[24626] | 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 | require AllList;
|
---|
| 32 |
|
---|
| 33 | use dbutil;
|
---|
| 34 | use gsprintf;
|
---|
| 35 | use strict; no strict 'subs';
|
---|
| 36 |
|
---|
| 37 |
|
---|
| 38 | sub gsprintf
|
---|
| 39 | {
|
---|
| 40 | return &gsprintf::gsprintf(@_);
|
---|
| 41 | }
|
---|
| 42 |
|
---|
| 43 |
|
---|
| 44 | sub load_classifier_for_info {
|
---|
| 45 | my ($classifier) = shift @_;
|
---|
| 46 |
|
---|
| 47 | # find the classifier
|
---|
| 48 | # my $customclassname;
|
---|
| 49 | # if (defined($ENV{'GSDLCOLLECTION'}))
|
---|
| 50 | # {
|
---|
| 51 | # $customclassname = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "custom", $ENV{'GSDLCOLLECTION'},
|
---|
| 52 | # "perllib", "classify", "${classifier}.pm");
|
---|
| 53 | # }
|
---|
| 54 | # my $colclassname = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "perllib", "classify", "${classifier}.pm");
|
---|
| 55 | # my $mainclassname = &util::filename_cat($ENV{'GSDLHOME'}, "perllib", "classify", "${classifier}.pm");
|
---|
| 56 | #
|
---|
| 57 | # if (defined($customclassname) && -e $customclassname) { require $customclassname; }
|
---|
| 58 | # elsif (-e $colclassname) { require $colclassname; }
|
---|
| 59 | # elsif (-e $mainclassname) { require $mainclassname; }
|
---|
| 60 | # else {
|
---|
| 61 | # &gsprintf(STDERR, "{classify.could_not_find_classifier}\n", $classifier) && die "\n";
|
---|
| 62 | # }
|
---|
| 63 |
|
---|
| 64 | # = Extensions Support =
|
---|
| 65 | # Used to have hardcoded list of places to load classifier from. We should,
|
---|
| 66 | # instead, try loading from all of the perllib places on the library path.
|
---|
| 67 | # Special cases nedded for collection specific and custom classifier.
|
---|
| 68 | my @possible_class_paths;
|
---|
| 69 | if (defined($ENV{'GSDLCOLLECTION'}))
|
---|
| 70 | {
|
---|
| 71 | push(@possible_class_paths, &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, 'custom', $ENV{'GSDLCOLLECTION'}, 'perllib', 'classify', $classifier . '.pm'));
|
---|
| 72 | }
|
---|
| 73 | # (why does GSDLCOLLECTDIR get set to GSDLHOME for classinfo calls?)
|
---|
| 74 | if ($ENV{'GSDLCOLLECTDIR'} ne $ENV{'GSDLHOME'})
|
---|
| 75 | {
|
---|
| 76 | push(@possible_class_paths, &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, 'perllib', 'classify', $classifier . '.pm'));
|
---|
| 77 | }
|
---|
| 78 | foreach my $library_path (@INC)
|
---|
| 79 | {
|
---|
| 80 | # only interested in classify paths found in the library paths
|
---|
| 81 | if ($library_path =~ /classify$/)
|
---|
| 82 | {
|
---|
| 83 | push(@possible_class_paths, &util::filename_cat($library_path, $classifier . '.pm'));
|
---|
| 84 | }
|
---|
| 85 | }
|
---|
| 86 | my $found_class = 0;
|
---|
| 87 | foreach my $possible_class_path (@possible_class_paths)
|
---|
| 88 | {
|
---|
| 89 | if (-e $possible_class_path)
|
---|
| 90 | {
|
---|
| 91 | require $possible_class_path;
|
---|
| 92 | $found_class = 1;
|
---|
| 93 | last;
|
---|
| 94 | }
|
---|
| 95 | }
|
---|
| 96 | if (!$found_class)
|
---|
| 97 | {
|
---|
| 98 | &gsprintf(STDERR, "{classify.could_not_find_classifier}\n", $classifier) && die "\n";
|
---|
| 99 | }
|
---|
| 100 |
|
---|
| 101 | my ($classobj);
|
---|
| 102 | my $options = "-gsdlinfo";
|
---|
| 103 | eval ("\$classobj = new \$classifier([],[$options])");
|
---|
| 104 | die "$@" if $@;
|
---|
| 105 |
|
---|
| 106 | return $classobj;
|
---|
| 107 | }
|
---|
| 108 |
|
---|
| 109 | sub load_classifiers {
|
---|
| 110 | my ($classify_list, $build_dir, $outhandle) = @_;
|
---|
| 111 | my @classify_objects = ();
|
---|
| 112 | my $classify_number = 1;
|
---|
| 113 |
|
---|
| 114 | my $colclassdir = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},"perllib/classify");
|
---|
| 115 | my $inc_paths = join(':',@INC);
|
---|
| 116 | if ($inc_paths !~ /$colclassdir/)
|
---|
| 117 | {
|
---|
| 118 | unshift (@INC, $colclassdir); # [jmt12]
|
---|
| 119 | }
|
---|
| 120 |
|
---|
| 121 | foreach my $classifyoption (@$classify_list) {
|
---|
| 122 |
|
---|
| 123 | # get the classifier name
|
---|
| 124 | my $classname = shift @$classifyoption;
|
---|
| 125 | next unless defined $classname;
|
---|
| 126 |
|
---|
| 127 | # find the classifier
|
---|
| 128 | # my $customclassname = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "custom", $ENV{'GSDLCOLLECTION'},
|
---|
| 129 | # "perllib", "classify", "${classname}.pm");
|
---|
| 130 | # my $colclassname = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "perllib", "classify", "${classname}.pm");
|
---|
| 131 | # my $mainclassname = &util::filename_cat($ENV{'GSDLHOME'}, "perllib", "classify", "${classname}.pm");
|
---|
| 132 |
|
---|
| 133 | # if (-e $customclassname) { require $customclassname; }
|
---|
| 134 | # elsif (-e $colclassname) { require $colclassname; }
|
---|
| 135 | # elsif (-e $mainclassname) { require $mainclassname; }
|
---|
| 136 | # else { &gsprintf(STDERR, "{classify.could_not_find_classifier}\n", $classname) && die "\n";
|
---|
| 137 | # die "ERROR - couldn't find classifier \"$classname\"\n";
|
---|
| 138 | # }
|
---|
| 139 |
|
---|
| 140 | # = Extensions Support =
|
---|
| 141 | # Used to have hardcoded list of places to load classifier from. We
|
---|
| 142 | # should, instead, try loading from all of the perllib places on the
|
---|
| 143 | # library path. Special cases nedded for collection specific and custom
|
---|
| 144 | # classifier. [jmt12]
|
---|
| 145 | my @possible_class_paths;
|
---|
| 146 | if (defined($ENV{'GSDLCOLLECTION'}))
|
---|
| 147 | {
|
---|
| 148 | push(@possible_class_paths, &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, 'custom', $ENV{'GSDLCOLLECTION'}, 'perllib', 'classify', $classname . '.pm'));
|
---|
| 149 | }
|
---|
| 150 | #push(@possible_class_paths,&util::filename_cat($ENV{'GSDLCOLLECTDIR'}, 'perllib', 'classify', $classname . '.pm'));
|
---|
| 151 | # (why is the collection specific classify path in INC multiple times?)
|
---|
| 152 | foreach my $library_path (@INC)
|
---|
| 153 | {
|
---|
| 154 | # only interested in classify paths found in the library paths
|
---|
| 155 | if ($library_path =~ /classify$/)
|
---|
| 156 | {
|
---|
| 157 | push(@possible_class_paths, &util::filename_cat($library_path, $classname . '.pm'));
|
---|
| 158 | }
|
---|
| 159 | }
|
---|
| 160 | my $found_class = 0;
|
---|
| 161 | foreach my $possible_class_path (@possible_class_paths)
|
---|
| 162 | {
|
---|
| 163 | if (-e $possible_class_path)
|
---|
| 164 | {
|
---|
| 165 | require $possible_class_path;
|
---|
| 166 | $found_class = 1;
|
---|
| 167 | last;
|
---|
| 168 | }
|
---|
| 169 | }
|
---|
| 170 | if (!$found_class)
|
---|
| 171 | {
|
---|
| 172 | &gsprintf(STDERR, "{classify.could_not_find_classifier}\n", $classname) && die "\n";
|
---|
| 173 | }
|
---|
| 174 |
|
---|
| 175 | # create the classify object
|
---|
| 176 | my ($classobj);
|
---|
| 177 |
|
---|
| 178 | my @newoptions;
|
---|
| 179 |
|
---|
| 180 | # do these first so they can be overriden by user supplied options
|
---|
| 181 | push @newoptions, "-builddir", "$build_dir" if ($build_dir);
|
---|
| 182 | push @newoptions, "-outhandle", "$outhandle" if ($outhandle);
|
---|
| 183 | push @newoptions, "-verbosity", "2";
|
---|
| 184 |
|
---|
| 185 | # backwards compatability hack: if the classifier options are
|
---|
| 186 | # in "x=y" format, convert them to parsearg ("-x y") format.
|
---|
| 187 | my ($opt, $key, $value);
|
---|
| 188 | foreach $opt (@$classifyoption) {
|
---|
| 189 | # if ($opt =~ /^(\w+)=(.*)$/) {
|
---|
| 190 | # push @newoptions, "-$1", $2;
|
---|
| 191 | # } else {
|
---|
| 192 | push @newoptions, $opt;
|
---|
| 193 | #}
|
---|
| 194 | }
|
---|
| 195 |
|
---|
| 196 | map { $_ = "\"$_\""; } @newoptions;
|
---|
| 197 | my $options .= join (",", @newoptions);
|
---|
| 198 |
|
---|
| 199 |
|
---|
| 200 | eval ("\$classobj = new \$classname([],[$options])");
|
---|
| 201 | die "$@" if $@;
|
---|
| 202 |
|
---|
| 203 | $classobj->set_number($classify_number);
|
---|
| 204 | $classify_number ++;
|
---|
| 205 |
|
---|
| 206 | # add this object to the list
|
---|
| 207 | push (@classify_objects, $classobj);
|
---|
| 208 | }
|
---|
| 209 |
|
---|
| 210 | my ($classobj);
|
---|
| 211 | eval ("\$classobj = new AllList()");
|
---|
| 212 | die "$@" if $@;
|
---|
| 213 | push (@classify_objects, $classobj);
|
---|
| 214 |
|
---|
| 215 | return \@classify_objects;
|
---|
| 216 | }
|
---|
| 217 |
|
---|
| 218 | # init_classifiers resets all the classifiers and readys them to process
|
---|
| 219 | # the documents.
|
---|
| 220 | sub init_classifiers {
|
---|
| 221 | my ($classifiers) = @_;
|
---|
| 222 |
|
---|
| 223 | foreach my $classobj (@$classifiers) {
|
---|
| 224 | $classobj->init();
|
---|
| 225 | }
|
---|
| 226 | }
|
---|
| 227 |
|
---|
| 228 |
|
---|
| 229 |
|
---|
| 230 | # takes a hashref containing the metadata for an infodb entry, and extracts
|
---|
| 231 | # the childrens numbers (from the 'contains' entry).
|
---|
| 232 | # assumes format is ".1;".2;".3
|
---|
| 233 | sub get_children {
|
---|
| 234 | my ($doc_db_hash) = @_;
|
---|
| 235 |
|
---|
| 236 | my $children = undef;
|
---|
| 237 |
|
---|
| 238 | my $contains = $doc_db_hash->{'contains'};
|
---|
| 239 | if (defined ($contains)) {
|
---|
| 240 | $contains =~ s/\@$//; #remove trailing @
|
---|
| 241 | $contains =~ s/^\"\.//; #remove initial ".
|
---|
| 242 | @$children = split /\;\"\./, $contains;
|
---|
| 243 | }
|
---|
| 244 |
|
---|
| 245 | return $children;
|
---|
| 246 | }
|
---|
| 247 |
|
---|
| 248 |
|
---|
| 249 | sub recurse_sections {
|
---|
| 250 | my ($doc_obj, $children, $parentoid, $parentsection, $database_recs) = @_;
|
---|
| 251 |
|
---|
| 252 | return if (!defined $children);
|
---|
| 253 |
|
---|
| 254 | foreach my $child (sort { $a <=> $b} @$children) {
|
---|
| 255 | $doc_obj->create_named_section("$parentsection.$child");
|
---|
| 256 | my $doc_db_rec = $database_recs->{"$parentoid.$child"};
|
---|
| 257 | my $doc_db_hash = db_rec_to_hash($doc_db_rec);
|
---|
| 258 |
|
---|
| 259 | # get child's children
|
---|
| 260 | my $newchildren = &get_children($doc_db_hash);
|
---|
| 261 |
|
---|
| 262 | # add content for current section
|
---|
| 263 | add_section_content($doc_obj, "$parentsection.$child", $doc_db_hash);
|
---|
| 264 |
|
---|
| 265 | # process all the children if there are any
|
---|
| 266 | if (defined ($newchildren))
|
---|
| 267 | {
|
---|
| 268 | recurse_sections($doc_obj, $newchildren, "$parentoid.$child",
|
---|
| 269 | "$parentsection.$child", $database_recs);
|
---|
| 270 | }
|
---|
| 271 | }
|
---|
| 272 | }
|
---|
| 273 |
|
---|
| 274 |
|
---|
| 275 | sub add_section_content {
|
---|
| 276 | my ($doc_obj, $cursection, $doc_db_hash) = @_;
|
---|
| 277 |
|
---|
| 278 | foreach my $key (keys %$doc_db_hash) {
|
---|
| 279 | #don't need to store these metadata
|
---|
| 280 | next if $key =~ /(thistype|childtype|contains|docnum|doctype|classifytype)/i;
|
---|
| 281 | # but do want things like hastxt and archivedir
|
---|
| 282 | my @items = split /@/, $doc_db_hash->{$key};
|
---|
| 283 | # metadata is all from the info database so should already be in utf8
|
---|
| 284 | map {$doc_obj->add_utf8_metadata ($cursection, $key, $_); } @items;
|
---|
| 285 |
|
---|
| 286 | }
|
---|
| 287 | }
|
---|
| 288 |
|
---|
| 289 |
|
---|
| 290 | # gets all the metadata from an infodb entry, and puts it into a hashref
|
---|
| 291 | sub db_rec_to_hash {
|
---|
| 292 |
|
---|
| 293 | my ($infodb_str_ref) = @_;
|
---|
| 294 |
|
---|
| 295 | my $hashref = {};
|
---|
| 296 |
|
---|
| 297 | my @entries = split(/\n/, $infodb_str_ref);
|
---|
| 298 | foreach my $entry (@entries) {
|
---|
| 299 | my($key, $value) = ($entry =~ /^<([^>]*)>(.*?)$/ );
|
---|
| 300 | $hashref->{$key} .= '@' if defined $hashref->{$key};
|
---|
| 301 | $hashref->{$key} .= $value;
|
---|
| 302 |
|
---|
| 303 | }
|
---|
| 304 |
|
---|
| 305 | return $hashref;
|
---|
| 306 | }
|
---|
| 307 |
|
---|
| 308 |
|
---|
| 309 | sub reconstruct_doc_objs_metadata
|
---|
| 310 | {
|
---|
| 311 | my $infodb_type = shift(@_);
|
---|
| 312 | my $infodb_file_path = shift(@_);
|
---|
| 313 | my $database_recs = shift(@_);
|
---|
| 314 |
|
---|
| 315 | # dig out top level doc sections
|
---|
| 316 | my %top_sections = ();
|
---|
| 317 | my %top_docnums = ();
|
---|
| 318 | foreach my $key ( keys %$database_recs )
|
---|
| 319 | {
|
---|
| 320 | my $md_rec = $database_recs->{$key};
|
---|
| 321 | my $md_hash = db_rec_to_hash($md_rec);
|
---|
| 322 |
|
---|
| 323 | if ((defined $md_hash->{'doctype'}) && ($md_hash->{'doctype'} eq "doc")) {
|
---|
| 324 | next if ($key =~ m/\./);
|
---|
| 325 | $top_sections{$key} = $md_hash;
|
---|
| 326 | $top_docnums{$key} = $md_hash->{'docnum'};
|
---|
| 327 | }
|
---|
| 328 | }
|
---|
| 329 |
|
---|
| 330 | # for greenstone document objects based on metadata in database file
|
---|
| 331 | my @all_docs = ();
|
---|
| 332 | # we need to make sure the documents were processed in the same order as
|
---|
| 333 | # before, so sort based on their docnums
|
---|
| 334 | foreach my $oid ( sort { $top_docnums{$a} <=> $top_docnums{$b} } keys %top_sections )
|
---|
| 335 | {
|
---|
| 336 | my $doc_db_hash = $top_sections{$oid};
|
---|
| 337 |
|
---|
| 338 | my $doc_obj = new doc();
|
---|
| 339 | $doc_obj->set_OID($oid);
|
---|
| 340 | my $top = $doc_obj->get_top_section();
|
---|
| 341 | add_section_content ($doc_obj, $top, $doc_db_hash);
|
---|
| 342 | my $children = &get_children($doc_db_hash);
|
---|
| 343 | recurse_sections($doc_obj, $children, $oid, $top, $database_recs);
|
---|
| 344 |
|
---|
| 345 | push(@all_docs,$doc_obj);
|
---|
| 346 | }
|
---|
| 347 |
|
---|
| 348 | return \@all_docs;
|
---|
| 349 | }
|
---|
| 350 |
|
---|
| 351 |
|
---|
| 352 |
|
---|
| 353 |
|
---|
| 354 |
|
---|
| 355 | # classify_doc lets each of the classifiers classify a document
|
---|
| 356 | sub classify_doc {
|
---|
| 357 | my ($classifiers, $doc_obj) = @_;
|
---|
| 358 |
|
---|
| 359 | foreach my $classobj (@$classifiers) {
|
---|
| 360 | my $title = $classobj->{'title'};
|
---|
| 361 |
|
---|
| 362 | $classobj->classify($doc_obj);
|
---|
| 363 | }
|
---|
| 364 | }
|
---|
| 365 |
|
---|
| 366 |
|
---|
| 367 | our $next_classify_num = 1;
|
---|
| 368 |
|
---|
| 369 | # output_classify_info outputs all the info needed for the classification
|
---|
| 370 | # to the database
|
---|
| 371 | sub output_classify_info
|
---|
| 372 | {
|
---|
| 373 | my ($classifiers, $infodb_type, $infodb_handle, $remove_empty_classifications, $gli) = @_;
|
---|
| 374 |
|
---|
| 375 | $gli = 0 unless defined $gli;
|
---|
| 376 |
|
---|
| 377 | # create a classification containing all the info
|
---|
| 378 | my $classifyinfo = { 'classifyOID'=> 'browse',
|
---|
| 379 | 'contains' => [] };
|
---|
| 380 |
|
---|
| 381 | # get each of the classifications
|
---|
| 382 | foreach my $classifier (@$classifiers)
|
---|
| 383 | {
|
---|
| 384 | my $classifier_info = $classifier->get_classify_info($gli);
|
---|
| 385 | $classifier_info->{'classifyOID'} = "CL$next_classify_num" unless defined($classifier_info->{'classifyOID'});
|
---|
| 386 | print STDERR "*** outputting information for classifier: $classifier_info->{'classifyOID'}\n";
|
---|
| 387 |
|
---|
| 388 | push(@{$classifyinfo->{'contains'}}, $classifier_info);
|
---|
| 389 | $next_classify_num++;
|
---|
| 390 | }
|
---|
| 391 |
|
---|
| 392 | &print_classify_info($infodb_type, $infodb_handle, $classifyinfo, "", $remove_empty_classifications);
|
---|
| 393 | }
|
---|
| 394 |
|
---|
| 395 |
|
---|
| 396 | sub print_classify_info
|
---|
| 397 | {
|
---|
| 398 | my ($infodb_type, $infodb_handle, $classifyinfo, $OID, $remove_empty_classifications) = @_;
|
---|
| 399 |
|
---|
| 400 | $OID =~ s/^\.+//; # just for good luck
|
---|
| 401 |
|
---|
| 402 | # book information is printed elsewhere
|
---|
| 403 | return if (defined ($classifyinfo->{'OID'}));
|
---|
| 404 |
|
---|
| 405 | # don't want empty classifications
|
---|
| 406 | return if (&check_contents ($classifyinfo, $remove_empty_classifications) == 0 && $remove_empty_classifications);
|
---|
| 407 |
|
---|
| 408 | $OID = $classifyinfo->{'classifyOID'} if defined ($classifyinfo->{'classifyOID'});
|
---|
| 409 |
|
---|
| 410 | my %classify_infodb = ();
|
---|
| 411 | $classify_infodb{"doctype"} = [ "classify" ];
|
---|
| 412 | $classify_infodb{"hastxt"} = [ "0" ];
|
---|
| 413 | $classify_infodb{"childtype"} = [ $classifyinfo->{'childtype'} ]
|
---|
| 414 | if defined $classifyinfo->{'childtype'};
|
---|
| 415 | $classify_infodb{"Title"} = [ $classifyinfo->{'Title'} ]
|
---|
| 416 | if defined $classifyinfo->{'Title'};
|
---|
| 417 | $classify_infodb{"numleafdocs"} = [ $classifyinfo->{'numleafdocs'} ]
|
---|
| 418 | if defined $classifyinfo->{'numleafdocs'};
|
---|
| 419 | $classify_infodb{"thistype"} = [ $classifyinfo->{'thistype'} ]
|
---|
| 420 | if defined $classifyinfo->{'thistype'};
|
---|
| 421 | $classify_infodb{"parameters"} = [ $classifyinfo->{'parameters'} ]
|
---|
| 422 | if defined $classifyinfo->{'parameters'};
|
---|
| 423 | $classify_infodb{"supportsmemberof"} = [ $classifyinfo->{'supportsmemberof'} ]
|
---|
| 424 | if defined $classifyinfo->{'supportsmemberof'};
|
---|
| 425 |
|
---|
| 426 | my $contains_text = "";
|
---|
| 427 | my $mdoffset_text = "";
|
---|
| 428 |
|
---|
| 429 | my $next_subOID = 1;
|
---|
| 430 | my $first = 1;
|
---|
| 431 | foreach my $tempinfo (@{$classifyinfo->{'contains'}}) {
|
---|
| 432 | # empty contents were made undefined by clean_contents()
|
---|
| 433 | next unless defined $tempinfo;
|
---|
| 434 | if (!defined ($tempinfo->{'classifyOID'}) ||
|
---|
| 435 | $tempinfo->{'classifyOID'} ne "oai") {
|
---|
| 436 | $contains_text .= ";" unless $first;
|
---|
| 437 | }
|
---|
| 438 | $mdoffset_text .= ";" unless $first;
|
---|
| 439 | $first = 0;
|
---|
| 440 |
|
---|
| 441 | if (defined ($tempinfo->{'classifyOID'}))
|
---|
| 442 | {
|
---|
| 443 | if ($tempinfo->{'classifyOID'} ne "oai")
|
---|
| 444 | {
|
---|
| 445 | $contains_text .= $tempinfo->{'classifyOID'};
|
---|
| 446 | }
|
---|
| 447 |
|
---|
| 448 | &print_classify_info ($infodb_type, $infodb_handle, $tempinfo, $tempinfo->{'classifyOID'},
|
---|
| 449 | $remove_empty_classifications);
|
---|
| 450 | }
|
---|
| 451 | elsif (defined ($tempinfo->{'OID'}))
|
---|
| 452 | {
|
---|
| 453 | $contains_text .= $tempinfo->{'OID'};
|
---|
| 454 | $mdoffset_text .= $tempinfo->{'offset'} if (defined ($tempinfo->{'offset'}));
|
---|
| 455 | }
|
---|
| 456 | else
|
---|
| 457 | {
|
---|
| 458 | # Supress having top-level node in Collage classifier
|
---|
| 459 | # so no bookshelf icon appears, top-level, along with the
|
---|
| 460 | # applet
|
---|
| 461 | if (!defined ($tempinfo->{'Title'}) || $tempinfo->{'Title'} ne "Collage")
|
---|
| 462 | {
|
---|
| 463 | $contains_text .= "\".$next_subOID";
|
---|
| 464 | }
|
---|
| 465 |
|
---|
| 466 | &print_classify_info ($infodb_type, $infodb_handle, $tempinfo, "$OID.$next_subOID",
|
---|
| 467 | $remove_empty_classifications);
|
---|
| 468 | $next_subOID++;
|
---|
| 469 | }
|
---|
| 470 | }
|
---|
| 471 |
|
---|
| 472 | $classify_infodb{"contains"} = [ $contains_text ];
|
---|
| 473 | $classify_infodb{"mdtype"} = [ $classifyinfo->{'mdtype'} ]
|
---|
| 474 | if defined $classifyinfo->{'mdtype'};
|
---|
| 475 | $classify_infodb{"mdoffset"} = [ $mdoffset_text ]
|
---|
| 476 | if ($mdoffset_text !~ m/^;+$/);
|
---|
| 477 |
|
---|
| 478 | &dbutil::write_infodb_entry($infodb_type, $infodb_handle, $OID, \%classify_infodb);
|
---|
| 479 | }
|
---|
| 480 |
|
---|
| 481 |
|
---|
| 482 | sub check_contents {
|
---|
| 483 | my ($classifyinfo,$remove_empty_classifications) = @_;
|
---|
| 484 | $remove_empty_classifications = 0 unless ($remove_empty_classifications);
|
---|
| 485 | my $num_leaf_docs = 0;
|
---|
| 486 | my $sub_num_leaf_docs = 0;
|
---|
| 487 |
|
---|
| 488 | return $classifyinfo->{'numleafdocs'} if (defined $classifyinfo->{'numleafdocs'});
|
---|
| 489 |
|
---|
| 490 | foreach my $content (@{$classifyinfo->{'contains'}}) {
|
---|
| 491 | if (defined $content->{'OID'}) {
|
---|
| 492 | # found a book
|
---|
| 493 | $num_leaf_docs ++;
|
---|
| 494 | } elsif (($sub_num_leaf_docs = &check_contents ($content,$remove_empty_classifications)) > 0) {
|
---|
| 495 | # there's a book somewhere below
|
---|
| 496 | $num_leaf_docs += $sub_num_leaf_docs;
|
---|
| 497 | } else {
|
---|
| 498 | if ($remove_empty_classifications){
|
---|
| 499 | # section contains no books so we want to remove
|
---|
| 500 | # it from its parents contents
|
---|
| 501 | $content = undef;
|
---|
| 502 | }
|
---|
| 503 | }
|
---|
| 504 | }
|
---|
| 505 |
|
---|
| 506 | $classifyinfo->{'numleafdocs'} = $num_leaf_docs;
|
---|
| 507 | return $num_leaf_docs;
|
---|
| 508 | }
|
---|
| 509 |
|
---|
| 510 | 1;
|
---|