| 1 | ###########################################################################
|
|---|
| 2 | #
|
|---|
| 3 | # classify.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 | # functions to handle classifiers
|
|---|
| 27 |
|
|---|
| 28 | package classify;
|
|---|
| 29 |
|
|---|
| 30 | require util;
|
|---|
| 31 | use FileUtils;
|
|---|
| 32 | require AllList;
|
|---|
| 33 |
|
|---|
| 34 | use dbutil;
|
|---|
| 35 | use gsprintf;
|
|---|
| 36 | use strict; no strict 'subs';
|
|---|
| 37 |
|
|---|
| 38 |
|
|---|
| 39 | sub gsprintf
|
|---|
| 40 | {
|
|---|
| 41 | return &gsprintf::gsprintf(@_);
|
|---|
| 42 | }
|
|---|
| 43 |
|
|---|
| 44 |
|
|---|
| 45 | sub load_classifier_for_info {
|
|---|
| 46 | my ($classifier) = shift @_;
|
|---|
| 47 |
|
|---|
| 48 | # find the classifier
|
|---|
| 49 | # - used to have hardcoded list of places to load classifier from. We
|
|---|
| 50 | # should, instead, try loading from all of the perllib places on the
|
|---|
| 51 | # library path, as that improves support for extensions. Special cases
|
|---|
| 52 | # needed for collection specific and custom classifier. [jmt12]
|
|---|
| 53 | my @possible_class_paths;
|
|---|
| 54 | if (defined($ENV{'GSDLCOLLECTION'}))
|
|---|
| 55 | {
|
|---|
| 56 | push(@possible_class_paths, &FileUtils::filenameConcatenate($ENV{'GSDLCOLLECTDIR'}, 'custom', $ENV{'GSDLCOLLECTION'}, 'perllib', 'classify', $classifier . '.pm'));
|
|---|
| 57 | }
|
|---|
| 58 | # (why does GSDLCOLLECTDIR get set to GSDLHOME for classinfo calls?)
|
|---|
| 59 | if ($ENV{'GSDLCOLLECTDIR'} ne $ENV{'GSDLHOME'})
|
|---|
| 60 | {
|
|---|
| 61 | push(@possible_class_paths, &FileUtils::filenameConcatenate($ENV{'GSDLCOLLECTDIR'}, 'perllib', 'classify', $classifier . '.pm'));
|
|---|
| 62 | }
|
|---|
| 63 | foreach my $library_path (@INC)
|
|---|
| 64 | {
|
|---|
| 65 | # only interested in classify paths found in the library paths
|
|---|
| 66 | if ($library_path =~ /classify$/)
|
|---|
| 67 | {
|
|---|
| 68 | push(@possible_class_paths, &FileUtils::filenameConcatenate($library_path, $classifier . '.pm'));
|
|---|
| 69 | }
|
|---|
| 70 | }
|
|---|
| 71 | my $found_class = 0;
|
|---|
| 72 | foreach my $possible_class_path (@possible_class_paths)
|
|---|
| 73 | {
|
|---|
| 74 | if (-e $possible_class_path)
|
|---|
| 75 | {
|
|---|
| 76 | require $possible_class_path;
|
|---|
| 77 | $found_class = 1;
|
|---|
| 78 | last;
|
|---|
| 79 | }
|
|---|
| 80 | }
|
|---|
| 81 | if (!$found_class)
|
|---|
| 82 | {
|
|---|
| 83 | &gsprintf(STDERR, "{classify.could_not_find_classifier}\n", $classifier) && die "\n";
|
|---|
| 84 | }
|
|---|
| 85 |
|
|---|
| 86 | my ($classobj);
|
|---|
| 87 | my $options = "-gsdlinfo";
|
|---|
| 88 | eval ("\$classobj = new \$classifier([],[$options])");
|
|---|
| 89 | die "$@" if $@;
|
|---|
| 90 |
|
|---|
| 91 | return $classobj;
|
|---|
| 92 | }
|
|---|
| 93 |
|
|---|
| 94 | sub load_classifiers {
|
|---|
| 95 | my ($classify_list, $build_dir, $outhandle) = @_;
|
|---|
| 96 | my @classify_objects = ();
|
|---|
| 97 | my $classify_number = 1;
|
|---|
| 98 |
|
|---|
| 99 | my $colclassdir = &FileUtils::filenameConcatenate($ENV{'GSDLCOLLECTDIR'},"perllib/classify");
|
|---|
| 100 | # - ensure colclassdir doesn't already exist in INC before adding, other-
|
|---|
| 101 | # wise we risk clobbering classifier inheritence implied by order of paths
|
|---|
| 102 | # in INC [jmt12]
|
|---|
| 103 | my $inc_paths = join(':',@INC);
|
|---|
| 104 |
|
|---|
| 105 | $colclassdir = &FileUtils::filenameConcatenate($colclassdir); # no mix of / and \ slashes, only OS-specific slash
|
|---|
| 106 | $colclassdir = &util::filename_to_regex($colclassdir); # to regex protect windows filepaths
|
|---|
| 107 | # Normalise slashes in INC filepaths: ensure it has slashes going the same way as $colclassdir
|
|---|
| 108 | # and further doesn't have a mix of / and \ slashes.
|
|---|
| 109 | $inc_paths = &FileUtils::filenameConcatenate($inc_paths);
|
|---|
| 110 | print STDERR "**** INC: $colclassdir | $inc_paths\n";
|
|---|
| 111 | if ($inc_paths !~ m@$colclassdir@)
|
|---|
| 112 | {
|
|---|
| 113 | unshift (@INC, $colclassdir);
|
|---|
| 114 | }
|
|---|
| 115 |
|
|---|
| 116 | foreach my $classifyoption (@$classify_list) {
|
|---|
| 117 |
|
|---|
| 118 | # get the classifier name
|
|---|
| 119 | my $classname = shift @$classifyoption;
|
|---|
| 120 | next unless defined $classname;
|
|---|
| 121 |
|
|---|
| 122 | # find the classifier
|
|---|
| 123 | # - replaced as explained in load_classifier_for_info() [jmt12]
|
|---|
| 124 | my @possible_class_paths;
|
|---|
| 125 | if (defined($ENV{'GSDLCOLLECTION'}))
|
|---|
| 126 | {
|
|---|
| 127 | push(@possible_class_paths, &FileUtils::filenameConcatenate($ENV{'GSDLCOLLECTDIR'}, 'custom', $ENV{'GSDLCOLLECTION'}, 'perllib', 'classify', $classname . '.pm'));
|
|---|
| 128 | }
|
|---|
| 129 | # (why does GSDLCOLLECTDIR get set to GSDLHOME for classinfo calls?)
|
|---|
| 130 | if ($ENV{'GSDLCOLLECTDIR'} ne $ENV{'GSDLHOME'})
|
|---|
| 131 | {
|
|---|
| 132 | push(@possible_class_paths,&FileUtils::filenameConcatenate($ENV{'GSDLCOLLECTDIR'}, 'perllib', 'classify', $classname . '.pm'));
|
|---|
| 133 | }
|
|---|
| 134 | foreach my $library_path (@INC)
|
|---|
| 135 | {
|
|---|
| 136 | # only interested in classify paths found in the library paths
|
|---|
| 137 | if ($library_path =~ /classify$/)
|
|---|
| 138 | {
|
|---|
| 139 | push(@possible_class_paths, &FileUtils::filenameConcatenate($library_path, $classname . '.pm'));
|
|---|
| 140 | }
|
|---|
| 141 | }
|
|---|
| 142 | my $found_class = 0;
|
|---|
| 143 | foreach my $possible_class_path (@possible_class_paths)
|
|---|
| 144 | {
|
|---|
| 145 | if (-e $possible_class_path)
|
|---|
| 146 | {
|
|---|
| 147 | require $possible_class_path;
|
|---|
| 148 | $found_class = 1;
|
|---|
| 149 | last;
|
|---|
| 150 | }
|
|---|
| 151 | }
|
|---|
| 152 | if (!$found_class)
|
|---|
| 153 | {
|
|---|
| 154 | &gsprintf(STDERR, "{classify.could_not_find_classifier}\n", $classname) && die "\n";
|
|---|
| 155 | }
|
|---|
| 156 |
|
|---|
| 157 | # create the classify object
|
|---|
| 158 | my ($classobj);
|
|---|
| 159 |
|
|---|
| 160 | my @newoptions;
|
|---|
| 161 |
|
|---|
| 162 | # do these first so they can be overriden by user supplied options
|
|---|
| 163 | push @newoptions, "-builddir", "$build_dir" if ($build_dir);
|
|---|
| 164 | push @newoptions, "-outhandle", "$outhandle" if ($outhandle);
|
|---|
| 165 | push @newoptions, "-verbosity", "2";
|
|---|
| 166 |
|
|---|
| 167 | # backwards compatability hack: if the classifier options are
|
|---|
| 168 | # in "x=y" format, convert them to parsearg ("-x y") format.
|
|---|
| 169 | my ($opt, $key, $value);
|
|---|
| 170 | foreach $opt (@$classifyoption) {
|
|---|
| 171 | # if ($opt =~ /^(\w+)=(.*)$/) {
|
|---|
| 172 | # push @newoptions, "-$1", $2;
|
|---|
| 173 | # } else {
|
|---|
| 174 | push @newoptions, $opt;
|
|---|
| 175 | #}
|
|---|
| 176 | }
|
|---|
| 177 |
|
|---|
| 178 | map { $_ = "\"$_\""; } @newoptions;
|
|---|
| 179 | my $options .= join (",", @newoptions);
|
|---|
| 180 |
|
|---|
| 181 |
|
|---|
| 182 | eval ("\$classobj = new \$classname([],[$options])");
|
|---|
| 183 | die "$@" if $@;
|
|---|
| 184 |
|
|---|
| 185 | $classobj->set_number($classify_number);
|
|---|
| 186 | $classify_number ++;
|
|---|
| 187 |
|
|---|
| 188 | # add this object to the list
|
|---|
| 189 | push (@classify_objects, $classobj);
|
|---|
| 190 | }
|
|---|
| 191 |
|
|---|
| 192 | my ($classobj);
|
|---|
| 193 | eval ("\$classobj = new AllList()");
|
|---|
| 194 | die "$@" if $@;
|
|---|
| 195 | push (@classify_objects, $classobj);
|
|---|
| 196 |
|
|---|
| 197 | return \@classify_objects;
|
|---|
| 198 | }
|
|---|
| 199 |
|
|---|
| 200 | # init_classifiers resets all the classifiers and readys them to process
|
|---|
| 201 | # the documents.
|
|---|
| 202 | sub init_classifiers {
|
|---|
| 203 | my ($classifiers) = @_;
|
|---|
| 204 |
|
|---|
| 205 | foreach my $classobj (@$classifiers) {
|
|---|
| 206 | $classobj->init();
|
|---|
| 207 | }
|
|---|
| 208 | }
|
|---|
| 209 |
|
|---|
| 210 |
|
|---|
| 211 |
|
|---|
| 212 | # takes a hashref containing the metadata for an infodb entry, and extracts
|
|---|
| 213 | # the childrens numbers (from the 'contains' entry).
|
|---|
| 214 | # assumes format is ".1;".2;".3
|
|---|
| 215 | sub get_children {
|
|---|
| 216 | my ($doc_db_hash) = @_;
|
|---|
| 217 |
|
|---|
| 218 | my $children = undef;
|
|---|
| 219 |
|
|---|
| 220 | my $contains = $doc_db_hash->{'contains'};
|
|---|
| 221 | if (defined ($contains)) {
|
|---|
| 222 | $contains =~ s/\@$//; #remove trailing @
|
|---|
| 223 | $contains =~ s/^\"\.//; #remove initial ".
|
|---|
| 224 | @$children = split /\;\"\./, $contains;
|
|---|
| 225 | }
|
|---|
| 226 |
|
|---|
| 227 | return $children;
|
|---|
| 228 | }
|
|---|
| 229 |
|
|---|
| 230 |
|
|---|
| 231 | sub recurse_sections {
|
|---|
| 232 | my ($doc_obj, $children, $parentoid, $parentsection, $database_recs) = @_;
|
|---|
| 233 |
|
|---|
| 234 | return if (!defined $children);
|
|---|
| 235 |
|
|---|
| 236 | foreach my $child (sort { $a <=> $b} @$children) {
|
|---|
| 237 | $doc_obj->create_named_section("$parentsection.$child");
|
|---|
| 238 | my $doc_db_rec = $database_recs->{"$parentoid.$child"};
|
|---|
| 239 | my $doc_db_hash = db_rec_to_hash($doc_db_rec);
|
|---|
| 240 |
|
|---|
| 241 | # get child's children
|
|---|
| 242 | my $newchildren = &get_children($doc_db_hash);
|
|---|
| 243 |
|
|---|
| 244 | # add content for current section
|
|---|
| 245 | add_section_content($doc_obj, "$parentsection.$child", $doc_db_hash);
|
|---|
| 246 |
|
|---|
| 247 | # process all the children if there are any
|
|---|
| 248 | if (defined ($newchildren))
|
|---|
| 249 | {
|
|---|
| 250 | recurse_sections($doc_obj, $newchildren, "$parentoid.$child",
|
|---|
| 251 | "$parentsection.$child", $database_recs);
|
|---|
| 252 | }
|
|---|
| 253 | }
|
|---|
| 254 | }
|
|---|
| 255 |
|
|---|
| 256 |
|
|---|
| 257 | sub add_section_content {
|
|---|
| 258 | my ($doc_obj, $cursection, $doc_db_hash) = @_;
|
|---|
| 259 |
|
|---|
| 260 | foreach my $key (keys %$doc_db_hash) {
|
|---|
| 261 | #don't need to store these metadata
|
|---|
| 262 | next if $key =~ /(thistype|childtype|contains|docnum|doctype|classifytype)/i;
|
|---|
| 263 | # but do want things like hastxt and archivedir
|
|---|
| 264 | my @items = split /@/, $doc_db_hash->{$key};
|
|---|
| 265 | # metadata is all from the info database so should already be in utf8
|
|---|
| 266 | map {$doc_obj->add_utf8_metadata ($cursection, $key, $_); } @items;
|
|---|
| 267 |
|
|---|
| 268 | }
|
|---|
| 269 | }
|
|---|
| 270 |
|
|---|
| 271 |
|
|---|
| 272 | # gets all the metadata from an infodb entry, and puts it into a hashref
|
|---|
| 273 | sub db_rec_to_hash {
|
|---|
| 274 |
|
|---|
| 275 | my ($infodb_str_ref) = @_;
|
|---|
| 276 |
|
|---|
| 277 | my $hashref = {};
|
|---|
| 278 |
|
|---|
| 279 | my @entries = split(/\n/, $infodb_str_ref);
|
|---|
| 280 | foreach my $entry (@entries) {
|
|---|
| 281 | my($key, $value) = ($entry =~ /^<([^>]*)>(.*?)$/ );
|
|---|
| 282 | $hashref->{$key} .= '@' if defined $hashref->{$key};
|
|---|
| 283 | $hashref->{$key} .= $value;
|
|---|
| 284 |
|
|---|
| 285 | }
|
|---|
| 286 |
|
|---|
| 287 | return $hashref;
|
|---|
| 288 | }
|
|---|
| 289 |
|
|---|
| 290 |
|
|---|
| 291 | sub reconstruct_doc_objs_metadata
|
|---|
| 292 | {
|
|---|
| 293 | my $infodb_type = shift(@_);
|
|---|
| 294 | my $infodb_file_path = shift(@_);
|
|---|
| 295 | my $database_recs = shift(@_);
|
|---|
| 296 |
|
|---|
| 297 | # dig out top level doc sections
|
|---|
| 298 | my %top_sections = ();
|
|---|
| 299 | my %top_docnums = ();
|
|---|
| 300 | foreach my $key ( keys %$database_recs )
|
|---|
| 301 | {
|
|---|
| 302 | my $md_rec = $database_recs->{$key};
|
|---|
| 303 | my $md_hash = db_rec_to_hash($md_rec);
|
|---|
| 304 |
|
|---|
| 305 | if ((defined $md_hash->{'doctype'}) && ($md_hash->{'doctype'} eq "doc")) {
|
|---|
| 306 | next if ($key =~ m/\./);
|
|---|
| 307 | $top_sections{$key} = $md_hash;
|
|---|
| 308 | $top_docnums{$key} = $md_hash->{'docnum'};
|
|---|
| 309 | }
|
|---|
| 310 | }
|
|---|
| 311 |
|
|---|
| 312 | # for greenstone document objects based on metadata in database file
|
|---|
| 313 | my @all_docs = ();
|
|---|
| 314 | # we need to make sure the documents were processed in the same order as
|
|---|
| 315 | # before, so sort based on their docnums
|
|---|
| 316 | foreach my $oid ( sort { $top_docnums{$a} <=> $top_docnums{$b} } keys %top_sections )
|
|---|
| 317 | {
|
|---|
| 318 | my $doc_db_hash = $top_sections{$oid};
|
|---|
| 319 |
|
|---|
| 320 | my $doc_obj = new doc();
|
|---|
| 321 | $doc_obj->set_OID($oid);
|
|---|
| 322 | my $top = $doc_obj->get_top_section();
|
|---|
| 323 | add_section_content ($doc_obj, $top, $doc_db_hash);
|
|---|
| 324 | my $children = &get_children($doc_db_hash);
|
|---|
| 325 | recurse_sections($doc_obj, $children, $oid, $top, $database_recs);
|
|---|
| 326 |
|
|---|
| 327 | push(@all_docs,$doc_obj);
|
|---|
| 328 | }
|
|---|
| 329 |
|
|---|
| 330 | return \@all_docs;
|
|---|
| 331 | }
|
|---|
| 332 |
|
|---|
| 333 |
|
|---|
| 334 |
|
|---|
| 335 |
|
|---|
| 336 |
|
|---|
| 337 | # classify_doc lets each of the classifiers classify a document
|
|---|
| 338 | sub classify_doc {
|
|---|
| 339 | my ($classifiers, $doc_obj) = @_;
|
|---|
| 340 |
|
|---|
| 341 | foreach my $classobj (@$classifiers) {
|
|---|
| 342 | my $title = $classobj->{'title'};
|
|---|
| 343 |
|
|---|
| 344 | $classobj->classify($doc_obj);
|
|---|
| 345 | }
|
|---|
| 346 | }
|
|---|
| 347 |
|
|---|
| 348 |
|
|---|
| 349 | our $next_classify_num = 1;
|
|---|
| 350 |
|
|---|
| 351 | # output_classify_info outputs all the info needed for the classification
|
|---|
| 352 | # to the database
|
|---|
| 353 | sub output_classify_info
|
|---|
| 354 | {
|
|---|
| 355 | my ($classifiers, $infodb_type, $infodb_handle, $remove_empty_classifications, $gli) = @_;
|
|---|
| 356 |
|
|---|
| 357 | $gli = 0 unless defined $gli;
|
|---|
| 358 |
|
|---|
| 359 | # create a classification containing all the info
|
|---|
| 360 | my $classifyinfo = { 'classifyOID'=> 'browse',
|
|---|
| 361 | 'contains' => [] };
|
|---|
| 362 |
|
|---|
| 363 | # get each of the classifications
|
|---|
| 364 | foreach my $classifier (@$classifiers)
|
|---|
| 365 | {
|
|---|
| 366 | my $classifier_info = $classifier->get_classify_info($gli);
|
|---|
| 367 | $classifier_info->{'classifyOID'} = "CL$next_classify_num" unless defined($classifier_info->{'classifyOID'});
|
|---|
| 368 | print STDERR "*** outputting information for classifier: $classifier_info->{'classifyOID'}\n";
|
|---|
| 369 |
|
|---|
| 370 | push(@{$classifyinfo->{'contains'}}, $classifier_info);
|
|---|
| 371 | $next_classify_num++;
|
|---|
| 372 | }
|
|---|
| 373 |
|
|---|
| 374 | &print_classify_info($infodb_type, $infodb_handle, $classifyinfo, "", $remove_empty_classifications);
|
|---|
| 375 | }
|
|---|
| 376 |
|
|---|
| 377 |
|
|---|
| 378 | sub print_classify_info
|
|---|
| 379 | {
|
|---|
| 380 | my ($infodb_type, $infodb_handle, $classifyinfo, $OID, $remove_empty_classifications) = @_;
|
|---|
| 381 |
|
|---|
| 382 | $OID =~ s/^\.+//; # just for good luck
|
|---|
| 383 |
|
|---|
| 384 | # book information is printed elsewhere
|
|---|
| 385 | return if (defined ($classifyinfo->{'OID'}));
|
|---|
| 386 |
|
|---|
| 387 | # don't want empty classifications
|
|---|
| 388 | return if (&check_contents ($classifyinfo, $remove_empty_classifications) == 0 && $remove_empty_classifications);
|
|---|
| 389 |
|
|---|
| 390 | $OID = $classifyinfo->{'classifyOID'} if defined ($classifyinfo->{'classifyOID'});
|
|---|
| 391 |
|
|---|
| 392 | my %classify_infodb = ();
|
|---|
| 393 | $classify_infodb{"doctype"} = [ "classify" ];
|
|---|
| 394 | $classify_infodb{"hastxt"} = [ "0" ];
|
|---|
| 395 | $classify_infodb{"childtype"} = [ $classifyinfo->{'childtype'} ]
|
|---|
| 396 | if defined $classifyinfo->{'childtype'};
|
|---|
| 397 | $classify_infodb{"Title"} = [ $classifyinfo->{'Title'} ]
|
|---|
| 398 | if defined $classifyinfo->{'Title'};
|
|---|
| 399 | $classify_infodb{"numleafdocs"} = [ $classifyinfo->{'numleafdocs'} ]
|
|---|
| 400 | if defined $classifyinfo->{'numleafdocs'};
|
|---|
| 401 | $classify_infodb{"thistype"} = [ $classifyinfo->{'thistype'} ]
|
|---|
| 402 | if defined $classifyinfo->{'thistype'};
|
|---|
| 403 | $classify_infodb{"parameters"} = [ $classifyinfo->{'parameters'} ]
|
|---|
| 404 | if defined $classifyinfo->{'parameters'};
|
|---|
| 405 | $classify_infodb{"supportsmemberof"} = [ $classifyinfo->{'supportsmemberof'} ]
|
|---|
| 406 | if defined $classifyinfo->{'supportsmemberof'};
|
|---|
| 407 |
|
|---|
| 408 | my $contains_text = "";
|
|---|
| 409 | my $mdoffset_text = "";
|
|---|
| 410 |
|
|---|
| 411 | my $next_subOID = 1;
|
|---|
| 412 | my $first = 1;
|
|---|
| 413 | foreach my $tempinfo (@{$classifyinfo->{'contains'}}) {
|
|---|
| 414 | # empty contents were made undefined by clean_contents()
|
|---|
| 415 | next unless defined $tempinfo;
|
|---|
| 416 | if (!defined ($tempinfo->{'classifyOID'}) ||
|
|---|
| 417 | $tempinfo->{'classifyOID'} ne "oai") {
|
|---|
| 418 | $contains_text .= ";" unless $first;
|
|---|
| 419 | }
|
|---|
| 420 | $mdoffset_text .= ";" unless $first;
|
|---|
| 421 | $first = 0;
|
|---|
| 422 |
|
|---|
| 423 | if (defined ($tempinfo->{'classifyOID'}))
|
|---|
| 424 | {
|
|---|
| 425 | if ($tempinfo->{'classifyOID'} ne "oai")
|
|---|
| 426 | {
|
|---|
| 427 | $contains_text .= $tempinfo->{'classifyOID'};
|
|---|
| 428 | }
|
|---|
| 429 |
|
|---|
| 430 | &print_classify_info ($infodb_type, $infodb_handle, $tempinfo, $tempinfo->{'classifyOID'},
|
|---|
| 431 | $remove_empty_classifications);
|
|---|
| 432 | }
|
|---|
| 433 | elsif (defined ($tempinfo->{'OID'}))
|
|---|
| 434 | {
|
|---|
| 435 | $contains_text .= $tempinfo->{'OID'};
|
|---|
| 436 | $mdoffset_text .= $tempinfo->{'offset'} if (defined ($tempinfo->{'offset'}));
|
|---|
| 437 | }
|
|---|
| 438 | else
|
|---|
| 439 | {
|
|---|
| 440 | # Supress having top-level node in Collage classifier
|
|---|
| 441 | # so no bookshelf icon appears, top-level, along with the
|
|---|
| 442 | # applet
|
|---|
| 443 | if (!defined ($tempinfo->{'Title'}) || $tempinfo->{'Title'} ne "Collage")
|
|---|
| 444 | {
|
|---|
| 445 | $contains_text .= "\".$next_subOID";
|
|---|
| 446 | }
|
|---|
| 447 |
|
|---|
| 448 | &print_classify_info ($infodb_type, $infodb_handle, $tempinfo, "$OID.$next_subOID",
|
|---|
| 449 | $remove_empty_classifications);
|
|---|
| 450 | $next_subOID++;
|
|---|
| 451 | }
|
|---|
| 452 | }
|
|---|
| 453 |
|
|---|
| 454 | $classify_infodb{"contains"} = [ $contains_text ];
|
|---|
| 455 | $classify_infodb{"mdtype"} = [ $classifyinfo->{'mdtype'} ]
|
|---|
| 456 | if defined $classifyinfo->{'mdtype'};
|
|---|
| 457 | $classify_infodb{"mdoffset"} = [ $mdoffset_text ]
|
|---|
| 458 | if ($mdoffset_text !~ m/^;+$/);
|
|---|
| 459 |
|
|---|
| 460 | &dbutil::write_infodb_entry($infodb_type, $infodb_handle, $OID, \%classify_infodb);
|
|---|
| 461 | }
|
|---|
| 462 |
|
|---|
| 463 |
|
|---|
| 464 | sub check_contents {
|
|---|
| 465 | my ($classifyinfo,$remove_empty_classifications) = @_;
|
|---|
| 466 | $remove_empty_classifications = 0 unless ($remove_empty_classifications);
|
|---|
| 467 | my $num_leaf_docs = 0;
|
|---|
| 468 | my $sub_num_leaf_docs = 0;
|
|---|
| 469 |
|
|---|
| 470 | return $classifyinfo->{'numleafdocs'} if (defined $classifyinfo->{'numleafdocs'});
|
|---|
| 471 |
|
|---|
| 472 | foreach my $content (@{$classifyinfo->{'contains'}}) {
|
|---|
| 473 | if (defined $content->{'OID'}) {
|
|---|
| 474 | # found a book
|
|---|
| 475 | $num_leaf_docs ++;
|
|---|
| 476 | } elsif (($sub_num_leaf_docs = &check_contents ($content,$remove_empty_classifications)) > 0) {
|
|---|
| 477 | # there's a book somewhere below
|
|---|
| 478 | $num_leaf_docs += $sub_num_leaf_docs;
|
|---|
| 479 | } else {
|
|---|
| 480 | if ($remove_empty_classifications){
|
|---|
| 481 | # section contains no books so we want to remove
|
|---|
| 482 | # it from its parents contents
|
|---|
| 483 | $content = undef;
|
|---|
| 484 | }
|
|---|
| 485 | }
|
|---|
| 486 | }
|
|---|
| 487 |
|
|---|
| 488 | $classifyinfo->{'numleafdocs'} = $num_leaf_docs;
|
|---|
| 489 | return $num_leaf_docs;
|
|---|
| 490 | }
|
|---|
| 491 |
|
|---|
| 492 | 1;
|
|---|
