return ($thistype, $childtype);
|
---|
502 | }
|
---|
503 |
|
---|
504 | sub assoc_files {
|
---|
505 | my $self = shift (@_);
|
---|
506 | my ($doc_obj, $archivedir) = @_;
|
---|
507 | my ($afile);
|
---|
508 |
|
---|
509 | foreach my $assoc_file (@{$doc_obj->get_assoc_files()}) {
|
---|
510 | # if assoc file starts with a slash, we put it relative to the assoc
|
---|
511 | # dir, otherwise it is relative to the HASH... directory
|
---|
512 | if ($assoc_file->[1] =~ m@^[/\\]@) {
|
---|
513 | $afile = &util::filename_cat($self->{'assocdir'},$assoc_file->[1]);
|
---|
514 | } else {
|
---|
515 | $afile = &util::filename_cat($self->{'assocdir'}, $archivedir, $assoc_file->[1]);
|
---|
516 | }
|
---|
517 | &util::hard_link ($assoc_file->[0], $afile);
|
---|
518 | }
|
---|
519 | }
|
---|
520 |
|
---|