[537] | 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 |
|
---|
[214] | 26 | # functions to handle classifiers
|
---|
| 27 |
|
---|
| 28 | package classify;
|
---|
| 29 |
|
---|
| 30 | require util;
|
---|
[8220] | 31 | require AllList;
|
---|
[5682] | 32 | use gsprintf;
|
---|
[214] | 33 |
|
---|
[12559] | 34 | #use GDBM_File;
|
---|
| 35 | use unbuildutil;
|
---|
[214] | 36 |
|
---|
[11994] | 37 |
|
---|
[5682] | 38 | sub gsprintf
|
---|
| 39 | {
|
---|
| 40 | return &gsprintf::gsprintf(@_);
|
---|
| 41 | }
|
---|
| 42 |
|
---|
| 43 |
|
---|
[315] | 44 | $next_classify_num = 1;
|
---|
[12844] | 45 | $oid_to_clids = {};
|
---|
| 46 |
|
---|
[6967] | 47 | sub load_classifier_for_info {
|
---|
| 48 | my ($classifier) = shift @_;
|
---|
[214] | 49 |
|
---|
[6967] | 50 | # find the classifier
|
---|
[14112] | 51 | my $customclassname = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "custom", $ENV{'GSDLCOLLECTION'},
|
---|
| 52 | "perllib", "classify", "${classifier}.pm");
|
---|
| 53 | my $colclassname = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "perllib", "classify", "${classifier}.pm");
|
---|
| 54 | my $mainclassname = &util::filename_cat($ENV{'GSDLHOME'}, "perllib", "classify", "${classifier}.pm");
|
---|
[6967] | 55 |
|
---|
[14112] | 56 | if (-e $customclassname) { require $customclassname; }
|
---|
| 57 | elsif (-e $colclassname) { require $colclassname; }
|
---|
[6967] | 58 | elsif (-e $mainclassname) { require $mainclassname; }
|
---|
| 59 | else {
|
---|
| 60 | &gsprintf(STDERR, "{classify.could_not_find_classifier}\n", $classifier) && die "\n";
|
---|
| 61 | }
|
---|
| 62 | my ($classobj);
|
---|
| 63 | my $options = "-gsdlinfo";
|
---|
[10218] | 64 | eval ("\$classobj = new \$classifier([],[$options])");
|
---|
[6967] | 65 | die "$@" if $@;
|
---|
| 66 |
|
---|
| 67 | return $classobj;
|
---|
| 68 | }
|
---|
| 69 |
|
---|
[811] | 70 | sub load_classifiers {
|
---|
[1839] | 71 | my ($classify_list, $build_dir, $outhandle) = @_;
|
---|
[811] | 72 | my @classify_objects = ();
|
---|
[8220] | 73 | my $classify_number = 1;
|
---|
[13933] | 74 |
|
---|
| 75 | my $colclassdir = &util::filename_cat($ENV{'GSDLCOLLECTDIR'},"perllib/classify");
|
---|
| 76 | unshift (@INC, $colclassdir);
|
---|
[811] | 77 |
|
---|
| 78 | foreach $classifyoption (@$classify_list) {
|
---|
[214] | 79 |
|
---|
[811] | 80 | # get the classifier name
|
---|
| 81 | my $classname = shift @$classifyoption;
|
---|
| 82 | next unless defined $classname;
|
---|
[1839] | 83 |
|
---|
[811] | 84 | # find the classifier
|
---|
[14112] | 85 | my $customclassname = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "custom", $ENV{'GSDLCOLLECTION'},
|
---|
| 86 | "perllib", "classify", "${classname}.pm");
|
---|
| 87 | my $colclassname = &util::filename_cat($ENV{'GSDLCOLLECTDIR'}, "perllib", "classify", "${classname}.pm");
|
---|
| 88 | my $mainclassname = &util::filename_cat($ENV{'GSDLHOME'}, "perllib", "classify", "${classname}.pm");
|
---|
[214] | 89 |
|
---|
[14112] | 90 | if (-e $customclassname) { require $customclassname; }
|
---|
| 91 | elsif (-e $colclassname) { require $colclassname; }
|
---|
[811] | 92 | elsif (-e $mainclassname) { require $mainclassname; }
|
---|
[5682] | 93 | else { &gsprintf(STDERR, "{classify.could_not_find_classifier}\n", $classname) && die "\n";
|
---|
| 94 | # die "ERROR - couldn't find classifier \"$classname\"\n";
|
---|
[14112] | 95 | }
|
---|
[214] | 96 |
|
---|
[811] | 97 | # create the classify object
|
---|
| 98 | my ($classobj);
|
---|
[1839] | 99 |
|
---|
| 100 | my @newoptions;
|
---|
[6964] | 101 |
|
---|
| 102 | # do these first so they can be overriden by user supplied options
|
---|
| 103 | push @newoptions, "-builddir", "$build_dir" if ($build_dir);
|
---|
| 104 | push @newoptions, "-outhandle", "$outhandle" if ($outhandle);
|
---|
| 105 | push @newoptions, "-verbosity", "2";
|
---|
| 106 |
|
---|
[6967] | 107 | # backwards compatability hack: if the classifier options are
|
---|
| 108 | # in "x=y" format, convert them to parsearg ("-x y") format.
|
---|
| 109 | my ($opt, $key, $value);
|
---|
[1839] | 110 | foreach $opt (@$classifyoption) {
|
---|
[11644] | 111 | # if ($opt =~ /^(\w+)=(.*)$/) {
|
---|
| 112 | # push @newoptions, "-$1", $2;
|
---|
| 113 | # } else {
|
---|
[1839] | 114 | push @newoptions, $opt;
|
---|
[11644] | 115 | #}
|
---|
[1839] | 116 | }
|
---|
| 117 |
|
---|
| 118 | map { $_ = "\"$_\""; } @newoptions;
|
---|
| 119 | my $options .= join (",", @newoptions);
|
---|
| 120 |
|
---|
[10218] | 121 |
|
---|
| 122 | eval ("\$classobj = new \$classname([],[$options])");
|
---|
[811] | 123 | die "$@" if $@;
|
---|
[1839] | 124 |
|
---|
[8220] | 125 | $classobj->set_number($classify_number);
|
---|
| 126 | $classify_number ++;
|
---|
| 127 |
|
---|
[1839] | 128 | # add this object to the list
|
---|
[811] | 129 | push (@classify_objects, $classobj);
|
---|
| 130 | }
|
---|
| 131 |
|
---|
[8220] | 132 | my ($classobj);
|
---|
| 133 | eval ("\$classobj = new AllList()");
|
---|
| 134 | die "$@" if $@;
|
---|
| 135 | push (@classify_objects, $classobj);
|
---|
| 136 |
|
---|
[811] | 137 | return \@classify_objects;
|
---|
[214] | 138 | }
|
---|
| 139 |
|
---|
| 140 | # init_classifiers resets all the classifiers and readys them to process
|
---|
[315] | 141 | # the documents.
|
---|
[214] | 142 | sub init_classifiers {
|
---|
| 143 | my ($classifiers) = @_;
|
---|
| 144 |
|
---|
| 145 | foreach $classobj (@$classifiers) {
|
---|
| 146 | $classobj->init();
|
---|
| 147 | }
|
---|
| 148 | }
|
---|
| 149 |
|
---|
[11994] | 150 |
|
---|
| 151 |
|
---|
| 152 | # takes a hashref containing the metadata for a gdbmfile entry, and extracts
|
---|
| 153 | # the childrens numbers (from the 'contains' entry).
|
---|
| 154 | # assumes format is ".1;".2;".3
|
---|
| 155 | sub get_children {
|
---|
| 156 | my ($doc_db_hash) = @_;
|
---|
| 157 |
|
---|
| 158 | my $children = undef;
|
---|
| 159 |
|
---|
| 160 | $childs = $doc_db_hash->{'contains'};
|
---|
| 161 | if (defined ($childs)) {
|
---|
| 162 | $childs =~ s/\@$//; #remove trailing @
|
---|
| 163 | $childs =~ s/^\"\.//; #remove initial ".
|
---|
| 164 | @$children = split /\;\"\./, $childs;
|
---|
| 165 |
|
---|
| 166 | }
|
---|
| 167 |
|
---|
| 168 | return $children;
|
---|
| 169 | }
|
---|
| 170 |
|
---|
| 171 |
|
---|
| 172 | sub recurse_sections {
|
---|
| 173 | my ($doc_obj, $children, $parentoid, $parentsection, $gdbm_recs) = @_;
|
---|
| 174 |
|
---|
| 175 | return if (!defined $children);
|
---|
| 176 |
|
---|
| 177 | foreach my $child (sort { $a <=> $b} @$children) {
|
---|
| 178 | $doc_obj->create_named_section("$parentsection.$child");
|
---|
| 179 | my $doc_db_rec = $gdbm_recs->{"$parentoid.$child"};
|
---|
| 180 | my $doc_db_hash = db_rec_to_hash($doc_db_rec);
|
---|
| 181 |
|
---|
| 182 | # get child's children
|
---|
| 183 | my $newchildren = &get_children($doc_db_hash);
|
---|
| 184 |
|
---|
| 185 | # add content for current section
|
---|
| 186 | add_section_content($doc_obj, "$parentsection.$child", $doc_db_hash);
|
---|
| 187 |
|
---|
| 188 | # process all the children if there are any
|
---|
| 189 | if (defined ($newchildren))
|
---|
| 190 | {
|
---|
| 191 | recurse_sections($doc_obj, $newchildren, "$parentoid.$child",
|
---|
| 192 | "$parentsection.$child", $gdbm_recs);
|
---|
| 193 | }
|
---|
| 194 | }
|
---|
| 195 | }
|
---|
| 196 |
|
---|
| 197 |
|
---|
| 198 | sub add_section_content {
|
---|
| 199 | my ($doc_obj, $cursection, $doc_db_hash) = @_;
|
---|
| 200 |
|
---|
| 201 | foreach $key (keys %$doc_db_hash) {
|
---|
| 202 | #don't need to store these metadata
|
---|
| 203 | next if $key =~ /(thistype|childtype|contains|docnum|doctype|classifytype)/i;
|
---|
| 204 | # but do want things like hastxt and archivedir
|
---|
| 205 | my @items = split /@/, $doc_db_hash->{$key};
|
---|
| 206 | map {$doc_obj->add_metadata ($cursection, $key, $_); } @items;
|
---|
| 207 |
|
---|
| 208 | }
|
---|
| 209 | }
|
---|
| 210 |
|
---|
| 211 |
|
---|
| 212 | # gets all the metadata from a gdbm file entry, and puts it into a hashref
|
---|
| 213 | sub db_rec_to_hash {
|
---|
| 214 |
|
---|
| 215 | my ($gdb_str_ref) = @_;
|
---|
| 216 |
|
---|
| 217 | my $hashref = {};
|
---|
| 218 |
|
---|
| 219 | my @entries = split(/\n/, $gdb_str_ref);
|
---|
| 220 | foreach $entry (@entries) {
|
---|
| 221 | my($key, $value) = ($entry =~ /^<([^>]*)>(.*?)$/ );
|
---|
| 222 | $hashref->{$key} .= '@' if defined $hashref->{$key};
|
---|
| 223 | $hashref->{$key} .= $value;
|
---|
| 224 |
|
---|
| 225 | }
|
---|
| 226 |
|
---|
| 227 | return $hashref;
|
---|
| 228 | }
|
---|
| 229 |
|
---|
| 230 |
|
---|
| 231 | sub reconstruct_doc_objs_metadata
|
---|
| 232 | {
|
---|
| 233 | my ($fulldbname) = @_;
|
---|
| 234 |
|
---|
[12559] | 235 | # tie %gdbm_recs, 'GDBM_File', $fulldbname, &GDBM_WRCREAT, 0640;
|
---|
[11994] | 236 |
|
---|
[12559] | 237 | my %gdbm_recs;
|
---|
| 238 | &unbuildutil::read_gdbm($fulldbname,\%gdbm_recs);
|
---|
| 239 |
|
---|
| 240 |
|
---|
[11994] | 241 | # dig out top level doc sections
|
---|
| 242 | my %top_sections = ();
|
---|
[13068] | 243 | my %top_docnums = ();
|
---|
[11994] | 244 | foreach my $key ( keys %gdbm_recs )
|
---|
| 245 | {
|
---|
| 246 | my $md_rec = $gdbm_recs{$key};
|
---|
| 247 | my $md_hash = db_rec_to_hash($md_rec);
|
---|
| 248 |
|
---|
| 249 | if ((defined $md_hash->{'doctype'}) && ($md_hash->{'doctype'} eq "doc")) {
|
---|
| 250 | next if ($key =~ m/\./);
|
---|
| 251 | $top_sections{$key} = $md_hash;
|
---|
[13068] | 252 | $top_docnums{$key} = $md_hash->{'docnum'};
|
---|
[11994] | 253 | }
|
---|
| 254 | }
|
---|
| 255 |
|
---|
| 256 | # for greenstone document objects based on metadata in gdbm file
|
---|
| 257 | my @all_docs = ();
|
---|
[13068] | 258 | # we need to make sure the documents were processed in the same order as
|
---|
| 259 | # before, so sort based on their docnums
|
---|
| 260 | foreach my $oid ( sort { $top_docnums{$a} <=> $top_docnums{$b} } keys %top_sections )
|
---|
[11994] | 261 | {
|
---|
| 262 | my $doc_db_hash = $top_sections{$oid};
|
---|
| 263 |
|
---|
| 264 | my $doc_obj = new doc();
|
---|
| 265 | $doc_obj->set_OID($oid);
|
---|
| 266 | my $top = $doc_obj->get_top_section();
|
---|
| 267 | add_section_content ($doc_obj, $top, $doc_db_hash);
|
---|
| 268 | my $children = &get_children($doc_db_hash);
|
---|
| 269 | recurse_sections($doc_obj, $children, $oid, $top, \%gdbm_recs);
|
---|
| 270 |
|
---|
| 271 | push(@all_docs,$doc_obj);
|
---|
| 272 | }
|
---|
| 273 |
|
---|
[12559] | 274 | # untie %gdbm_recs;
|
---|
[11994] | 275 |
|
---|
| 276 | return \@all_docs;
|
---|
| 277 | }
|
---|
| 278 |
|
---|
| 279 |
|
---|
| 280 |
|
---|
| 281 |
|
---|
| 282 |
|
---|
[214] | 283 | # classify_doc lets each of the classifiers classify a document
|
---|
| 284 | sub classify_doc {
|
---|
| 285 | my ($classifiers, $doc_obj) = @_;
|
---|
| 286 |
|
---|
| 287 | foreach $classobj (@$classifiers) {
|
---|
[8220] | 288 | my $title = $classobj->{'title'};
|
---|
[214] | 289 | $classobj->classify($doc_obj);
|
---|
| 290 | }
|
---|
| 291 | }
|
---|
| 292 |
|
---|
| 293 | # output_classify_info outputs all the info needed for the classification
|
---|
| 294 | # to the gdbm
|
---|
| 295 | sub output_classify_info {
|
---|
[8361] | 296 | my ($classifiers, $handle, $remove_empty_classifications, $gli) = @_;
|
---|
[214] | 297 | # $handle = "main::STDOUT";
|
---|
| 298 |
|
---|
[6332] | 299 | $gli = 0 unless defined $gli;
|
---|
| 300 |
|
---|
[315] | 301 | # create a classification containing all the info
|
---|
| 302 | my $classifyinfo = {'classifyOID'=>'browse',
|
---|
| 303 | 'contains'=>[]};
|
---|
| 304 |
|
---|
| 305 | # get each of the classifications
|
---|
[8275] | 306 | foreach $classobj (@$classifiers) {
|
---|
[6332] | 307 | my $tempinfo = $classobj->get_classify_info($gli);
|
---|
[8220] | 308 | my $classID = $tempinfo->{'classifyOID'};
|
---|
| 309 |
|
---|
| 310 | $tempinfo->{'classifyOID'} = "CL$next_classify_num" unless defined($tempinfo->{'classifyOID'});
|
---|
[315] | 311 | $next_classify_num++;
|
---|
[12844] | 312 |
|
---|
| 313 | print STDERR "*** outputting information for classifier: $tempinfo->{'classifyOID'}\n";
|
---|
| 314 |
|
---|
[315] | 315 | push (@{$classifyinfo->{'contains'}}, $tempinfo);
|
---|
[214] | 316 | }
|
---|
| 317 |
|
---|
[8361] | 318 | &print_classify_info ($handle, $classifyinfo, "", $remove_empty_classifications);
|
---|
[214] | 319 | }
|
---|
| 320 |
|
---|
[315] | 321 | sub print_classify_info {
|
---|
[8361] | 322 | my ($handle, $classifyinfo, $OID, $remove_empty_classifications) = @_;
|
---|
[831] | 323 |
|
---|
[315] | 324 | $OID =~ s/^\.+//; # just for good luck
|
---|
[214] | 325 |
|
---|
[315] | 326 | # book information is printed elsewhere
|
---|
| 327 | return if (defined ($classifyinfo->{'OID'}));
|
---|
| 328 |
|
---|
| 329 | # don't want empty classifications
|
---|
[8445] | 330 | return if (&check_contents ($classifyinfo, $remove_empty_classifications) == 0 && $remove_empty_classifications);
|
---|
[315] | 331 |
|
---|
[8361] | 332 | $OID = $classifyinfo->{'classifyOID'} if defined ($classifyinfo->{'classifyOID'});
|
---|
[315] | 333 |
|
---|
[8361] | 334 | my $outputtext = "[$OID]\n";
|
---|
| 335 | $outputtext .= "<doctype>classify\n";
|
---|
| 336 | $outputtext .= "<hastxt>0\n";
|
---|
| 337 | $outputtext .= "<childtype>$classifyinfo->{'childtype'}\n"
|
---|
| 338 | if defined $classifyinfo->{'childtype'};
|
---|
| 339 | $outputtext .= "<Title>$classifyinfo->{'Title'}\n"
|
---|
| 340 | if defined $classifyinfo->{'Title'};
|
---|
| 341 | $outputtext .= "<numleafdocs>$classifyinfo->{'numleafdocs'}\n"
|
---|
| 342 | if defined $classifyinfo->{'numleafdocs'};
|
---|
| 343 | $outputtext .= "<thistype>$classifyinfo->{'thistype'}\n"
|
---|
| 344 | if defined $classifyinfo->{'thistype'};
|
---|
| 345 | $outputtext .= "<parameters>$classifyinfo->{'parameters'}\n"
|
---|
| 346 | if defined $classifyinfo->{'parameters'};
|
---|
| 347 | $outputtext .= "<supportsmemberof>$classifyinfo->{'supportsmemberof'}\n"
|
---|
| 348 | if defined $classifyinfo->{'supportsmemberof'};
|
---|
| 349 |
|
---|
| 350 | my $contains_text = "<contains>";
|
---|
| 351 | my $mdoffset_text = "<mdoffset>";
|
---|
| 352 |
|
---|
| 353 | my $next_subOID = 1;
|
---|
| 354 | my $first = 1;
|
---|
| 355 | foreach $tempinfo (@{$classifyinfo->{'contains'}}) {
|
---|
| 356 | # empty contents were made undefined by clean_contents()
|
---|
| 357 | next unless defined $tempinfo;
|
---|
[315] | 358 |
|
---|
[8361] | 359 | if (!defined ($tempinfo->{'classifyOID'}) ||
|
---|
| 360 | $tempinfo->{'classifyOID'} ne "oai") {
|
---|
| 361 | $contains_text .= ";" unless $first;
|
---|
| 362 | }
|
---|
| 363 | $mdoffset_text .= ";" unless $first;
|
---|
| 364 | $first = 0;
|
---|
[315] | 365 |
|
---|
[8361] | 366 | if (defined ($tempinfo->{'classifyOID'})) {
|
---|
| 367 | if ($tempinfo->{'classifyOID'} ne "oai") {
|
---|
| 368 | $contains_text .= $tempinfo->{'classifyOID'};
|
---|
[8275] | 369 | }
|
---|
[12844] | 370 |
|
---|
| 371 | # Extra code for incremental building.
|
---|
| 372 | # We need to store a listing of the classifiers each DOI is in
|
---|
| 373 | my $clids = [];
|
---|
| 374 | #rint STDERR "==1. Recording reverse lookup for " . $tempinfo->{'classifyOID'} . "==\n";
|
---|
| 375 | if(defined($oid_to_clids->{$tempinfo->{'classifyOID'}})) {
|
---|
| 376 | #rint STDERR "Found existing array!\n";
|
---|
| 377 | $clids = $oid_to_clids->{$tempinfo->{'classifyOID'}};
|
---|
| 378 | }
|
---|
| 379 | #rint STDERR "Appended $OID to \"" . join(";", @{$clids}) . "\"\n";
|
---|
| 380 | push(@{$clids}, $OID);
|
---|
| 381 | $oid_to_clids->{$tempinfo->{'classifyOID'}} = $clids;
|
---|
| 382 | #rint STDERR "Result: \"" . join(";", @{$clids}) . "\"\n";
|
---|
| 383 |
|
---|
[8361] | 384 | &print_classify_info ($handle, $tempinfo, $tempinfo->{'classifyOID'},
|
---|
| 385 | $remove_empty_classifications);
|
---|
| 386 | } elsif (defined ($tempinfo->{'OID'})) {
|
---|
| 387 | $contains_text .= $tempinfo->{'OID'};
|
---|
[12844] | 388 | $mdoffset_text .= $tempinfo->{'offset'} if (defined ($tempinfo->{'offset'}));
|
---|
| 389 |
|
---|
| 390 |
|
---|
| 391 | # note: we don't want to print the contents of the books
|
---|
| 392 | # Extra code for incremental building.
|
---|
| 393 | # We need to store a listing of the classifiers each DOI is in
|
---|
| 394 | my $clids = [];
|
---|
| 395 | #rint STDERR "==2. Recording reverse lookup for " . $tempinfo->{'OID'} . "==\n";
|
---|
| 396 | if(defined($oid_to_clids->{$tempinfo->{'OID'}})) {
|
---|
| 397 | #rint STDERR "Found existing array!\n";
|
---|
| 398 | $clids = $oid_to_clids->{$tempinfo->{'OID'}};
|
---|
| 399 | }
|
---|
| 400 | #rint STDERR "Appended $OID to \"" . join(";", @{$clids}) . "\"\n";
|
---|
| 401 | push(@{$clids}, $OID);
|
---|
| 402 | $oid_to_clids->{$tempinfo->{'OID'}} = $clids;
|
---|
| 403 | #rint STDERR "Result: \"" . join(";", @{$clids}) . "\"\n";
|
---|
| 404 |
|
---|
| 405 |
|
---|
[315] | 406 | } else {
|
---|
[7346] | 407 |
|
---|
| 408 | # Supress having top-level node in Collage classifier
|
---|
| 409 | # so no bookshelf icon appears, top-level, along with the
|
---|
| 410 | # applet
|
---|
[8361] | 411 |
|
---|
[7346] | 412 | if (!defined ($tempinfo->{'Title'}) || $tempinfo->{'Title'} ne "Collage") {
|
---|
| 413 | $contains_text .= "\".$next_subOID";
|
---|
| 414 | }
|
---|
[12844] | 415 |
|
---|
| 416 | # Extra code for incremental building.
|
---|
| 417 | # We need to store a listing of the classifiers each DOI is in
|
---|
| 418 | my $clids = [];
|
---|
| 419 | #rint STDERR "==3. Recording reverse lookup for $OID.$next_subOID==\n";
|
---|
| 420 | if(defined($oid_to_clids->{$OID . "." . $next_subOID})) {
|
---|
| 421 | #rint STDERR "Found existing array!\n";
|
---|
| 422 | $clids = $oid_to_clids->{$OID . "." . $next_subOID};
|
---|
| 423 | }
|
---|
| 424 | #rint STDERR "Appended $OID to \"" . join(";", @{$clids}) . "\"\n";
|
---|
| 425 | push(@{$clids}, $OID);
|
---|
| 426 | $oid_to_clids->{$OID . "." . $next_subOID} = $clids;
|
---|
| 427 | #rint STDERR "Result: \"" . join(";", @{$clids}) . "\"\n";
|
---|
[8361] | 428 |
|
---|
[315] | 429 | &print_classify_info ($handle, $tempinfo, "$OID.$next_subOID",
|
---|
[8361] | 430 | $remove_empty_classifications);
|
---|
[315] | 431 | $next_subOID++;
|
---|
| 432 | }
|
---|
| 433 | }
|
---|
[8361] | 434 |
|
---|
| 435 | $outputtext .= "$contains_text\n";
|
---|
| 436 | $outputtext .= "<mdtype>$classifyinfo->{'mdtype'}\n"
|
---|
| 437 | if defined $classifyinfo->{'mdtype'};
|
---|
| 438 | $outputtext .= "$mdoffset_text\n"
|
---|
| 439 | if ($mdoffset_text !~ m/^<mdoffset>;+$/);
|
---|
| 440 |
|
---|
| 441 | $outputtext .= '-' x 70 . "\n";
|
---|
| 442 |
|
---|
| 443 | print $handle $outputtext;
|
---|
| 444 |
|
---|
[315] | 445 | }
|
---|
| 446 |
|
---|
[637] | 447 | sub check_contents {
|
---|
[8445] | 448 | my ($classifyinfo,$remove_empty_classifications) = @_;
|
---|
| 449 | $remove_empty_classifications = 0 unless ($remove_empty_classifications);
|
---|
[637] | 450 | my $num_leaf_docs = 0;
|
---|
| 451 | my $sub_num_leaf_docs = 0;
|
---|
[315] | 452 |
|
---|
[637] | 453 | return $classifyinfo->{'numleafdocs'} if (defined $classifyinfo->{'numleafdocs'});
|
---|
| 454 |
|
---|
[315] | 455 | foreach $content (@{$classifyinfo->{'contains'}}) {
|
---|
| 456 | if (defined $content->{'OID'}) {
|
---|
| 457 | # found a book
|
---|
[637] | 458 | $num_leaf_docs ++;
|
---|
[9790] | 459 | } elsif (($sub_num_leaf_docs = &check_contents ($content,$remove_empty_classifications)) > 0) {
|
---|
[315] | 460 | # there's a book somewhere below
|
---|
[637] | 461 | $num_leaf_docs += $sub_num_leaf_docs;
|
---|
[315] | 462 | } else {
|
---|
[8445] | 463 | if ($remove_empty_classifications){
|
---|
| 464 | # section contains no books so we want to remove
|
---|
| 465 | # it from its parents contents
|
---|
| 466 | $content = undef;
|
---|
| 467 | }
|
---|
[315] | 468 | }
|
---|
| 469 | }
|
---|
[637] | 470 |
|
---|
| 471 | $classifyinfo->{'numleafdocs'} = $num_leaf_docs;
|
---|
| 472 | return $num_leaf_docs;
|
---|
[315] | 473 | }
|
---|
| 474 |
|
---|
[214] | 475 | 1;
|
---|