1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # IncrementalBuildUtils.pm -- API to assist incremental building
|
---|
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 (C) 2006 DL Consulting Ltd and 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 | # /** Initial versions of these functions by John Thompson, revisions by
|
---|
26 | # * and turning it into a package by John Rowe. Used heavily by
|
---|
27 | # * basebuilder::remove_document() and getdocument.pl
|
---|
28 | # *
|
---|
29 | # * @version 1.0 Initial version by John Thompson
|
---|
30 | # * @version 1.1 Addition of get_document and change of get_document_as_xml
|
---|
31 | # * by John Rowe
|
---|
32 | # * @version 2.0 Package version including seperation from calling code and
|
---|
33 | # * modularisation by creating gdbmget, gdbmset and
|
---|
34 | # * get_database_path by John Rowe
|
---|
35 | # *
|
---|
36 | # * @author John Thompson, DL Consulting Ltd.
|
---|
37 | # * @author John Rowe, DL Consulting Ltd.
|
---|
38 | # */
|
---|
39 | ###########################################################################
|
---|
40 | package IncrementalBuildUtils;
|
---|
41 |
|
---|
42 | BEGIN {
|
---|
43 | die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
|
---|
44 | die "GSDLOS not set\n" unless defined $ENV{'GSDLOS'};
|
---|
45 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib");
|
---|
46 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/cpan");
|
---|
47 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugins");
|
---|
48 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/classify");
|
---|
49 | }
|
---|
50 |
|
---|
51 | use doc;
|
---|
52 | use cfgread;
|
---|
53 | use colcfg;
|
---|
54 | use strict;
|
---|
55 | use util;
|
---|
56 |
|
---|
57 | use ClassifyTreeModel;
|
---|
58 | use IncrementalDocument;
|
---|
59 |
|
---|
60 | # Change debugging to 1 if you want verbose debugging output
|
---|
61 | my $debug = 1;
|
---|
62 |
|
---|
63 | # Ensure the collection specific binaries are on the search path
|
---|
64 | my $path_separator = ":";
|
---|
65 | if($ENV{'GSDLOS'} =~ /win/) {
|
---|
66 | $path_separator = ";";
|
---|
67 | }
|
---|
68 | $ENV{'PATH'} = &util::filename_cat($ENV{'GSDLHOME'}, "bin", $ENV{'GSDLOS'}) . $path_separator . &util::filename_cat($ENV{'GSDLHOME'}, "bin", "script") . $path_separator.$ENV{'PATH'};
|
---|
69 |
|
---|
70 | # /**
|
---|
71 | # */
|
---|
72 | sub addDocument()
|
---|
73 | {
|
---|
74 | my ($collection, $doc_obj, $section, $updateindex) = @_;
|
---|
75 |
|
---|
76 | $updateindex = 0 unless defined($updateindex);
|
---|
77 |
|
---|
78 | print STDERR "IncrementalBuildUtils::addDocument('$collection',$doc_obj,'$section')\n" unless !$debug;
|
---|
79 | # Gonna need to know in several places whether this is the top section
|
---|
80 | # of the document or not
|
---|
81 | my $is_top = ($section eq $doc_obj->get_top_section());
|
---|
82 |
|
---|
83 | # Retrieve all of the metadata from this document object only - not any
|
---|
84 | # child documents
|
---|
85 | my $metadata = $doc_obj->get_all_metadata($section);
|
---|
86 | # Check and add the docnum first
|
---|
87 | my $found_docnum = 0;
|
---|
88 | foreach my $pair (@$metadata)
|
---|
89 | {
|
---|
90 | my ($key, $value) = (@$pair);
|
---|
91 | if ($key eq "docnum")
|
---|
92 | {
|
---|
93 | &setDocumentMetadata($collection, $doc_obj->get_OID() . "$section", $key, "", $value, $updateindex);
|
---|
94 | $found_docnum = 1;
|
---|
95 | }
|
---|
96 | }
|
---|
97 |
|
---|
98 | if (!$found_docnum)
|
---|
99 | {
|
---|
100 | die("Fatal Error! Tried to add document without providing docnum");
|
---|
101 | }
|
---|
102 |
|
---|
103 | # Add it piece by piece - this depends on the loading of a blank document
|
---|
104 | # working the way it should.
|
---|
105 | foreach my $pair (@$metadata)
|
---|
106 | {
|
---|
107 | my ($key, $value) = (@$pair);
|
---|
108 | if ($key ne "Identifier" && $key ne "docnum" && $key !~ /^gsdl/ && defined $value && $value ne "")
|
---|
109 | {
|
---|
110 | # escape problematic stuff
|
---|
111 | $value =~ s/\\/\\\\/g;
|
---|
112 | $value =~ s/\n/\\n/g;
|
---|
113 | $value =~ s/\r/\\r/g;
|
---|
114 | if ($value =~ /-{70,}/)
|
---|
115 | {
|
---|
116 | # if value contains 70 or more hyphens in a row we need
|
---|
117 | # to escape them to prevent txt2db from treating them
|
---|
118 | # as a separator
|
---|
119 | $value =~ s/-/&\#045;/gi;
|
---|
120 | }
|
---|
121 | # Go ahead and set the metadata
|
---|
122 | &setDocumentMetadata($collection, $doc_obj->get_OID() . "$section", $key, "", $value, $updateindex);
|
---|
123 | }
|
---|
124 | }
|
---|
125 | # We now have to load the browselist node too. We create a ClassifyTreeNode
|
---|
126 | # based on a dummy model.
|
---|
127 | # Note: only if section is the top section
|
---|
128 | if ($is_top)
|
---|
129 | {
|
---|
130 | my $dummy_model = new ClassifyTreeModel($collection, "");
|
---|
131 | my $browselist_node = new ClassifyTreeNode($dummy_model, "browselist");
|
---|
132 | # Add the document
|
---|
133 | $browselist_node->addDocument($doc_obj->get_OID());
|
---|
134 | }
|
---|
135 | # We now recursively move through the document objects child sections,
|
---|
136 | # adding them too. As we do this we build up a contains list for this
|
---|
137 | # document.
|
---|
138 | my $section_ptr = $doc_obj->_lookup_section($section);
|
---|
139 | my @contains = ();
|
---|
140 | if (defined $section_ptr)
|
---|
141 | {
|
---|
142 | foreach my $subsection (@{$section_ptr->{'subsection_order'}}) {
|
---|
143 | &addDocument($collection, $doc_obj, "$section.$subsection");
|
---|
144 | push(@contains, "\".$subsection");
|
---|
145 | }
|
---|
146 | }
|
---|
147 | # Done - clean up
|
---|
148 | }
|
---|
149 | # /** addDocument() **/
|
---|
150 |
|
---|
151 | # /** Sets the metadata attached to a given document. This will update, at most,
|
---|
152 | # * three different locations:
|
---|
153 | # * 1. The Lucene index must be updated. This will involve removing any
|
---|
154 | # * existing value and, if required, adding a new value in its place.
|
---|
155 | # * 2. The GDBM database must be updated. Again any existing value will be
|
---|
156 | # * removed and, if required, a new value added.
|
---|
157 | # * 3. Finally a check against the collect.cfg will be done to determine if
|
---|
158 | # * the changed metadata would have an effect on a classifier and, if so
|
---|
159 | # * the classifier tree will be updated to remove, add or replace any
|
---|
160 | # * tree nodes or node 'contains lists' as necessary.
|
---|
161 | # *
|
---|
162 | # * Pseudo Code:
|
---|
163 | # * ------------
|
---|
164 | # * To add metadata to the document NT1
|
---|
165 | # * A. Establish connection to Lucene
|
---|
166 | # * B. Create a IncrementalDocument object for 'NT1' loading the information
|
---|
167 | # * from the GDBM
|
---|
168 | # * C. Check to see if this metadata is used to build a classifier(s) and if
|
---|
169 | # * so create the appropriate ClassifyTreeModel(s)
|
---|
170 | # * D. If removing or replacing metadata:
|
---|
171 | # * i/ Call ??? to remove key-value from Lucene index
|
---|
172 | # * ii/ Use removeMetadata() to clear value in IncrementalDocument
|
---|
173 | # * iii/ Call removeDocument() in ClassifyTreeModel(s) as necessary
|
---|
174 | # * E. If adding or replacing metadata:
|
---|
175 | # * i/ Call ??? to add key-value from Lucene index
|
---|
176 | # * ii/ Use addMetadata() to add value in IncrementalDocument
|
---|
177 | # * iii/ Call addDocument() in ClassifyTreeModel(s) as necessary
|
---|
178 | # * F. Complete Lucene transaction
|
---|
179 | # * G. Save IncrementalDocument to GDBM
|
---|
180 | # * Note: ClassifyTreeModel automatically updates GDBM as necessary.
|
---|
181 | # *
|
---|
182 | # * @param $collection The name of the collection to update as a string
|
---|
183 | # * @param $oid The unique identifier of a Greenstone document as a
|
---|
184 | # * string
|
---|
185 | # * @param $key The key of the metadata being added as a string
|
---|
186 | # * @param $old_value The value of the metadata being removed/replaced
|
---|
187 | # * or an empty string if adding metadata
|
---|
188 | # * @param $new_value The value of the metadata being added/replacing
|
---|
189 | # * or an empty string if removing metadata
|
---|
190 | # * @param $updateindex 1 to get the index updated. This is used to prevent
|
---|
191 | # * the indexes being changed when doing an incremental
|
---|
192 | # * addition of a new document.
|
---|
193 | # *
|
---|
194 | # * @author John Thompson, DL Consulting Ltd.
|
---|
195 | # */
|
---|
196 | sub setDocumentMetadata()
|
---|
197 | {
|
---|
198 | my ($collection, $oid, $key, $old_value, $new_value, $updateindex) = @_;
|
---|
199 | print STDERR "IncrementalBuildUtils::setDocumentMetadata('$collection','$oid','$key','$old_value','$new_value',$updateindex)\n" unless !$debug;
|
---|
200 | # A. Establish connection to Lucene
|
---|
201 | # This isn't required at the moment, but might be later if we implement
|
---|
202 | # Lucene daemon.
|
---|
203 | # B. Create a IncrementalDocument object for 'NT1' loading the information
|
---|
204 | # from the GDBM
|
---|
205 | print STDERR "* creating incremental document for $oid\n" unless !$debug;
|
---|
206 | my $doc_obj = new IncrementalDocument($collection, $oid);
|
---|
207 | $doc_obj->loadDocument();
|
---|
208 | # C. Check to see if this metadata is used to build a classifier(s) and if
|
---|
209 | # so create the appropriate ClassifyTreeModel(s)
|
---|
210 | print STDERR "* load collection configuration\n" unless !$debug;
|
---|
211 | my $config_obj = &getConfigObj($collection);
|
---|
212 | my $clidx = 1;
|
---|
213 | my @classifier_tree_models = ();
|
---|
214 | foreach my $classifier (@{$config_obj->{'classify'}})
|
---|
215 | {
|
---|
216 | my $index = 0;
|
---|
217 | my $option_count = scalar(@{$classifier});
|
---|
218 | for ($index = 0; $index < $option_count; $index++)
|
---|
219 | {
|
---|
220 | if ($index + 1 < $option_count && @{$classifier}[$index] eq "-metadata" && @{$classifier}[$index + 1] eq $key)
|
---|
221 | {
|
---|
222 | # Create a tree model for this classifier
|
---|
223 | print STDERR "* creating a tree model for classifier: CL$clidx\n" unless !$debug;
|
---|
224 | my $tree_model_obj = new ClassifyTreeModel($collection, "CL" . $clidx);
|
---|
225 | # And store it for later
|
---|
226 | push(@classifier_tree_models, $tree_model_obj);
|
---|
227 | }
|
---|
228 | }
|
---|
229 | $clidx++;
|
---|
230 | }
|
---|
231 | # D. If removing or replacing metadata:
|
---|
232 | if (defined($old_value) && $old_value =~ /[\w\d]+/)
|
---|
233 | {
|
---|
234 | print STDERR "* removing '$key'='$old_value' from GDBM database for document $oid\n" unless !$debug;
|
---|
235 | # i/ Call ??? to remove key-value from Lucene index
|
---|
236 | # Moved elsewhere
|
---|
237 | # ii/ Use removeMetadata() to clear value in IncrementalDocument
|
---|
238 | $doc_obj->removeMetadata($key, $old_value);
|
---|
239 | # iii/ Call removeDocument() in ClassifyTreeModel(s) as necessary
|
---|
240 | foreach my $classifier_tree_model (@classifier_tree_models)
|
---|
241 | {
|
---|
242 | print STDERR "* removing '$old_value' from classifier tree\n" unless !$debug;
|
---|
243 | $classifier_tree_model->removeDocument($old_value, $oid, 1);
|
---|
244 | }
|
---|
245 | }
|
---|
246 | # E. If adding or replacing metadata:
|
---|
247 | if (defined($new_value) && $new_value =~ /[\w\d]+/)
|
---|
248 | {
|
---|
249 | print STDERR "* adding '$key'='$new_value' to GDBM database for document $oid\n" unless !$debug;
|
---|
250 | # i/ Call ??? to add key-value from Lucene index
|
---|
251 | # Moved elsewhere
|
---|
252 | # ii/ Use addMetadata() to add value in IncrementalDocument
|
---|
253 | $doc_obj->addMetadata($key, $new_value);
|
---|
254 | # iii/ Call addDocument() in ClassifyTreeModel(s) as necessary
|
---|
255 | foreach my $classifier_tree_model (@classifier_tree_models)
|
---|
256 | {
|
---|
257 | print STDERR "* adding '$new_value' to classifier tree\n" unless !$debug;
|
---|
258 | $classifier_tree_model->addDocument($new_value, $oid);
|
---|
259 | }
|
---|
260 | }
|
---|
261 | # F. Complete Lucene transaction
|
---|
262 | if(defined($updateindex) && $updateindex)
|
---|
263 | {
|
---|
264 | print STDERR "* updating Lucene indexes\n" unless !$debug;
|
---|
265 | &callGS2LuceneEditor($collection, $doc_obj->getDocNum, $key, $old_value, $new_value);
|
---|
266 | }
|
---|
267 | # G. Save IncrementalDocument to GDBM
|
---|
268 | $doc_obj->saveDocument();
|
---|
269 | $doc_obj = 0;
|
---|
270 | }
|
---|
271 | # /** setDocumentMetadata() **/
|
---|
272 |
|
---|
273 | # /**
|
---|
274 | # *
|
---|
275 | # */
|
---|
276 | sub callGS2LuceneDelete()
|
---|
277 | {
|
---|
278 | my ($collection, $docnum) = @_;
|
---|
279 |
|
---|
280 | # Some path information that is the same for all indexes
|
---|
281 | my $classpath = &util::filename_cat($ENV{'GSDLHOME'},"bin","java","LuceneWrap.jar");
|
---|
282 | my $java_lucene = "org.nzdl.gsdl.LuceneWrap.GS2LuceneDelete";
|
---|
283 | my $indexpath = &util::filename_cat($ENV{'GSDLHOME'},"collect",$collection,"index");
|
---|
284 | # Determine what indexes need to be changed by opening the collections
|
---|
285 | # index path and searching for directories named *idx
|
---|
286 | # If the directory doesn't exist, then there is no built index, and nothing
|
---|
287 | # for us to do.
|
---|
288 | if(opendir(INDEXDIR, $indexpath))
|
---|
289 | {
|
---|
290 | my @index_files = readdir(INDEXDIR);
|
---|
291 | closedir(INDEXDIR);
|
---|
292 | # For each index that matches or pattern, we call the java application
|
---|
293 | # to change the index (as necessary - not every index will include the
|
---|
294 | # document we have been asked to modify)
|
---|
295 | foreach my $actual_index_dir (@index_files)
|
---|
296 | {
|
---|
297 | next unless $actual_index_dir =~ /idx$/;
|
---|
298 | # Determine the path to the index to modify
|
---|
299 | my $full_index_dir = &util::filename_cat($indexpath, $actual_index_dir);
|
---|
300 | # Call java to remove the document
|
---|
301 | my $cmd = "java -classpath \"$classpath\" $java_lucene --index $full_index_dir --nodeid $docnum";
|
---|
302 | print STDERR "CMD: " . $cmd . "\n" unless !$debug;
|
---|
303 | # Run command
|
---|
304 | my $result = `$cmd 2>&1`;
|
---|
305 | print STDERR $result unless !$debug;
|
---|
306 | }
|
---|
307 | }
|
---|
308 | # Done
|
---|
309 | }
|
---|
310 | # /** callGS2LuceneDelete() **/
|
---|
311 |
|
---|
312 | # /**
|
---|
313 | # */
|
---|
314 | sub callGS2LuceneEditor()
|
---|
315 | {
|
---|
316 | my ($collection, $docnum, $key, $old_value, $new_value) = @_;
|
---|
317 |
|
---|
318 | # Some path information that is the same for all indexes
|
---|
319 | my $classpath = &util::filename_cat($ENV{'GSDLHOME'},"collect",$collection,"java","classes");
|
---|
320 | my $jarpath = &util::filename_cat($ENV{'GSDLHOME'},"bin","java","LuceneWrap.jar");
|
---|
321 | my $java_lucene = "org.nzdl.gsdl.LuceneWrap.GS2LuceneEditor";
|
---|
322 | my $indexpath = &util::filename_cat($ENV{'GSDLHOME'},"collect",$collection,"index");
|
---|
323 | # And some commands that don't change
|
---|
324 | my $java_args = "";
|
---|
325 | # Append the node id
|
---|
326 | $java_args .= "--nodeid $docnum ";
|
---|
327 | # We have to convert the given metadata key into its two letter field code.
|
---|
328 | # We do this by looking in the build.cfg file.
|
---|
329 | my $field = &getFieldFromBuildCFG($indexpath, $key);
|
---|
330 | # The metadata field to change
|
---|
331 | $java_args .= "--field $field ";
|
---|
332 | # And the old and new values as necessary
|
---|
333 | if(defined($old_value) && $old_value =~ /[\w\d]+/)
|
---|
334 | {
|
---|
335 | $java_args .= "--oldvalue \"$old_value\" ";
|
---|
336 | }
|
---|
337 | if(defined($new_value) && $new_value =~ /[\w\d]+/)
|
---|
338 | {
|
---|
339 | $java_args .= "--newvalue \"$new_value\" ";
|
---|
340 | }
|
---|
341 | # Determine what indexes need to be changed by opening the collections
|
---|
342 | # index path and searching for directories named *idx
|
---|
343 | # If the directory doesn't exist, then there is no built index, and nothing
|
---|
344 | # for us to do.
|
---|
345 | # We also check if the field is something other than "". It is entirely
|
---|
346 | # possible that we have been asked to update a metadata field that isn't
|
---|
347 | # part of any index, so this is where we break out of editing the index if
|
---|
348 | # we have
|
---|
349 | if($field =~ /^\w\w$/ && opendir(INDEXDIR, $indexpath))
|
---|
350 | {
|
---|
351 | my @index_files = readdir(INDEXDIR);
|
---|
352 | closedir(INDEXDIR);
|
---|
353 | # For each index that matches or pattern, we call the java application
|
---|
354 | # to change the index (as necessary - not every index will include the
|
---|
355 | # document we have been asked to modify)
|
---|
356 | foreach my $actual_index_dir (@index_files)
|
---|
357 | {
|
---|
358 | next unless $actual_index_dir =~ /idx$/;
|
---|
359 | # Determine the path to the index to modify
|
---|
360 | my $full_index_dir = &util::filename_cat($indexpath, $actual_index_dir);
|
---|
361 | # And prepend to the command java arguments
|
---|
362 | my $cur_java_args = "--index $full_index_dir " . $java_args;
|
---|
363 | print STDERR "CMD: java -classpath \"$classpath:$jarpath\" $java_lucene $cur_java_args 2>&1\n" unless !$debug;
|
---|
364 | # Run command
|
---|
365 | my $result = `java -classpath \"$classpath:$jarpath\" $java_lucene $cur_java_args 2>&1`;
|
---|
366 | print STDERR $result unless !$debug;
|
---|
367 | }
|
---|
368 | }
|
---|
369 | # Done
|
---|
370 | }
|
---|
371 | # /** callGS2LuceneEditor() **/
|
---|
372 |
|
---|
373 | ## Remove a document from the GDBM and Index.
|
---|
374 | #
|
---|
375 | # @param collection The collection to alter
|
---|
376 | # @param oid The unique identifier of the document to be removed
|
---|
377 | ##
|
---|
378 | sub deleteDocument()
|
---|
379 | {
|
---|
380 | my ($collection, $oid) = @_;
|
---|
381 | # Load the incremental document to go with this oid, as we need some
|
---|
382 | # information from it.
|
---|
383 | my $doc_obj = new IncrementalDocument($collection, $oid);
|
---|
384 | $doc_obj->loadDocument();
|
---|
385 | # Check if this object even exists by retrieving the docnum.
|
---|
386 | my $doc_num = $doc_obj->getDocNum();
|
---|
387 | print STDERR "Removing document docnum: $doc_num\n" unless !$debug;
|
---|
388 | if ($doc_num > -1)
|
---|
389 | {
|
---|
390 | # Now use the GDBM utils to write a blank string to this oid in the
|
---|
391 | # database
|
---|
392 | &GDBMUtils::gdbmCachedCollectionSet($collection, $oid, "");
|
---|
393 | # Remove reverse lookup
|
---|
394 | &GDBMUtils::gdbmCachedCollectionSet($collection, $doc_num, "");
|
---|
395 | # And remove from the database
|
---|
396 | &callGS2LuceneDelete($collection, $doc_num);
|
---|
397 |
|
---|
398 | # Regenerate the classifier trees.
|
---|
399 | print STDERR "* load collection configuration\n";# unless !$debug;
|
---|
400 | my $config_obj = &getConfigObj($collection);
|
---|
401 | my $clidx = 1;
|
---|
402 | my %classifier_tree_models = ();
|
---|
403 | foreach my $classifier (@{$config_obj->{'classify'}})
|
---|
404 | {
|
---|
405 | my $index = 0;
|
---|
406 | my $option_count = scalar(@{$classifier});
|
---|
407 | for ($index = 0; $index < $option_count; $index++)
|
---|
408 | {
|
---|
409 | if ($index + 1 < $option_count && @{$classifier}[$index] eq "-metadata")
|
---|
410 | {
|
---|
411 | my $key = @{$classifier}[$index + 1];
|
---|
412 | # Create a tree model for this classifier
|
---|
413 | print STDERR "* creating a tree model for classifier: CL" . $clidx . " [" . $key . "]\n";# unless !$debug;
|
---|
414 | my $tree_model_obj = new ClassifyTreeModel($collection, "CL" . $clidx);
|
---|
415 | # And store it against its key for later
|
---|
416 | $classifier_tree_models{$key} = $tree_model_obj;
|
---|
417 | }
|
---|
418 | }
|
---|
419 | $clidx++;
|
---|
420 | }
|
---|
421 |
|
---|
422 | # For each piece of metadata assigned to this document, if there is a
|
---|
423 | # matching classifier tree, remove the path from the tree.
|
---|
424 | print STDERR "* searching for classifier paths to be removed\n";
|
---|
425 |
|
---|
426 | my $metadata = $doc_obj->getAllMetadata();
|
---|
427 | foreach my $pair (@$metadata)
|
---|
428 | {
|
---|
429 | my ($key, $value) = @$pair;
|
---|
430 | print STDERR "* testing " . $key . "=>" . $value . "\n";
|
---|
431 | if (defined($classifier_tree_models{$key}))
|
---|
432 | {
|
---|
433 | my $model = $classifier_tree_models{$key};
|
---|
434 | print STDERR "* removing '" . $value . "' from classifier " . $model->getRootNode()->getCLID() . "\n";
|
---|
435 | $model->removeDocument($value, $oid, 1);
|
---|
436 | }
|
---|
437 | }
|
---|
438 |
|
---|
439 | # We also have to remove from browselist - the reverse process of
|
---|
440 | # adding to browselist shown above.
|
---|
441 | my $dummy_model = new ClassifyTreeModel($collection, "");
|
---|
442 | my $browselist_node = new ClassifyTreeNode($dummy_model, "browselist");
|
---|
443 | # Add the document
|
---|
444 | $browselist_node->removeDocument($oid);
|
---|
445 | # Clean up
|
---|
446 | }
|
---|
447 | # else, no document, no need to delete.
|
---|
448 | }
|
---|
449 | ## deleteDocument() ##
|
---|
450 |
|
---|
451 | # /**
|
---|
452 | # */
|
---|
453 | sub getFieldFromBuildCFG()
|
---|
454 | {
|
---|
455 | my ($indexpath, $key) = @_;
|
---|
456 | my $field = "";
|
---|
457 | my $build_cfg = &util::filename_cat($indexpath, "build.cfg");
|
---|
458 | # If there isn't a build.cfg then the index hasn't been built and there is
|
---|
459 | # nothing to do
|
---|
460 | if(open(BUILDCFG, $build_cfg))
|
---|
461 | {
|
---|
462 | # For each line of the build configuration
|
---|
463 | my $line;
|
---|
464 | while($line = <BUILDCFG>)
|
---|
465 | {
|
---|
466 | # Only interested in the indexfieldmap line
|
---|
467 | if($line =~ /^indexfieldmap\s+/)
|
---|
468 | {
|
---|
469 | # Extract the field information by looking up the key pair
|
---|
470 | if($line =~ /\s$key->(\w\w)/)
|
---|
471 | {
|
---|
472 | $field = $1;
|
---|
473 | }
|
---|
474 | }
|
---|
475 | }
|
---|
476 | # Done with file
|
---|
477 | close(BUILDCFG);
|
---|
478 | }
|
---|
479 | # Return whatever we found
|
---|
480 | return $field;
|
---|
481 | }
|
---|
482 | # /** getFieldFromBuildCFG() **/
|
---|
483 |
|
---|
484 |
|
---|
485 |
|
---|
486 |
|
---|
487 |
|
---|
488 | # /** Retrieve an object (associative array) containing information about the
|
---|
489 | # * collection configuration.
|
---|
490 | # * @param $collection The shortname of the collection as a string
|
---|
491 | # * @return An associative array containing information from the collect.cfg
|
---|
492 | # * @author John Thompson, DL Consulting Ltd.
|
---|
493 | # */
|
---|
494 | sub getConfigObj()
|
---|
495 | {
|
---|
496 | my ($collection) = @_;
|
---|
497 |
|
---|
498 | #rint STDERR "getConfigObj()\n" unless !$debug;
|
---|
499 |
|
---|
500 | my $colcfgname = &util::filename_cat($ENV{'GSDLHOME'}, "collect", $collection, "etc", "collect.cfg");
|
---|
501 | if (!-e $colcfgname)
|
---|
502 | {
|
---|
503 | die "IncrementalBuildUtils - couldn't find collect.cfg for collection $collection\n";
|
---|
504 | }
|
---|
505 | return &colcfg::read_collect_cfg ($colcfgname);
|
---|
506 | }
|
---|
507 | # /** getConfigObj() **/
|
---|
508 |
|
---|
509 | 1;
|
---|