| 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 John Rowe
|
|---|
| 34 | # *
|
|---|
| 35 | # * @author John Thompson, DL Consulting Ltd.
|
|---|
| 36 | # * @author John Rowe, DL Consulting Ltd.
|
|---|
| 37 | # */
|
|---|
| 38 | ###########################################################################
|
|---|
| 39 | package IncrementalBuildUtils;
|
|---|
| 40 |
|
|---|
| 41 | BEGIN {
|
|---|
| 42 | die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
|
|---|
| 43 | die "GSDLOS not set\n" unless defined $ENV{'GSDLOS'};
|
|---|
| 44 |
|
|---|
| 45 | my $current_library_paths = join(":",@INC);
|
|---|
| 46 | my $gsdl_perllib_path = $ENV{'GSDLHOME'} . '/perllib';
|
|---|
| 47 | if ($current_library_paths !~ /$gsdl_perllib_path/)
|
|---|
| 48 | {
|
|---|
| 49 | print "IncrementalBuildUtils setting up environment!\n";
|
|---|
| 50 | unshift (@INC, $gsdl_perllib_path); # [jmt12]
|
|---|
| 51 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/cpan"); # [jmt12]
|
|---|
| 52 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/plugins"); # [jmt12]
|
|---|
| 53 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/classify"); # [jmt12]
|
|---|
| 54 | }
|
|---|
| 55 | }
|
|---|
| 56 |
|
|---|
| 57 | use doc;
|
|---|
| 58 | use cfgread;
|
|---|
| 59 | use colcfg;
|
|---|
| 60 | use strict;
|
|---|
| 61 | use util;
|
|---|
| 62 |
|
|---|
| 63 | use ClassifyTreeModel;
|
|---|
| 64 | use IncrementalDocument;
|
|---|
| 65 |
|
|---|
| 66 | # Change debugging to 1 if you want verbose debugging output
|
|---|
| 67 | my $debug = 1;
|
|---|
| 68 |
|
|---|
| 69 | # Ensure the collection specific binaries are on the search path
|
|---|
| 70 | my $path_separator = ":";
|
|---|
| 71 | if($ENV{'GSDLOS'} =~ /win/) {
|
|---|
| 72 | $path_separator = ";";
|
|---|
| 73 | }
|
|---|
| 74 | my $os_binary_path = &util::filename_cat($ENV{'GSDLHOME'}, "bin", $ENV{'GSDLOS'});
|
|---|
| 75 | if ($ENV{'PATH'} !~ /$os_binary_path/)
|
|---|
| 76 | {
|
|---|
| 77 | $ENV{'PATH'} = $os_binary_path . $path_separator . $ENV{'PATH'};
|
|---|
| 78 | }
|
|---|
| 79 | my $script_path = &util::filename_cat($ENV{'GSDLHOME'}, "bin", "script");
|
|---|
| 80 | if ($ENV{'PATH'} !~ /$script_path/)
|
|---|
| 81 | {
|
|---|
| 82 | $ENV{'PATH'} = $script_path . $path_separator . $ENV{'PATH'};
|
|---|
| 83 | }
|
|---|
| 84 |
|
|---|
| 85 | # /**
|
|---|
| 86 | # */
|
|---|
| 87 | sub addDocument()
|
|---|
| 88 | {
|
|---|
| 89 | my ($collection, $infodbtype, $doc_obj, $section, $updateindex) = @_;
|
|---|
| 90 |
|
|---|
| 91 | $updateindex = 0 unless defined($updateindex);
|
|---|
| 92 |
|
|---|
| 93 | print STDERR "IncrementalBuildUtils::addDocument('$collection',$infodbtype,$doc_obj,'$section')\n" unless !$debug;
|
|---|
| 94 | # Gonna need to know in several places whether this is the top section
|
|---|
| 95 | # of the document or not
|
|---|
| 96 | my $is_top = ($section eq $doc_obj->get_top_section());
|
|---|
| 97 |
|
|---|
| 98 | # Retrieve all of the metadata from this document object only - not any
|
|---|
| 99 | # child documents
|
|---|
| 100 | my $metadata = $doc_obj->get_all_metadata($section);
|
|---|
| 101 | # Check and add the docnum first
|
|---|
| 102 | my $found_docnum = 0;
|
|---|
| 103 | foreach my $pair (@$metadata)
|
|---|
| 104 | {
|
|---|
| 105 | my ($key, $value) = (@$pair);
|
|---|
| 106 | if ($key eq "docnum")
|
|---|
| 107 | {
|
|---|
| 108 | &setDocumentMetadata($collection, $infodbtype, $doc_obj->get_OID() . "$section", $key, "", $value, $updateindex);
|
|---|
| 109 | $found_docnum = 1;
|
|---|
| 110 | }
|
|---|
| 111 | }
|
|---|
| 112 |
|
|---|
| 113 | if (!$found_docnum)
|
|---|
| 114 | {
|
|---|
| 115 | die("Fatal Error! Tried to add document without providing docnum");
|
|---|
| 116 | }
|
|---|
| 117 |
|
|---|
| 118 | # Add it piece by piece - this depends on the loading of a blank document
|
|---|
| 119 | # working the way it should.
|
|---|
| 120 | foreach my $pair (@$metadata)
|
|---|
| 121 | {
|
|---|
| 122 | my ($key, $value) = (@$pair);
|
|---|
| 123 | if ($key ne "Identifier" && $key ne "docnum" && $key !~ /^gsdl/ && defined $value && $value ne "")
|
|---|
| 124 | {
|
|---|
| 125 | # escape problematic stuff
|
|---|
| 126 | $value =~ s/\\/\\\\/g;
|
|---|
| 127 | $value =~ s/\n/\\n/g;
|
|---|
| 128 | $value =~ s/\r/\\r/g;
|
|---|
| 129 | if ($value =~ /-{70,}/)
|
|---|
| 130 | {
|
|---|
| 131 | # if value contains 70 or more hyphens in a row we need
|
|---|
| 132 | # to escape them to prevent txt2db from treating them
|
|---|
| 133 | # as a separator
|
|---|
| 134 | $value =~ s/-/&\#045;/gi;
|
|---|
| 135 | }
|
|---|
| 136 | # Go ahead and set the metadata
|
|---|
| 137 | &setDocumentMetadata($collection, $infodbtype, $doc_obj->get_OID() . "$section", $key, "", $value, $updateindex);
|
|---|
| 138 | }
|
|---|
| 139 | }
|
|---|
| 140 | # We now have to load the browselist node too. We create a ClassifyTreeNode
|
|---|
| 141 | # based on a dummy model.
|
|---|
| 142 | # Note: only if section is the top section
|
|---|
| 143 | if ($is_top)
|
|---|
| 144 | {
|
|---|
| 145 | my $dummy_model = new ClassifyTreeModel($collection, $infodbtype, "");
|
|---|
| 146 | my $browselist_node = new ClassifyTreeNode($dummy_model, "browselist");
|
|---|
| 147 | # Add the document
|
|---|
| 148 | $browselist_node->addDocument($doc_obj->get_OID());
|
|---|
| 149 | }
|
|---|
| 150 | # We now recursively move through the document objects child sections,
|
|---|
| 151 | # adding them too. As we do this we build up a contains list for this
|
|---|
| 152 | # document.
|
|---|
| 153 | my $section_ptr = $doc_obj->_lookup_section($section);
|
|---|
| 154 | my @contains = ();
|
|---|
| 155 | if (defined $section_ptr)
|
|---|
| 156 | {
|
|---|
| 157 | foreach my $subsection (@{$section_ptr->{'subsection_order'}}) {
|
|---|
| 158 | &addDocument($collection, $infodbtype, $doc_obj, "$section.$subsection");
|
|---|
| 159 | push(@contains, "\".$subsection");
|
|---|
| 160 | }
|
|---|
| 161 | }
|
|---|
| 162 | # Done - clean up
|
|---|
| 163 | }
|
|---|
| 164 | # /** addDocument() **/
|
|---|
| 165 |
|
|---|
| 166 | # /** Sets the metadata attached to a given document. This will update, at most,
|
|---|
| 167 | # * three different locations:
|
|---|
| 168 | # * 1. The Lucene index must be updated. This will involve removing any
|
|---|
| 169 | # * existing value and, if required, adding a new value in its place.
|
|---|
| 170 | # * 2. The info database must be updated. Again any existing value will be
|
|---|
| 171 | # * removed and, if required, a new value added.
|
|---|
| 172 | # * 3. Finally a check against the collect.cfg will be done to determine if
|
|---|
| 173 | # * the changed metadata would have an effect on a classifier and, if so
|
|---|
| 174 | # * the classifier tree will be updated to remove, add or replace any
|
|---|
| 175 | # * tree nodes or node 'contains lists' as necessary.
|
|---|
| 176 | # *
|
|---|
| 177 | # * Pseudo Code:
|
|---|
| 178 | # * ------------
|
|---|
| 179 | # * To add metadata to the document NT1
|
|---|
| 180 | # * A. Establish connection to Lucene
|
|---|
| 181 | # * B. Create a IncrementalDocument object for 'NT1' loading the information
|
|---|
| 182 | # * from the info database
|
|---|
| 183 | # * C. Check to see if this metadata is used to build a classifier(s) and if
|
|---|
| 184 | # * so create the appropriate ClassifyTreeModel(s)
|
|---|
| 185 | # * D. If removing or replacing metadata:
|
|---|
| 186 | # * i/ Call ??? to remove key-value from Lucene index
|
|---|
| 187 | # * ii/ Use removeMetadata() to clear value in IncrementalDocument
|
|---|
| 188 | # * iii/ Call removeDocument() in ClassifyTreeModel(s) as necessary
|
|---|
| 189 | # * E. If adding or replacing metadata:
|
|---|
| 190 | # * i/ Call ??? to add key-value from Lucene index
|
|---|
| 191 | # * ii/ Use addMetadata() to add value in IncrementalDocument
|
|---|
| 192 | # * iii/ Call addDocument() in ClassifyTreeModel(s) as necessary
|
|---|
| 193 | # * F. Complete Lucene transaction
|
|---|
| 194 | # * G. Save IncrementalDocument to info database
|
|---|
| 195 | # * Note: ClassifyTreeModel automatically updates the info database as necessary.
|
|---|
| 196 | # *
|
|---|
| 197 | # * @param $collection The name of the collection to update as a string
|
|---|
| 198 | # * @param $oid The unique identifier of a Greenstone document as a
|
|---|
| 199 | # * string
|
|---|
| 200 | # * @param $key The key of the metadata being added as a string
|
|---|
| 201 | # * @param $old_value The value of the metadata being removed/replaced
|
|---|
| 202 | # * or an empty string if adding metadata
|
|---|
| 203 | # * @param $new_value The value of the metadata being added/replacing
|
|---|
| 204 | # * or an empty string if removing metadata
|
|---|
| 205 | # * @param $updateindex 1 to get the index updated. This is used to prevent
|
|---|
| 206 | # * the indexes being changed when doing an incremental
|
|---|
| 207 | # * addition of a new document.
|
|---|
| 208 | # *
|
|---|
| 209 | # * @author John Thompson, DL Consulting Ltd.
|
|---|
| 210 | # */
|
|---|
| 211 | sub setDocumentMetadata()
|
|---|
| 212 | {
|
|---|
| 213 | my ($collection, $infodbtype, $oid, $key, $old_value, $new_value, $updateindex) = @_;
|
|---|
| 214 | print STDERR "IncrementalBuildUtils::setDocumentMetadata('$collection',$infodbtype,'$oid','$key','$old_value','$new_value',$updateindex)\n" unless !$debug;
|
|---|
| 215 | # A. Establish connection to Lucene
|
|---|
| 216 | # This isn't required at the moment, but might be later if we implement
|
|---|
| 217 | # Lucene daemon.
|
|---|
| 218 | # B. Create a IncrementalDocument object for 'NT1' loading the information
|
|---|
| 219 | # from the info database
|
|---|
| 220 | print STDERR "* creating incremental document for $oid\n" unless !$debug;
|
|---|
| 221 | my $doc_obj = new IncrementalDocument($collection, $infodbtype, $oid);
|
|---|
| 222 | $doc_obj->loadDocument();
|
|---|
| 223 | # C. Check to see if this metadata is used to build a classifier(s) and if
|
|---|
| 224 | # so create the appropriate ClassifyTreeModel(s)
|
|---|
| 225 | print STDERR "* load collection configuration\n" unless !$debug;
|
|---|
| 226 | my $config_obj = &getConfigObj($collection);
|
|---|
| 227 | my $clidx = 1;
|
|---|
| 228 | my @classifier_tree_models = ();
|
|---|
| 229 | foreach my $classifier (@{$config_obj->{'classify'}})
|
|---|
| 230 | {
|
|---|
| 231 | my $index = 0;
|
|---|
| 232 | my $option_count = scalar(@{$classifier});
|
|---|
| 233 | for ($index = 0; $index < $option_count; $index++)
|
|---|
| 234 | {
|
|---|
| 235 | if ($index + 1 < $option_count && @{$classifier}[$index] eq "-metadata" && @{$classifier}[$index + 1] eq $key)
|
|---|
| 236 | {
|
|---|
| 237 | # Create a tree model for this classifier
|
|---|
| 238 | print STDERR "* creating a tree model for classifier: CL$clidx\n" unless !$debug;
|
|---|
| 239 | my $tree_model_obj = new ClassifyTreeModel($collection, $infodbtype, "CL" . $clidx);
|
|---|
| 240 | # And store it for later
|
|---|
| 241 | push(@classifier_tree_models, $tree_model_obj);
|
|---|
| 242 | }
|
|---|
| 243 | }
|
|---|
| 244 | $clidx++;
|
|---|
| 245 | }
|
|---|
| 246 | # D. If removing or replacing metadata:
|
|---|
| 247 | if (defined($old_value) && $old_value =~ /[\w\d]+/)
|
|---|
| 248 | {
|
|---|
| 249 | print STDERR "* removing '$key'='$old_value' from info database for document $oid\n" unless !$debug;
|
|---|
| 250 | # i/ Call ??? to remove key-value from Lucene index
|
|---|
| 251 | # Moved elsewhere
|
|---|
| 252 | # ii/ Use removeMetadata() to clear value in IncrementalDocument
|
|---|
| 253 | $doc_obj->removeMetadata($key, $old_value);
|
|---|
| 254 | # iii/ Call removeDocument() in ClassifyTreeModel(s) as necessary
|
|---|
| 255 | foreach my $classifier_tree_model (@classifier_tree_models)
|
|---|
| 256 | {
|
|---|
| 257 | print STDERR "* removing '$old_value' from classifier tree\n" unless !$debug;
|
|---|
| 258 | $classifier_tree_model->removeDocument($old_value, $oid, 1);
|
|---|
| 259 | }
|
|---|
| 260 | }
|
|---|
| 261 | # E. If adding or replacing metadata:
|
|---|
| 262 | if (defined($new_value) && $new_value =~ /[\w\d]+/)
|
|---|
| 263 | {
|
|---|
| 264 | print STDERR "* adding '$key'='$new_value' to info database for document $oid\n" unless !$debug;
|
|---|
| 265 | # i/ Call ??? to add key-value from Lucene index
|
|---|
| 266 | # Moved elsewhere
|
|---|
| 267 | # ii/ Use addMetadata() to add value in IncrementalDocument
|
|---|
| 268 | $doc_obj->addMetadata($key, $new_value);
|
|---|
| 269 | # iii/ Call addDocument() in ClassifyTreeModel(s) as necessary
|
|---|
| 270 | foreach my $classifier_tree_model (@classifier_tree_models)
|
|---|
| 271 | {
|
|---|
| 272 | print STDERR "* adding '$new_value' to classifier tree\n" unless !$debug;
|
|---|
| 273 | $classifier_tree_model->addDocument($new_value, $oid);
|
|---|
| 274 | }
|
|---|
| 275 | }
|
|---|
| 276 | # F. Complete Lucene transaction
|
|---|
| 277 | if(defined($updateindex) && $updateindex)
|
|---|
| 278 | {
|
|---|
| 279 | print STDERR "* updating Lucene indexes\n" unless !$debug;
|
|---|
| 280 | &callGS2LuceneEditor($collection, $doc_obj->getDocNum, $key, $old_value, $new_value);
|
|---|
| 281 | }
|
|---|
| 282 | # G. Save IncrementalDocument to info database
|
|---|
| 283 | $doc_obj->saveDocument();
|
|---|
| 284 | $doc_obj = 0;
|
|---|
| 285 | }
|
|---|
| 286 | # /** setDocumentMetadata() **/
|
|---|
| 287 |
|
|---|
| 288 | # /**
|
|---|
| 289 | # *
|
|---|
| 290 | # */
|
|---|
| 291 | sub callGS2LuceneDelete()
|
|---|
| 292 | {
|
|---|
| 293 | my ($collection, $docnum) = @_;
|
|---|
| 294 |
|
|---|
| 295 | # Some path information that is the same for all indexes
|
|---|
| 296 | my $classpath = &util::filename_cat($ENV{'GSDLHOME'},"bin","java","LuceneWrap.jar");
|
|---|
| 297 | my $java_lucene = "org.nzdl.gsdl.LuceneWrap.GS2LuceneDelete";
|
|---|
| 298 | my $indexpath = &util::filename_cat($ENV{'GSDLHOME'},"collect",$collection,"index");
|
|---|
| 299 | # Determine what indexes need to be changed by opening the collections
|
|---|
| 300 | # index path and searching for directories named *idx
|
|---|
| 301 | # If the directory doesn't exist, then there is no built index, and nothing
|
|---|
| 302 | # for us to do.
|
|---|
| 303 | if(opendir(INDEXDIR, $indexpath))
|
|---|
| 304 | {
|
|---|
| 305 | my @index_files = readdir(INDEXDIR);
|
|---|
| 306 | closedir(INDEXDIR);
|
|---|
| 307 | # For each index that matches or pattern, we call the java application
|
|---|
| 308 | # to change the index (as necessary - not every index will include the
|
|---|
| 309 | # document we have been asked to modify)
|
|---|
| 310 | foreach my $actual_index_dir (@index_files)
|
|---|
| 311 | {
|
|---|
| 312 | next unless $actual_index_dir =~ /idx$/;
|
|---|
| 313 | # Determine the path to the index to modify
|
|---|
| 314 | my $full_index_dir = &util::filename_cat($indexpath, $actual_index_dir);
|
|---|
| 315 | # Call java to remove the document
|
|---|
| 316 | my $cmd = "java -classpath \"$classpath\" $java_lucene --index $full_index_dir --nodeid $docnum";
|
|---|
| 317 | print STDERR "CMD: " . $cmd . "\n" unless !$debug;
|
|---|
| 318 | # Run command
|
|---|
| 319 | my $result = `$cmd 2>&1`;
|
|---|
| 320 | print STDERR $result unless !$debug;
|
|---|
| 321 | }
|
|---|
| 322 | }
|
|---|
| 323 | # Done
|
|---|
| 324 | }
|
|---|
| 325 | # /** callGS2LuceneDelete() **/
|
|---|
| 326 |
|
|---|
| 327 | # /**
|
|---|
| 328 | # */
|
|---|
| 329 | sub callGS2LuceneEditor()
|
|---|
| 330 | {
|
|---|
| 331 | my ($collection, $docnum, $key, $old_value, $new_value) = @_;
|
|---|
| 332 |
|
|---|
| 333 | # Some path information that is the same for all indexes
|
|---|
| 334 | my $classpath = &util::filename_cat($ENV{'GSDLHOME'},"collect",$collection,"java","classes");
|
|---|
| 335 | my $jarpath = &util::filename_cat($ENV{'GSDLHOME'},"bin","java","LuceneWrap.jar");
|
|---|
| 336 | my $java_lucene = "org.nzdl.gsdl.LuceneWrap.GS2LuceneEditor";
|
|---|
| 337 | my $indexpath = &util::filename_cat($ENV{'GSDLHOME'},"collect",$collection,"index");
|
|---|
| 338 | # And some commands that don't change
|
|---|
| 339 | my $java_args = "";
|
|---|
| 340 | # Append the node id
|
|---|
| 341 | $java_args .= "--nodeid $docnum ";
|
|---|
| 342 | # We have to convert the given metadata key into its two letter field code.
|
|---|
| 343 | # We do this by looking in the build.cfg file.
|
|---|
| 344 | my $field = &getFieldFromBuildCFG($indexpath, $key);
|
|---|
| 345 | # The metadata field to change
|
|---|
| 346 | $java_args .= "--field $field ";
|
|---|
| 347 | # And the old and new values as necessary
|
|---|
| 348 | if(defined($old_value) && $old_value =~ /[\w\d]+/)
|
|---|
| 349 | {
|
|---|
| 350 | $java_args .= "--oldvalue \"$old_value\" ";
|
|---|
| 351 | }
|
|---|
| 352 | if(defined($new_value) && $new_value =~ /[\w\d]+/)
|
|---|
| 353 | {
|
|---|
| 354 | $java_args .= "--newvalue \"$new_value\" ";
|
|---|
| 355 | }
|
|---|
| 356 | # Determine what indexes need to be changed by opening the collections
|
|---|
| 357 | # index path and searching for directories named *idx
|
|---|
| 358 | # If the directory doesn't exist, then there is no built index, and nothing
|
|---|
| 359 | # for us to do.
|
|---|
| 360 | # We also check if the field is something other than "". It is entirely
|
|---|
| 361 | # possible that we have been asked to update a metadata field that isn't
|
|---|
| 362 | # part of any index, so this is where we break out of editing the index if
|
|---|
| 363 | # we have
|
|---|
| 364 | if($field =~ /^\w\w$/ && opendir(INDEXDIR, $indexpath))
|
|---|
| 365 | {
|
|---|
| 366 | my @index_files = readdir(INDEXDIR);
|
|---|
| 367 | closedir(INDEXDIR);
|
|---|
| 368 | # For each index that matches or pattern, we call the java application
|
|---|
| 369 | # to change the index (as necessary - not every index will include the
|
|---|
| 370 | # document we have been asked to modify)
|
|---|
| 371 | foreach my $actual_index_dir (@index_files)
|
|---|
| 372 | {
|
|---|
| 373 | next unless $actual_index_dir =~ /idx$/;
|
|---|
| 374 | # Determine the path to the index to modify
|
|---|
| 375 | my $full_index_dir = &util::filename_cat($indexpath, $actual_index_dir);
|
|---|
| 376 | # And prepend to the command java arguments
|
|---|
| 377 | my $cur_java_args = "--index $full_index_dir " . $java_args;
|
|---|
| 378 | print STDERR "CMD: java -classpath \"$classpath:$jarpath\" $java_lucene $cur_java_args 2>&1\n" unless !$debug;
|
|---|
| 379 | # Run command
|
|---|
| 380 | my $result = `java -classpath \"$classpath:$jarpath\" $java_lucene $cur_java_args 2>&1`;
|
|---|
| 381 | print STDERR $result unless !$debug;
|
|---|
| 382 | }
|
|---|
| 383 | }
|
|---|
| 384 | # Done
|
|---|
| 385 | }
|
|---|
| 386 | # /** callGS2LuceneEditor() **/
|
|---|
| 387 |
|
|---|
| 388 | ## Remove a document from the info database and Index.
|
|---|
| 389 | #
|
|---|
| 390 | # @param collection The collection to alter
|
|---|
| 391 | # @param oid The unique identifier of the document to be removed
|
|---|
| 392 | ##
|
|---|
| 393 | sub deleteDocument()
|
|---|
| 394 | {
|
|---|
| 395 | my ($collection, $infodbtype, $oid) = @_;
|
|---|
| 396 | # Load the incremental document to go with this oid, as we need some
|
|---|
| 397 | # information from it.
|
|---|
| 398 | my $doc_obj = new IncrementalDocument($collection, $infodbtype, $oid);
|
|---|
| 399 | $doc_obj->loadDocument();
|
|---|
| 400 | # Check if this object even exists by retrieving the docnum.
|
|---|
| 401 | my $doc_num = $doc_obj->getDocNum();
|
|---|
| 402 | print STDERR "Removing document docnum: $doc_num\n" unless !$debug;
|
|---|
| 403 | if ($doc_num > -1)
|
|---|
| 404 | {
|
|---|
| 405 | # Now write a blank string to this oid in the info database
|
|---|
| 406 | my $index_text_directory_path = &util::filename_cat($ENV{'GSDLHOME'}, "collect", $collection, "index", "text");
|
|---|
| 407 | my $infodb_file_path = &dbutil::get_infodb_file_path($infodbtype, $collection, $index_text_directory_path);
|
|---|
| 408 | my $infodb_file_handle = &dbutil::open_infodb_write_handle($infodbtype, $infodb_file_path, "append");
|
|---|
| 409 | &dbutil::write_infodb_entry($infodbtype, $infodb_file_handle, $oid, &dbutil::convert_infodb_string_to_hash(""));
|
|---|
| 410 | # Remove reverse lookup
|
|---|
| 411 | &dbutil::write_infodb_entry($infodbtype, $infodb_file_handle, $doc_num, &dbutil::convert_infodb_string_to_hash(""));
|
|---|
| 412 | &dbutil::close_infodb_write_handle($infodbtype, $infodb_file_handle);
|
|---|
| 413 |
|
|---|
| 414 | # And remove from the database
|
|---|
| 415 | &callGS2LuceneDelete($collection, $doc_num);
|
|---|
| 416 |
|
|---|
| 417 | # Regenerate the classifier trees.
|
|---|
| 418 | print STDERR "* load collection configuration\n";# unless !$debug;
|
|---|
| 419 | my $config_obj = &getConfigObj($collection);
|
|---|
| 420 | my $clidx = 1;
|
|---|
| 421 | my %classifier_tree_models = ();
|
|---|
| 422 | foreach my $classifier (@{$config_obj->{'classify'}})
|
|---|
| 423 | {
|
|---|
| 424 | my $index = 0;
|
|---|
| 425 | my $option_count = scalar(@{$classifier});
|
|---|
| 426 | for ($index = 0; $index < $option_count; $index++)
|
|---|
| 427 | {
|
|---|
| 428 | if ($index + 1 < $option_count && @{$classifier}[$index] eq "-metadata")
|
|---|
| 429 | {
|
|---|
| 430 | my $key = @{$classifier}[$index + 1];
|
|---|
| 431 | # Create a tree model for this classifier
|
|---|
| 432 | print STDERR "* creating a tree model for classifier: CL" . $clidx . " [" . $key . "]\n";# unless !$debug;
|
|---|
| 433 | my $tree_model_obj = new ClassifyTreeModel($collection, $infodbtype, "CL" . $clidx);
|
|---|
| 434 | # And store it against its key for later
|
|---|
| 435 | $classifier_tree_models{$key} = $tree_model_obj;
|
|---|
| 436 | }
|
|---|
| 437 | }
|
|---|
| 438 | $clidx++;
|
|---|
| 439 | }
|
|---|
| 440 |
|
|---|
| 441 | # For each piece of metadata assigned to this document, if there is a
|
|---|
| 442 | # matching classifier tree, remove the path from the tree.
|
|---|
| 443 | print STDERR "* searching for classifier paths to be removed\n";
|
|---|
| 444 |
|
|---|
| 445 | my $metadata = $doc_obj->getAllMetadata();
|
|---|
| 446 | foreach my $pair (@$metadata)
|
|---|
| 447 | {
|
|---|
| 448 | my ($key, $value) = @$pair;
|
|---|
| 449 | print STDERR "* testing " . $key . "=>" . $value . "\n";
|
|---|
| 450 | if (defined($classifier_tree_models{$key}))
|
|---|
| 451 | {
|
|---|
| 452 | my $model = $classifier_tree_models{$key};
|
|---|
| 453 | print STDERR "* removing '" . $value . "' from classifier " . $model->getRootNode()->getCLID() . "\n";
|
|---|
| 454 | $model->removeDocument($value, $oid, 1);
|
|---|
| 455 | }
|
|---|
| 456 | }
|
|---|
| 457 |
|
|---|
| 458 | # We also have to remove from browselist - the reverse process of
|
|---|
| 459 | # adding to browselist shown above.
|
|---|
| 460 | my $dummy_model = new ClassifyTreeModel($collection, $infodbtype, "");
|
|---|
| 461 | my $browselist_node = new ClassifyTreeNode($dummy_model, "browselist");
|
|---|
| 462 | # Add the document
|
|---|
| 463 | $browselist_node->removeDocument($oid);
|
|---|
| 464 | # Clean up
|
|---|
| 465 | }
|
|---|
| 466 | # else, no document, no need to delete.
|
|---|
| 467 | }
|
|---|
| 468 | ## deleteDocument() ##
|
|---|
| 469 |
|
|---|
| 470 | # /**
|
|---|
| 471 | # */
|
|---|
| 472 | sub getFieldFromBuildCFG()
|
|---|
| 473 | {
|
|---|
| 474 | my ($indexpath, $key) = @_;
|
|---|
| 475 | my $field = "";
|
|---|
| 476 | my $build_cfg = &util::filename_cat($indexpath, "build.cfg");
|
|---|
| 477 | # If there isn't a build.cfg then the index hasn't been built and there is
|
|---|
| 478 | # nothing to do
|
|---|
| 479 | if(open(BUILDCFG, $build_cfg))
|
|---|
| 480 | {
|
|---|
| 481 | # For each line of the build configuration
|
|---|
| 482 | my $line;
|
|---|
| 483 | while($line = <BUILDCFG>)
|
|---|
| 484 | {
|
|---|
| 485 | # Only interested in the indexfieldmap line
|
|---|
| 486 | if($line =~ /^indexfieldmap\s+/)
|
|---|
| 487 | {
|
|---|
| 488 | # Extract the field information by looking up the key pair
|
|---|
| 489 | if($line =~ /\s$key->(\w\w)/)
|
|---|
| 490 | {
|
|---|
| 491 | $field = $1;
|
|---|
| 492 | }
|
|---|
| 493 | }
|
|---|
| 494 | }
|
|---|
| 495 | # Done with file
|
|---|
| 496 | close(BUILDCFG);
|
|---|
| 497 | }
|
|---|
| 498 | # Return whatever we found
|
|---|
| 499 | return $field;
|
|---|
| 500 | }
|
|---|
| 501 | # /** getFieldFromBuildCFG() **/
|
|---|
| 502 |
|
|---|
| 503 |
|
|---|
| 504 |
|
|---|
| 505 |
|
|---|
| 506 |
|
|---|
| 507 | # /** Retrieve an object (associative array) containing information about the
|
|---|
| 508 | # * collection configuration.
|
|---|
| 509 | # * @param $collection The shortname of the collection as a string
|
|---|
| 510 | # * @return An associative array containing information from the collect.cfg
|
|---|
| 511 | # * @author John Thompson, DL Consulting Ltd.
|
|---|
| 512 | # */
|
|---|
| 513 | sub getConfigObj()
|
|---|
| 514 | {
|
|---|
| 515 | my ($collection) = @_;
|
|---|
| 516 |
|
|---|
| 517 | #rint STDERR "getConfigObj()\n" unless !$debug;
|
|---|
| 518 |
|
|---|
| 519 | my $colcfgname = &util::filename_cat($ENV{'GSDLHOME'}, "collect", $collection, "etc", "collect.cfg");
|
|---|
| 520 | if (!-e $colcfgname)
|
|---|
| 521 | {
|
|---|
| 522 | die "IncrementalBuildUtils - couldn't find collect.cfg for collection $collection\n";
|
|---|
| 523 | }
|
|---|
| 524 | return &colcfg::read_collect_cfg ($colcfgname);
|
|---|
| 525 | }
|
|---|
| 526 | # /** getConfigObj() **/
|
|---|
| 527 |
|
|---|
| 528 | 1;
|
|---|
