1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # List.pm -- A general and flexible list classifier with most of
|
---|
4 | # the abilities of AZCompactList, and better Unicode,
|
---|
5 | # metadata and sorting capabilities.
|
---|
6 | #
|
---|
7 | # A component of the Greenstone digital library software
|
---|
8 | # from the New Zealand Digital Library Project at the
|
---|
9 | # University of Waikato, New Zealand.
|
---|
10 | #
|
---|
11 | # Author: Michael Dewsnip, NZDL Project, University of Waikato, NZ
|
---|
12 | #
|
---|
13 | # Copyright (C) 2005 New Zealand Digital Library Project
|
---|
14 | #
|
---|
15 | # This program is free software; you can redistribute it and/or modify
|
---|
16 | # it under the terms of the GNU General Public License as published by
|
---|
17 | # the Free Software Foundation; either version 2 of the License, or
|
---|
18 | # (at your option) any later version.
|
---|
19 | #
|
---|
20 | # This program is distributed in the hope that it will be useful,
|
---|
21 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
22 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
23 | # GNU General Public License for more details.
|
---|
24 | #
|
---|
25 | # You should have received a copy of the GNU General Public License
|
---|
26 | # along with this program; if not, write to the Free Software
|
---|
27 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
28 | #
|
---|
29 | # TO DO: - Remove punctuation from metadata values before sorting.
|
---|
30 | # - Add an AZCompactList-style hlist option?
|
---|
31 | #
|
---|
32 | ###########################################################################
|
---|
33 |
|
---|
34 |
|
---|
35 | package List;
|
---|
36 |
|
---|
37 |
|
---|
38 | use BaseClassifier;
|
---|
39 |
|
---|
40 | use strict;
|
---|
41 |
|
---|
42 |
|
---|
43 | sub BEGIN {
|
---|
44 | @List::ISA = ('BaseClassifier');
|
---|
45 | }
|
---|
46 |
|
---|
47 | my $meta_select_type_list =
|
---|
48 | [
|
---|
49 | { 'name' => "firstvalue",
|
---|
50 | 'desc' => "{List.metadata_selection.firstvalue}"},
|
---|
51 | { 'name' => "firstvalidmetadata",
|
---|
52 | 'desc' => "{List.metadata_selection.firstvalidmetadata}"},
|
---|
53 | { 'name' => "allvalues",
|
---|
54 | 'desc' => "{List.metadata_selection.allvalues}"} ];
|
---|
55 | my $valid_meta_select_types = { 'firstvalue' => 1,
|
---|
56 | 'firstvalidmetadata' => 1,
|
---|
57 | 'allvalues' => 1 };
|
---|
58 | my $partition_type_list =
|
---|
59 | [ { 'name' => "per_letter",
|
---|
60 | 'desc' => "{List.level_partition.per_letter}" },
|
---|
61 | { 'name' => "approximate_size",
|
---|
62 | 'desc' => "{List.level_partition.approximate_size}"},
|
---|
63 | { 'name' => "constant_size",
|
---|
64 | 'desc' => "{List.level_partition.constant_size}" },
|
---|
65 | { 'name' => "none",
|
---|
66 | 'desc' => "{List.level_partition.none}" } ];
|
---|
67 |
|
---|
68 | # following used to check types later on
|
---|
69 | my $valid_partition_types = { 'per_letter' => 1,
|
---|
70 | 'constant_size' => 1,
|
---|
71 | 'per_letter_fixed_size' => 1,
|
---|
72 | 'approximate_size' => 1,
|
---|
73 | 'none' => 1};
|
---|
74 |
|
---|
75 | my $bookshelf_type_list =
|
---|
76 | [ { 'name' => "always",
|
---|
77 | 'desc' => "{List.bookshelf_type.always}" },
|
---|
78 | { 'name' => "duplicate_only",
|
---|
79 | 'desc' => "{List.bookshelf_type.duplicate_only}" },
|
---|
80 | { 'name' => "never",
|
---|
81 | 'desc' => "{List.bookshelf_type.never}" } ];
|
---|
82 |
|
---|
83 | my $arguments =
|
---|
84 | [ { 'name' => "metadata",
|
---|
85 | 'desc' => "{List.metadata}",
|
---|
86 | 'type' => "metadata",
|
---|
87 | 'reqd' => "yes" },
|
---|
88 | { 'name' => "metadata_selection_mode",
|
---|
89 | 'desc' => "{List.metadata_selection_mode}",
|
---|
90 | 'type' => "enumstring", # Must be enumstring because multiple values can be specified (separated by '/')
|
---|
91 | 'list' => $meta_select_type_list,
|
---|
92 | 'deft' => "firstvalidmetadata" },
|
---|
93 | # The interesting options
|
---|
94 | { 'name' => "bookshelf_type",
|
---|
95 | 'desc' => "{List.bookshelf_type}",
|
---|
96 | 'type' => "enum",
|
---|
97 | 'list' => $bookshelf_type_list,
|
---|
98 | 'deft' => "never" },
|
---|
99 | { 'name' => "classify_sections",
|
---|
100 | 'desc' => "{List.classify_sections}",
|
---|
101 | 'type' => "flag" },
|
---|
102 | { 'name' => "partition_type_within_level",
|
---|
103 | 'desc' => "{List.partition_type_within_level}",
|
---|
104 | 'type' => "enumstring", # Must be enumstring because multiple values can be specified (separated by '/')
|
---|
105 | 'list' => $partition_type_list,
|
---|
106 | 'deft' => "per_letter" },
|
---|
107 | { 'name' => "partition_size_within_level",
|
---|
108 | 'desc' => "{List.partition_size_within_level}",
|
---|
109 | 'type' => "string" }, # Must be string because multiple values can be specified (separated by '/')
|
---|
110 | { 'name' => "partition_name_length",
|
---|
111 | 'desc' => "{List.partition_name_length}",
|
---|
112 | 'type' => "string" },
|
---|
113 | { 'name' => "sort_leaf_nodes_using",
|
---|
114 | 'desc' => "{List.sort_leaf_nodes_using}",
|
---|
115 | 'type' => "metadata",
|
---|
116 | 'deft' => "Title" },
|
---|
117 | { 'name' => "reverse_sort_leaf_nodes",
|
---|
118 | 'desc' => "{List.reverse_sort_leaf_nodes}",
|
---|
119 | 'type' => "flag"},
|
---|
120 | { 'name' => "sort_using_unicode_collation",
|
---|
121 | 'desc' => "{List.sort_using_unicode_collation}",
|
---|
122 | 'type' => "flag" },
|
---|
123 | { 'name' => "use_hlist_for",
|
---|
124 | 'desc' => "{List.use_hlist_for}",
|
---|
125 | 'type' => "string" },
|
---|
126 | {'name' => "filter_metadata",
|
---|
127 | 'desc' => "{List.filter_metadata}",
|
---|
128 | 'type' => "metadata"},
|
---|
129 | {'name' => "filter_regex",
|
---|
130 | 'desc' => "{List.filter_regex}",
|
---|
131 | 'type' => "regexp"},
|
---|
132 | { 'name' => "standardize_capitalization",
|
---|
133 | 'desc' => "{List.standardize_capitalization}",
|
---|
134 | 'type' => "flag"},
|
---|
135 | { 'name' => "removeprefix",
|
---|
136 | 'desc' => "{BasClas.removeprefix}",
|
---|
137 | 'type' => "regexp" },
|
---|
138 | { 'name' => "removesuffix",
|
---|
139 | 'desc' => "{BasClas.removesuffix}",
|
---|
140 | 'type' => "regexp" } ];
|
---|
141 |
|
---|
142 | my $options = { 'name' => "List",
|
---|
143 | 'desc' => "{List.desc}",
|
---|
144 | 'abstract' => "no",
|
---|
145 | 'inherits' => "yes",
|
---|
146 | 'args' => $arguments };
|
---|
147 |
|
---|
148 |
|
---|
149 | sub new
|
---|
150 | {
|
---|
151 | my ($class) = shift(@_);
|
---|
152 | my ($classifierslist, $inputargs, $hashArgOptLists) = @_;
|
---|
153 | push(@$classifierslist, $class);
|
---|
154 |
|
---|
155 | push(@{$hashArgOptLists->{"ArgList"}}, @{$arguments});
|
---|
156 | push(@{$hashArgOptLists->{"OptList"}}, $options);
|
---|
157 |
|
---|
158 | my $self = new BaseClassifier($classifierslist, $inputargs, $hashArgOptLists);
|
---|
159 |
|
---|
160 | if ($self->{'info_only'}) {
|
---|
161 | # don't worry about any options etc
|
---|
162 | return bless $self, $class;
|
---|
163 | }
|
---|
164 |
|
---|
165 | # The metadata elements to use (required)
|
---|
166 | if (!$self->{'metadata'}) {
|
---|
167 | die "Error: No metadata fields specified for List.\n";
|
---|
168 | }
|
---|
169 |
|
---|
170 | my @metadata_groups = split(/[\/]/, $self->{'metadata'});
|
---|
171 | $self->{'metadata_groups'} = \@metadata_groups;
|
---|
172 |
|
---|
173 | # The classifier button name (default: the first metadata element specified)
|
---|
174 | if (!$self->{'buttonname'}) {
|
---|
175 | my $first_metadata_group = $metadata_groups[0];
|
---|
176 | my $first_metadata_element = (split(/[\;|,\/]/, $first_metadata_group))[0];
|
---|
177 | $self->{'buttonname'} = $self->generate_title_from_metadata($first_metadata_element);
|
---|
178 | }
|
---|
179 |
|
---|
180 | # meta selection mode for each level
|
---|
181 | if (!$self->{'metadata_selection_mode'}) {
|
---|
182 | foreach my $metadata_group (@metadata_groups) {
|
---|
183 | $self->{$metadata_group . ".metadata_selection_mode"} = "firstvalidmetadata";
|
---|
184 | }
|
---|
185 | } else {
|
---|
186 | my @metadata_selection_mode_list = split(/\//, $self->{'metadata_selection_mode'});
|
---|
187 | foreach my $metadata_group (@metadata_groups) {
|
---|
188 | my $meta_select_mode = shift(@metadata_selection_mode_list);
|
---|
189 | if (defined($meta_select_mode) && defined $valid_meta_select_types->{$meta_select_mode}) {
|
---|
190 | $self->{$metadata_group . ".metadata_selection_mode"} = $meta_select_mode;
|
---|
191 | } else {
|
---|
192 | $self->{$metadata_group . ".metadata_selection_mode"} = "firstvalidmetadata";
|
---|
193 | }
|
---|
194 | }
|
---|
195 | }
|
---|
196 | # Whether to group items into a bookshelf, (must be 'always' for all metadata fields except the last)
|
---|
197 | foreach my $metadata_group (@metadata_groups) {
|
---|
198 | $self->{$metadata_group . ".bookshelf_type"} = "always";
|
---|
199 | }
|
---|
200 | my $last_metadata_group = $metadata_groups[$#metadata_groups];
|
---|
201 | # Default: duplicate_only, ie. leave leaf nodes ungrouped (equivalent to AZCompactList -mingroup 2)
|
---|
202 | $self->{$last_metadata_group . ".bookshelf_type"} = $self->{'bookshelf_type'};
|
---|
203 |
|
---|
204 | # Whether to use an hlist or a vlist for each level in the hierarchy (default: vlist)
|
---|
205 | foreach my $metadata_group (@metadata_groups) {
|
---|
206 | $self->{$metadata_group . ".list_type"} = "VList";
|
---|
207 | }
|
---|
208 | foreach my $metadata_group (split(/\,/, $self->{'use_hlist_for'})) {
|
---|
209 | $self->{$metadata_group . ".list_type"} = "HList";
|
---|
210 | }
|
---|
211 |
|
---|
212 | # How the items are grouped into partitions (default: no partition)
|
---|
213 | # for each level (metadata group), separated by '/'
|
---|
214 | if (!$self->{"partition_type_within_level"}) {
|
---|
215 | foreach my $metadata_group (@metadata_groups) {
|
---|
216 | $self->{$metadata_group . ".partition_type_within_level"} = "none";
|
---|
217 | }
|
---|
218 | } else {
|
---|
219 | my @partition_type_within_levellist = split(/\//, $self->{'partition_type_within_level'});
|
---|
220 |
|
---|
221 | my $first = 1;
|
---|
222 | foreach my $metadata_group (@metadata_groups) {
|
---|
223 | my $partition_type_within_levelelem = shift(@partition_type_within_levellist);
|
---|
224 | if (defined($partition_type_within_levelelem) && $partition_type_within_levelelem eq "per_letter_fixed_size") {
|
---|
225 | print STDERR "per letter fixed size, changing to approximate size\n";
|
---|
226 | $partition_type_within_levelelem = "approximate_size";
|
---|
227 | }
|
---|
228 | if (defined($partition_type_within_levelelem) && defined $valid_partition_types->{$partition_type_within_levelelem}) {
|
---|
229 | $self->{$metadata_group . ".partition_type_within_level"} = $partition_type_within_levelelem;
|
---|
230 | }
|
---|
231 | else {
|
---|
232 | if ($first) {
|
---|
233 | $self->{$metadata_group . ".partition_type_within_level"} = "none";
|
---|
234 | $first = 0;
|
---|
235 | } else {
|
---|
236 | $self->{$metadata_group . ".partition_type_within_level"} = $self->{$metadata_groups[0] . ".partition_type_within_level"};
|
---|
237 | }
|
---|
238 | if (defined($partition_type_within_levelelem)) {
|
---|
239 | # ie invalid entry
|
---|
240 | print STDERR "invalid partition type for level $metadata_group: $partition_type_within_levelelem, defaulting to ". $self->{$metadata_group . ".partition_type_within_level"} ."\n";
|
---|
241 | }
|
---|
242 | }
|
---|
243 | }
|
---|
244 | }
|
---|
245 |
|
---|
246 | # The number of items in each partition
|
---|
247 | if (!$self->{'partition_size_within_level'}) {
|
---|
248 | # Default: 20
|
---|
249 | foreach my $metadata_group (@metadata_groups) {
|
---|
250 | $self->{$metadata_group . ".partition_size_within_level"} = 20;
|
---|
251 | }
|
---|
252 | }
|
---|
253 | else {
|
---|
254 | my @partition_size_within_levellist = split(/\//, $self->{'partition_size_within_level'});
|
---|
255 |
|
---|
256 | # Assign values based on the partition_size_within_level parameter
|
---|
257 | foreach my $metadata_group (@metadata_groups) {
|
---|
258 | my $partition_size_within_levelelem = shift(@partition_size_within_levellist);
|
---|
259 | if (defined($partition_size_within_levelelem)) {
|
---|
260 | $self->{$metadata_group . ".partition_size_within_level"} = $partition_size_within_levelelem;
|
---|
261 | }
|
---|
262 | else {
|
---|
263 | $self->{$metadata_group . ".partition_size_within_level"} = $self->{$metadata_groups[0] . ".partition_size_within_level"};
|
---|
264 | }
|
---|
265 | }
|
---|
266 | }
|
---|
267 |
|
---|
268 | # The removeprefix and removesuffix expressions
|
---|
269 | if ($self->{'removeprefix'}) {
|
---|
270 | # If there are more than one expressions, use '' to quote each experession and '/' to separate
|
---|
271 | my @removeprefix_exprs_within_levellist = split(/'\/'/, $self->{'removeprefix'});
|
---|
272 |
|
---|
273 | foreach my $metadata_group (@metadata_groups) {
|
---|
274 | my $removeprefix_expr_within_levelelem = shift(@removeprefix_exprs_within_levellist);
|
---|
275 | if (defined($removeprefix_expr_within_levelelem) && $removeprefix_expr_within_levelelem ne "") {
|
---|
276 | # Remove the other ' at the beginning and the end if there is any
|
---|
277 | $removeprefix_expr_within_levelelem =~ s/^'//;
|
---|
278 | $removeprefix_expr_within_levelelem =~ s/'$//;
|
---|
279 | # Remove the extra ^ at the beginning
|
---|
280 | $removeprefix_expr_within_levelelem =~ s/^\^//;
|
---|
281 | $self->{$metadata_group . ".remove_prefix_expr"} = $removeprefix_expr_within_levelelem;
|
---|
282 | } else {
|
---|
283 | $self->{$metadata_group . ".remove_prefix_expr"} = $self->{$metadata_groups[0] . ".remove_prefix_expr"};
|
---|
284 | }
|
---|
285 | }
|
---|
286 | }
|
---|
287 | if ($self->{'removesuffix'}) {
|
---|
288 | my @removesuffix_exprs_within_levellist = split(/'\/'/, $self->{'removesuffix'});
|
---|
289 |
|
---|
290 | foreach my $metadata_group (@metadata_groups) {
|
---|
291 | my $removesuffix_expr_within_levelelem = shift(@removesuffix_exprs_within_levellist);
|
---|
292 | if (defined($removesuffix_expr_within_levelelem) && $removesuffix_expr_within_levelelem ne "") {
|
---|
293 | $removesuffix_expr_within_levelelem =~ s/^'//;
|
---|
294 | $removesuffix_expr_within_levelelem =~ s/'$//;
|
---|
295 | # Remove the extra $ at the end
|
---|
296 | $removesuffix_expr_within_levelelem =~ s/\$$//;
|
---|
297 | $self->{$metadata_group . ".remove_suffix_expr"} = $removesuffix_expr_within_levelelem;
|
---|
298 | } else {
|
---|
299 | $self->{$metadata_group . ".remove_suffix_expr"} = $self->{$metadata_groups[0] . ".remove_suffix_expr"};
|
---|
300 | }
|
---|
301 | }
|
---|
302 | }
|
---|
303 |
|
---|
304 | # The metadata elements to use to sort the leaf nodes (default: Title)
|
---|
305 | my @sort_leaf_nodes_using_metadata_groups = ( "Title" );
|
---|
306 | if ($self->{'sort_leaf_nodes_using'}) {
|
---|
307 | @sort_leaf_nodes_using_metadata_groups = split(/\|/, $self->{'sort_leaf_nodes_using'});
|
---|
308 | }
|
---|
309 | $self->{'sort_leaf_nodes_using_metadata_groups'} = \@sort_leaf_nodes_using_metadata_groups;
|
---|
310 | foreach my $sort_group (@sort_leaf_nodes_using_metadata_groups) {
|
---|
311 | # set metadata_select_type, if not already set - might be already set if the same group was used in -metadata
|
---|
312 | if (!defined $self->{$sort_group . ".metadata_selection_mode"}) {
|
---|
313 | $self->{$sort_group . ".metadata_selection_mode"} = "firstvalue";
|
---|
314 | }
|
---|
315 | }
|
---|
316 | # Create an instance of the Unicode::Collate object if better Unicode sorting is desired
|
---|
317 | if ($self->{'sort_using_unicode_collation'}) {
|
---|
318 | # To use this you first need to download the allkeys.txt file from
|
---|
319 | # http://www.unicode.org/Public/UCA/latest/allkeys.txt and put it in the Perl
|
---|
320 | # Unicode/Collate directory.
|
---|
321 | require Unicode::Collate;
|
---|
322 | $self->{'unicode_collator'} = Unicode::Collate->new();
|
---|
323 | }
|
---|
324 |
|
---|
325 | # An empty array for the document/section OIDs that we are classifying
|
---|
326 | $self->{'OIDs'} = [];
|
---|
327 | # A hash for all the doc ids that we have seen, so we don't classify something twice
|
---|
328 | $self->{'all_doc_OIDs'} = {};
|
---|
329 | return bless $self, $class;
|
---|
330 | }
|
---|
331 |
|
---|
332 |
|
---|
333 | sub init
|
---|
334 | {
|
---|
335 | # Nothing to do...
|
---|
336 | }
|
---|
337 |
|
---|
338 |
|
---|
339 | # Called for each document in the collection
|
---|
340 | sub classify
|
---|
341 | {
|
---|
342 | my $self = shift(@_);
|
---|
343 | my ($doc_obj) = @_;
|
---|
344 |
|
---|
345 | if (defined $self->{'all_doc_OIDs'}->{$doc_obj->get_OID()}) {
|
---|
346 | print STDERR "Warning, List classifier has already seen document ".$doc_obj->get_OID().", not classifying again\n";
|
---|
347 | return;
|
---|
348 | }
|
---|
349 | $self->{'all_doc_OIDs'}->{$doc_obj->get_OID()} = 1;
|
---|
350 | # check against filter here
|
---|
351 | if ($self->{'filter_metadata'}) {
|
---|
352 | #print STDERR "filtering documents on $self->{'filter_metadata'}\n";
|
---|
353 | my $meta = $doc_obj->get_metadata_element($doc_obj->get_top_section(), $self->{'filter_metadata'});
|
---|
354 | return unless defined $meta;
|
---|
355 | if ($self->{'filter_regex'} ne "" && $meta !~ /$self->{'filter_regex'}/) {
|
---|
356 | #print STDERR "doesn't match regex\n";
|
---|
357 | return;
|
---|
358 |
|
---|
359 | }
|
---|
360 | }
|
---|
361 | # if we get here, we have passed the test for filtering
|
---|
362 | # If "-classify_sections" is set, classify every section of the document
|
---|
363 | if ($self->{'classify_sections'}) {
|
---|
364 | my $section = $doc_obj->get_next_section($doc_obj->get_top_section());
|
---|
365 | while (defined $section) {
|
---|
366 | $self->classify_section($doc_obj, $doc_obj->get_OID() . ".$section", $section);
|
---|
367 | $section = $doc_obj->get_next_section($section);
|
---|
368 | }
|
---|
369 | }
|
---|
370 | # Otherwise just classify the top document section
|
---|
371 | else {
|
---|
372 | $self->classify_section($doc_obj, $doc_obj->get_OID(), $doc_obj->get_top_section());
|
---|
373 | }
|
---|
374 |
|
---|
375 | }
|
---|
376 |
|
---|
377 | sub classify_section
|
---|
378 | {
|
---|
379 | my $self = shift(@_);
|
---|
380 | my ($doc_obj,$section_OID,$section) = @_;
|
---|
381 |
|
---|
382 | my @metadata_groups = @{$self->{'metadata_groups'}};
|
---|
383 |
|
---|
384 |
|
---|
385 | # Only classify the section if it has a value for one of the metadata elements in the first group
|
---|
386 | my $classify_section = 0;
|
---|
387 | my $first_metadata_group = $metadata_groups[0];
|
---|
388 | my $remove_prefix_expr = $self->{$first_metadata_group . ".remove_prefix_expr"};
|
---|
389 | my $remove_suffix_expr = $self->{$first_metadata_group . ".remove_suffix_expr"};
|
---|
390 | foreach my $first_metadata_group_element (split(/\;|,/, $first_metadata_group)) {
|
---|
391 | my $real_first_metadata_group_element = $self->strip_ex_from_metadata($first_metadata_group_element);
|
---|
392 | my $first_metadata_group_element_value = $doc_obj->get_metadata_element($section, $real_first_metadata_group_element);
|
---|
393 |
|
---|
394 | # Remove prefix/suffix if requested
|
---|
395 | if (defined ($first_metadata_group_element_value)) {
|
---|
396 | if (defined $remove_prefix_expr && $remove_prefix_expr ne "") {
|
---|
397 | $first_metadata_group_element_value =~ s/^$remove_prefix_expr//;
|
---|
398 | }
|
---|
399 |
|
---|
400 | if (defined $remove_suffix_expr && $remove_suffix_expr ne "") {
|
---|
401 | $first_metadata_group_element_value =~ s/$remove_suffix_expr$//;
|
---|
402 | }
|
---|
403 | }
|
---|
404 | if (defined($first_metadata_group_element_value) && $first_metadata_group_element_value ne "") {
|
---|
405 | # This section must be included in the classifier
|
---|
406 | $classify_section = 1;
|
---|
407 | last;
|
---|
408 | }
|
---|
409 | }
|
---|
410 |
|
---|
411 | # We're not classifying this section because it doesn't have the required metadata
|
---|
412 | return if (!$classify_section);
|
---|
413 |
|
---|
414 | # Otherwise, include this section in the classifier
|
---|
415 |
|
---|
416 | push(@{$self->{'OIDs'}}, $section_OID);
|
---|
417 |
|
---|
418 | # Create a hash for the metadata values of each metadata element we're interested in
|
---|
419 | my %metadata_groups_done = ();
|
---|
420 | foreach my $metadata_group (@metadata_groups, @{$self->{'sort_leaf_nodes_using_metadata_groups'}}) {
|
---|
421 | # Take care not to do a metadata group more than once
|
---|
422 | unless ($metadata_groups_done{$metadata_group}) {
|
---|
423 | my $remove_prefix_expr = $self->{$metadata_group . ".remove_prefix_expr"};
|
---|
424 | my $remove_suffix_expr = $self->{$metadata_group . ".remove_suffix_expr"};
|
---|
425 | foreach my $metadata_element (split(/\;|,/, $metadata_group)) {
|
---|
426 | my $real_metadata_element = $self->strip_ex_from_metadata($metadata_element);
|
---|
427 |
|
---|
428 | my @metadata_values = @{$doc_obj->get_metadata($section, $real_metadata_element)};
|
---|
429 | foreach my $metadata_value (@metadata_values) {
|
---|
430 | # Strip leading and trailing whitespace
|
---|
431 | $metadata_value =~ s/^\s*//;
|
---|
432 | $metadata_value =~ s/\s*$//;
|
---|
433 |
|
---|
434 | # Remove prefix/suffix if requested
|
---|
435 | if (defined $remove_prefix_expr && $remove_prefix_expr ne "") {
|
---|
436 | $metadata_value =~ s/^$remove_prefix_expr//;
|
---|
437 | }
|
---|
438 | if (defined $remove_suffix_expr && $remove_suffix_expr ne "") {
|
---|
439 | $metadata_value =~ s/$remove_suffix_expr$//;
|
---|
440 | }
|
---|
441 |
|
---|
442 | # lowercase metadata both for sorting meta (d/D under D), and to allow CSS to be able to
|
---|
443 | # text-transform the stored lowercase values as capitalize or uppercase (can't CSS
|
---|
444 | # text-transform if stored uppercase). 2 CSS text-transforms have been added to core.css
|
---|
445 | my $lc_metadata_value = lc($metadata_value);
|
---|
446 |
|
---|
447 | # We are already working with unicode aware strings at this
|
---|
448 | # stage, so we no longer need to convert from utf8 to unicode
|
---|
449 | #my $metadata_value_unicode_string = $metadata_value; # $self->convert_utf8_string_to_unicode_string($metadata_value);
|
---|
450 |
|
---|
451 | # Add the metadata value into the list for this combination of metadata group and section
|
---|
452 | # text that we have some non-whitespace chars
|
---|
453 | if ($lc_metadata_value =~ /\S/) {
|
---|
454 |
|
---|
455 | push(@{$self->{$metadata_group . ".list"}->{$section_OID}}, $lc_metadata_value);
|
---|
456 |
|
---|
457 |
|
---|
458 | # add the actual value into the stored values so we can remember the case
|
---|
459 | if (!$self->{'standardize_capitalization'}) {
|
---|
460 | if (defined $self->{$metadata_group . ".actualvalues"}->{$lc_metadata_value}->{$metadata_value}) {
|
---|
461 | $self->{$metadata_group . ".actualvalues"}->{$lc_metadata_value}->{$metadata_value}++;
|
---|
462 | } else {
|
---|
463 | $self->{$metadata_group . ".actualvalues"}->{$lc_metadata_value}->{$metadata_value} = 1;
|
---|
464 | }
|
---|
465 | }
|
---|
466 | last if ($self->{$metadata_group . ".metadata_selection_mode"} eq "firstvalue");
|
---|
467 | }
|
---|
468 | } # foreach metadatavalue
|
---|
469 | last if ((@metadata_values > 0) && $self->{$metadata_group . ".metadata_selection_mode"} =~ /^(firstvalue|firstvalidmetadata)$/ );
|
---|
470 | } # foreach metadata element
|
---|
471 |
|
---|
472 | $metadata_groups_done{$metadata_group} = 1;
|
---|
473 | }
|
---|
474 | }
|
---|
475 | }
|
---|
476 |
|
---|
477 |
|
---|
478 | sub get_classify_info
|
---|
479 | {
|
---|
480 | my $self = shift(@_);
|
---|
481 |
|
---|
482 | # The metadata groups to classify by
|
---|
483 | my @metadata_groups = @{$self->{'metadata_groups'}};
|
---|
484 | my $first_metadata_group = $metadata_groups[0];
|
---|
485 |
|
---|
486 | # The OID values of the documents to include in the classifier
|
---|
487 | my @OIDs = @{$self->{'OIDs'}};
|
---|
488 |
|
---|
489 | # Create the root node of the classification hierarchy
|
---|
490 | my %classifier_node = ( 'thistype' => "Invisible",
|
---|
491 | 'childtype' => $self->{$first_metadata_group . ".list_type"},
|
---|
492 | 'Title' => $self->{'buttonname'},
|
---|
493 | 'contains' => [],
|
---|
494 | 'mdtype' => $first_metadata_group );
|
---|
495 |
|
---|
496 | # Recursively create the classification hierarchy, one level for each metadata group
|
---|
497 | $self->add_level(\@metadata_groups, \@OIDs, \%classifier_node);
|
---|
498 | return \%classifier_node;
|
---|
499 | }
|
---|
500 |
|
---|
501 |
|
---|
502 | sub add_level
|
---|
503 | {
|
---|
504 | my $self = shift(@_);
|
---|
505 | my @metadata_groups = @{shift(@_)};
|
---|
506 | my @OIDs = @{shift(@_)};
|
---|
507 | my $classifier_node = shift(@_);
|
---|
508 |
|
---|
509 | my $metadata_group = $metadata_groups[0];
|
---|
510 |
|
---|
511 | if (!defined($self->{$metadata_group . ".list"})) {
|
---|
512 | print STDERR "Warning: No metadata values assigned to $metadata_group.\n";
|
---|
513 | return;
|
---|
514 | }
|
---|
515 |
|
---|
516 | # Create a mapping from metadata value to OID
|
---|
517 | my $OID_to_metadata_values_hash_ref = $self->{$metadata_group . ".list"};
|
---|
518 | my %metadata_value_to_OIDs_hash = ();
|
---|
519 | foreach my $OID (@OIDs)
|
---|
520 | {
|
---|
521 | if ($OID_to_metadata_values_hash_ref->{$OID})
|
---|
522 | {
|
---|
523 | my @metadata_values = @{$OID_to_metadata_values_hash_ref->{$OID}};
|
---|
524 | foreach my $metadata_value (@metadata_values)
|
---|
525 | {
|
---|
526 | push(@{$metadata_value_to_OIDs_hash{$metadata_value}}, $OID);
|
---|
527 | }
|
---|
528 | }
|
---|
529 | }
|
---|
530 | #print STDERR "Number of distinct values: " . scalar(keys %metadata_value_to_OIDs_hash) . "\n";
|
---|
531 |
|
---|
532 | # Partition the values (if necessary)
|
---|
533 | my $partition_type_within_level = $self->{$metadata_group . ".partition_type_within_level"};
|
---|
534 | my $partition_size_within_level = $self->{$metadata_group . ".partition_size_within_level"};
|
---|
535 | if ($partition_type_within_level =~ /^per_letter$/i) {
|
---|
536 | # Generate one hlist for each letter
|
---|
537 | my @sortedmetadata_values = $self->sort_metadata_values_array(keys(%metadata_value_to_OIDs_hash));
|
---|
538 | my %metadata_value_to_OIDs_subhash = ();
|
---|
539 |
|
---|
540 | my $lastpartition = substr($sortedmetadata_values[0], 0, 1);
|
---|
541 | foreach my $metadata_value (@sortedmetadata_values) {
|
---|
542 | my $metadata_valuepartition = substr($metadata_value, 0, 1);
|
---|
543 |
|
---|
544 | # Is this the start of a new partition?
|
---|
545 | if ($metadata_valuepartition ne $lastpartition) {
|
---|
546 | $self->add_hlist_partition(\@metadata_groups, $classifier_node, $lastpartition, \%metadata_value_to_OIDs_subhash);
|
---|
547 | %metadata_value_to_OIDs_subhash = ();
|
---|
548 | $lastpartition = $metadata_valuepartition;
|
---|
549 | }
|
---|
550 |
|
---|
551 | $metadata_value_to_OIDs_subhash{$metadata_value} = $metadata_value_to_OIDs_hash{$metadata_value};
|
---|
552 | }
|
---|
553 |
|
---|
554 | # Don't forget to add the last partition
|
---|
555 | $self->add_hlist_partition(\@metadata_groups, $classifier_node, $lastpartition, \%metadata_value_to_OIDs_subhash);
|
---|
556 |
|
---|
557 | # The partitions are stored in an HList
|
---|
558 | $classifier_node->{'childtype'} = "HList";
|
---|
559 | }
|
---|
560 | elsif ($partition_type_within_level =~ /^approximate_size$/i && scalar(keys %metadata_value_to_OIDs_hash) > $partition_size_within_level) {
|
---|
561 | # Generate hlist based on the first letter of the metadata value (like per_letter) but with restriction on the partition size
|
---|
562 | # If a letter has fewer items than specified by the "partition_size_within_level", then group them together if possible
|
---|
563 | # If a letter has more items than specified, split into several hlists.
|
---|
564 | # Depends on the bookshelf_type, one item can be either a document (when bookshelf_type is "never") or a metadata value (otherwise)
|
---|
565 | my $partition_size_within_level = $self->{$metadata_group . ".partition_size_within_level"};
|
---|
566 | my @sortedmetadata_values = $self->sort_metadata_values_array(keys(%metadata_value_to_OIDs_hash));
|
---|
567 | my $bookshelf_type = $self->{$metadata_group . ".bookshelf_type"};
|
---|
568 |
|
---|
569 | # Separate values by their first letter, each form a bucket, like the per_letter partition type
|
---|
570 | my $last_partition = substr($sortedmetadata_values[0], 0, 1);
|
---|
571 | my %partition_buckets = ();
|
---|
572 | my @metadata_values_in_bucket = ();
|
---|
573 | my $num_items_in_bucket = 0;
|
---|
574 | foreach my $metadata_value (@sortedmetadata_values) {
|
---|
575 | my $metadata_valuepartition = substr($metadata_value, 0, 1);
|
---|
576 | if ($metadata_valuepartition ne $last_partition) {
|
---|
577 | my @temp_array = @metadata_values_in_bucket;
|
---|
578 | # Cache the values that belong to this bucket, and the number of items in this bucket, not necessary to be the same number as the metadata values
|
---|
579 | my %partition_info = ();
|
---|
580 | $partition_info{'metadata_values'} = \@temp_array;
|
---|
581 | $partition_info{'size'} = $num_items_in_bucket;
|
---|
582 | $partition_buckets{$last_partition} = \%partition_info;
|
---|
583 |
|
---|
584 | @metadata_values_in_bucket = ($metadata_value);
|
---|
585 | $num_items_in_bucket = $bookshelf_type eq "never" ? scalar(@{$metadata_value_to_OIDs_hash{$metadata_value}}) : scalar(@metadata_values_in_bucket);
|
---|
586 | $last_partition = $metadata_valuepartition;
|
---|
587 | } else {
|
---|
588 | $num_items_in_bucket += $bookshelf_type eq "never" ? scalar(@{$metadata_value_to_OIDs_hash{$metadata_value}}) : scalar(@metadata_values_in_bucket);
|
---|
589 | push (@metadata_values_in_bucket, $metadata_value);
|
---|
590 | }
|
---|
591 | }
|
---|
592 | # Last one
|
---|
593 | my %partition_info = ();
|
---|
594 | $partition_info{'metadata_values'} = \@metadata_values_in_bucket;
|
---|
595 | $partition_info{'size'} = $num_items_in_bucket;
|
---|
596 | $partition_buckets{$last_partition} = \%partition_info;
|
---|
597 |
|
---|
598 | my @partition_keys = $self->sort_metadata_values_array(keys(%partition_buckets));
|
---|
599 | for (my $i = 0; $i < scalar(@partition_keys) - 1; $i++) {
|
---|
600 | my $partition = $partition_keys[$i];
|
---|
601 | my $items_in_partition = $partition_buckets{$partition}->{'size'};
|
---|
602 | # Merge small buckets together, but keep the numeric bucket apart
|
---|
603 | if ($items_in_partition < $partition_size_within_level) {
|
---|
604 | my $items_in_next_partition = $partition_buckets{$partition_keys[$i+1]}->{'size'};
|
---|
605 | if ($items_in_partition + $items_in_next_partition <= $partition_size_within_level
|
---|
606 | && !(($partition =~ /^[^0-9]/ && $partition_keys[$i+1] =~ /^[0-9]/)
|
---|
607 | || ($partition =~ /^[0-9]/ && $partition_keys[$i+1] =~ /^[^0-9]/))) {
|
---|
608 | foreach my $metadata_value_to_merge (@{$partition_buckets{$partition}->{'metadata_values'}}) {
|
---|
609 | push(@{$partition_buckets{$partition_keys[$i+1]}->{'metadata_values'}}, $metadata_value_to_merge);
|
---|
610 | }
|
---|
611 | $partition_buckets{$partition_keys[$i+1]}->{'size'} += $items_in_partition;
|
---|
612 | delete $partition_buckets{$partition};
|
---|
613 | }
|
---|
614 | }
|
---|
615 | }
|
---|
616 | @partition_keys = $self->sort_metadata_values_array(keys(%partition_buckets));
|
---|
617 |
|
---|
618 | # Add partitions, and divide big bucket into several
|
---|
619 | my $last_partition_end = "";
|
---|
620 | my $partition_start = "";
|
---|
621 | foreach my $partition (@partition_keys) {
|
---|
622 | my @metadata_values = $self->sort_metadata_values_array(@{$partition_buckets{$partition}->{'metadata_values'}});
|
---|
623 | my $items_in_partition = $partition_buckets{$partition}->{'size'};
|
---|
624 | $partition_start = $self->generate_partition_start($metadata_values[0], $last_partition_end, $self->{"partition_name_length"});
|
---|
625 |
|
---|
626 | if ($items_in_partition > $partition_size_within_level) {
|
---|
627 | my $items_done = 0;
|
---|
628 | my %metadata_values_to_OIDs_subhashes = ();
|
---|
629 | for (my $i = 0; $i < scalar(@metadata_values); $i++) {
|
---|
630 | my $metadata_value = $metadata_values[$i];
|
---|
631 | # If the bookshelf_type is "never", count the documents, otherwise count the distinct metadata values
|
---|
632 | my $items_for_this_md_value = $bookshelf_type eq "never" ? scalar(@{$metadata_value_to_OIDs_hash{$metadata_value}}) : 1;
|
---|
633 |
|
---|
634 | my $partitionend = $self->generate_partition_end($metadata_value, $partition_start, $self->{"partition_name_length"});
|
---|
635 | my $partitionname = $partition_start;
|
---|
636 | if ($partitionend ne $partition_start) {
|
---|
637 | $partitionname = $partitionname . "-" . $partitionend;
|
---|
638 | }
|
---|
639 |
|
---|
640 | # Start a new partition
|
---|
641 | if ($items_done + $items_for_this_md_value > $partition_size_within_level && $items_done != 0) {
|
---|
642 | $self->add_hlist_partition(\@metadata_groups, $classifier_node, $partitionname, \%metadata_values_to_OIDs_subhashes);
|
---|
643 | $last_partition_end = $partitionend;
|
---|
644 | $partition_start = $self->generate_partition_start($metadata_value, $last_partition_end, $self->{"partition_name_length"});
|
---|
645 | $items_done = 0;
|
---|
646 | %metadata_values_to_OIDs_subhashes = ();
|
---|
647 | }
|
---|
648 |
|
---|
649 | # If bookshelf_type is "never" and the current metadata value holds too many items, need to split into several partitions
|
---|
650 | if ($bookshelf_type eq "never" && $items_for_this_md_value > $partition_size_within_level) {
|
---|
651 | my $partitionname_for_this_value = $self->generate_partition_start($metadata_value, $last_partition_end, $self->{"partition_name_length"});
|
---|
652 | # Get the number of partitions needed for this value
|
---|
653 | my $num_splits = int($items_for_this_md_value / $partition_size_within_level);
|
---|
654 | $num_splits++ if ($items_for_this_md_value / $partition_size_within_level > $num_splits);
|
---|
655 |
|
---|
656 | my @OIDs_for_this_value = @{$metadata_value_to_OIDs_hash{$metadata_value}};
|
---|
657 | for (my $i = 0; $i < $num_splits; $i++) {
|
---|
658 | my %OIDs_subhashes_for_this_value = ();
|
---|
659 | my @OIDs_for_this_partition = ();
|
---|
660 | for (my $d = $i * $partition_size_within_level; $d < (($i+1) * $partition_size_within_level > $items_for_this_md_value ? $items_for_this_md_value : ($i+1) * $partition_size_within_level); $d++) {
|
---|
661 | push (@OIDs_for_this_partition, $OIDs_for_this_value[$d]);
|
---|
662 | }
|
---|
663 |
|
---|
664 | # The last bucket might have only a few items and need to be merged with buckets for subsequent metadata values
|
---|
665 | if ($i == $num_splits - 1 && scalar(@OIDs_for_this_partition) < $partition_size_within_level) {
|
---|
666 | $metadata_values_to_OIDs_subhashes{$metadata_value} = \@OIDs_for_this_partition;
|
---|
667 | $items_done += scalar(@OIDs_for_this_partition);
|
---|
668 | next;
|
---|
669 | }
|
---|
670 |
|
---|
671 | # Add an HList for this bucket
|
---|
672 | $OIDs_subhashes_for_this_value{$metadata_value} = \@OIDs_for_this_partition;
|
---|
673 | $self->add_hlist_partition(\@metadata_groups, $classifier_node, $partitionname_for_this_value, \%OIDs_subhashes_for_this_value);
|
---|
674 | $last_partition_end = $partitionname_for_this_value;
|
---|
675 | }
|
---|
676 | next;
|
---|
677 | }
|
---|
678 |
|
---|
679 | $metadata_values_to_OIDs_subhashes{$metadata_value} = $metadata_value_to_OIDs_hash{$metadata_value};
|
---|
680 | $items_done += $bookshelf_type eq "never" ? scalar(@{$metadata_values_to_OIDs_subhashes{$metadata_value}}) : 1;
|
---|
681 |
|
---|
682 | # The last partition
|
---|
683 | if($i == scalar(@metadata_values) - 1) {
|
---|
684 | $self->add_hlist_partition(\@metadata_groups, $classifier_node, $partitionname, \%metadata_values_to_OIDs_subhashes);
|
---|
685 | }
|
---|
686 | }
|
---|
687 | }
|
---|
688 | else {
|
---|
689 | # The easier case, just add a partition
|
---|
690 | my %metadata_values_to_OIDs_subhashes = ();
|
---|
691 | for (my $i = 0; $i < scalar(@metadata_values); $i++) {
|
---|
692 | my $metadata_value = $metadata_values[$i];
|
---|
693 | $metadata_values_to_OIDs_subhashes{$metadata_value} = $metadata_value_to_OIDs_hash{$metadata_value};
|
---|
694 | }
|
---|
695 | my $last_metadata_value = $metadata_values[scalar(@metadata_values)-1];
|
---|
696 | my $partitionend = $self->generate_partition_end($last_metadata_value, $partition_start, $self->{"partition_name_length"});
|
---|
697 | my $partitionname = $partition_start;
|
---|
698 | if ($partitionend ne $partition_start) {
|
---|
699 | $partitionname = $partitionname . "-" . $partitionend;
|
---|
700 | }
|
---|
701 | $self->add_hlist_partition(\@metadata_groups, $classifier_node, $partitionname, \%metadata_values_to_OIDs_subhashes);
|
---|
702 | $last_partition_end = $partitionend;
|
---|
703 | }
|
---|
704 | }
|
---|
705 |
|
---|
706 | # The partitions are stored in an HList
|
---|
707 | $classifier_node->{'childtype'} = "HList";
|
---|
708 |
|
---|
709 | } # end approximate_size
|
---|
710 | else {
|
---|
711 | # Generate hlists of a certain size
|
---|
712 | if ($partition_type_within_level =~ /^constant_size$/i && scalar(keys %metadata_value_to_OIDs_hash) > $partition_size_within_level) {
|
---|
713 | my @sortedmetadata_values = $self->sort_metadata_values_array(keys(%metadata_value_to_OIDs_hash));
|
---|
714 | my $itemsdone = 0;
|
---|
715 | my %metadata_value_to_OIDs_subhash = ();
|
---|
716 | my $lastpartitionend = "";
|
---|
717 | my $partitionstart;
|
---|
718 | foreach my $metadata_value (@sortedmetadata_values) {
|
---|
719 | $metadata_value_to_OIDs_subhash{$metadata_value} = $metadata_value_to_OIDs_hash{$metadata_value};
|
---|
720 | $itemsdone++;
|
---|
721 | my $itemsinpartition = scalar(keys %metadata_value_to_OIDs_subhash);
|
---|
722 |
|
---|
723 | # Is this the start of a new partition?
|
---|
724 | if ($itemsinpartition == 1) {
|
---|
725 | $partitionstart = $self->generate_partition_start($metadata_value, $lastpartitionend, $self->{"partition_name_length"});
|
---|
726 | }
|
---|
727 |
|
---|
728 | # Is this the end of the partition?
|
---|
729 | if ($itemsinpartition == $partition_size_within_level || $itemsdone == @sortedmetadata_values) {
|
---|
730 | my $partitionend = $self->generate_partition_end($metadata_value, $partitionstart, $self->{"partition_name_length"});
|
---|
731 | my $partitionname = $partitionstart;
|
---|
732 | if ($partitionend ne $partitionstart) {
|
---|
733 | $partitionname = $partitionname . "-" . $partitionend;
|
---|
734 | }
|
---|
735 |
|
---|
736 | $self->add_hlist_partition(\@metadata_groups, $classifier_node, $partitionname, \%metadata_value_to_OIDs_subhash);
|
---|
737 | %metadata_value_to_OIDs_subhash = ();
|
---|
738 | $lastpartitionend = $partitionend;
|
---|
739 | }
|
---|
740 | }
|
---|
741 |
|
---|
742 | # The partitions are stored in an HList
|
---|
743 | $classifier_node->{'childtype'} = "HList";
|
---|
744 | }
|
---|
745 |
|
---|
746 | # Otherwise just add all the values to a VList
|
---|
747 | else {
|
---|
748 | $self->add_vlist(\@metadata_groups, $classifier_node, \%metadata_value_to_OIDs_hash);
|
---|
749 | }
|
---|
750 | }
|
---|
751 | }
|
---|
752 |
|
---|
753 |
|
---|
754 | sub generate_partition_start
|
---|
755 | {
|
---|
756 | my $self = shift(@_);
|
---|
757 | my $metadata_value = shift(@_);
|
---|
758 | my $lastpartitionend = shift(@_);
|
---|
759 | my $partition_name_length = shift(@_);
|
---|
760 |
|
---|
761 | if ($partition_name_length) {
|
---|
762 | return substr($metadata_value, 0, $partition_name_length);
|
---|
763 | }
|
---|
764 |
|
---|
765 | my $partitionstart = substr($metadata_value, 0, 1);
|
---|
766 | if ($partitionstart le $lastpartitionend) {
|
---|
767 | $partitionstart = substr($metadata_value, 0, 2);
|
---|
768 | # Give up after three characters
|
---|
769 | if ($partitionstart le $lastpartitionend) {
|
---|
770 | $partitionstart = substr($metadata_value, 0, 3);
|
---|
771 | }
|
---|
772 | }
|
---|
773 |
|
---|
774 | return $partitionstart;
|
---|
775 | }
|
---|
776 |
|
---|
777 |
|
---|
778 | sub generate_partition_end
|
---|
779 | {
|
---|
780 | my $self = shift(@_);
|
---|
781 | my $metadata_value = shift(@_);
|
---|
782 | my $partitionstart = shift(@_);
|
---|
783 | my $partition_name_length = shift(@_);
|
---|
784 |
|
---|
785 | if ($partition_name_length) {
|
---|
786 | return substr($metadata_value, 0, $partition_name_length);
|
---|
787 | }
|
---|
788 |
|
---|
789 | my $partitionend = substr($metadata_value, 0, length($partitionstart));
|
---|
790 | if ($partitionend gt $partitionstart) {
|
---|
791 | $partitionend = substr($metadata_value, 0, 1);
|
---|
792 | if ($partitionend le $partitionstart) {
|
---|
793 | $partitionend = substr($metadata_value, 0, 2);
|
---|
794 | # Give up after three characters
|
---|
795 | if ($partitionend le $partitionstart) {
|
---|
796 | $partitionend = substr($metadata_value, 0, 3);
|
---|
797 | }
|
---|
798 | }
|
---|
799 | }
|
---|
800 |
|
---|
801 | return $partitionend;
|
---|
802 | }
|
---|
803 |
|
---|
804 |
|
---|
805 | sub add_hlist_partition
|
---|
806 | {
|
---|
807 | my $self = shift(@_);
|
---|
808 | my @metadata_groups = @{shift(@_)};
|
---|
809 | my $classifier_node = shift(@_);
|
---|
810 | my $partitionname = shift(@_);
|
---|
811 | my $metadata_value_to_OIDs_hash_ref = shift(@_);
|
---|
812 |
|
---|
813 | # Create an hlist partition
|
---|
814 | # Note that we don't need to convert from unicode-aware strings
|
---|
815 | # to utf8 here, as that is handled elsewhere in the code
|
---|
816 | my %child_classifier_node = ( 'Title' => $partitionname, #'Title' => $self->convert_unicode_string_to_utf8_string($partitionname),
|
---|
817 | 'childtype' => "VList",
|
---|
818 | 'contains' => [] );
|
---|
819 |
|
---|
820 | # Add the children to the hlist partition
|
---|
821 | $self->add_vlist(\@metadata_groups, \%child_classifier_node, $metadata_value_to_OIDs_hash_ref);
|
---|
822 | push(@{$classifier_node->{'contains'}}, \%child_classifier_node);
|
---|
823 | }
|
---|
824 |
|
---|
825 |
|
---|
826 | sub add_vlist
|
---|
827 | {
|
---|
828 | my $self = shift(@_);
|
---|
829 | my @metadata_groups = @{shift(@_)};
|
---|
830 | my $classifier_node = shift(@_);
|
---|
831 | my $metadata_value_to_OIDs_hash_ref = shift(@_);
|
---|
832 |
|
---|
833 | my $metadata_group = shift(@metadata_groups);
|
---|
834 | $classifier_node->{'mdtype'} = $metadata_group;
|
---|
835 |
|
---|
836 | # Create an entry in the vlist for each value
|
---|
837 | foreach my $metadata_value ($self->sort_metadata_values_array(keys(%{$metadata_value_to_OIDs_hash_ref})))
|
---|
838 | {
|
---|
839 | my @OIDs = @{$metadata_value_to_OIDs_hash_ref->{$metadata_value}};
|
---|
840 | # If there is only one item and 'bookshelf_type' is not always (ie. never or duplicate_only), add the item to the list
|
---|
841 | if (@OIDs == 1 && $self->{$metadata_group . ".bookshelf_type"} ne "always") {
|
---|
842 | my $OID = $OIDs[0];
|
---|
843 | my $offset = $self->metadata_offset($metadata_group, $OID, $metadata_value);
|
---|
844 | push(@{$classifier_node->{'contains'}}, { 'OID' => $OID, 'offset' => $offset });
|
---|
845 | }
|
---|
846 | # If 'bookshelf_type' is 'never', list all the items even if there are duplicated values
|
---|
847 | elsif ($self->{$metadata_group . ".bookshelf_type"} eq "never") {
|
---|
848 | @OIDs = $self->sort_leaf_items(\@OIDs);
|
---|
849 | foreach my $OID (@OIDs) {
|
---|
850 | my $offset = $self->metadata_offset($metadata_group, $OID, $metadata_value);
|
---|
851 | push(@{$classifier_node->{'contains'}}, { 'OID' => $OID , 'offset' => $offset });
|
---|
852 | }
|
---|
853 |
|
---|
854 | }
|
---|
855 | # Otherwise create a sublist (bookshelf) for the metadata value
|
---|
856 | else {
|
---|
857 | my $metadata_value_display = $self->get_metadata_value_display($metadata_group, $metadata_value);
|
---|
858 | # Note that we don't need to convert from unicode-aware strings
|
---|
859 | # to utf8 here, as that is handled elsewhere in the code
|
---|
860 | my %child_classifier_node = ( 'Title' => $metadata_value_display, # 'Title' => $self->convert_unicode_string_to_utf8_string($metadata_value),
|
---|
861 | 'childtype' => "VList",
|
---|
862 | 'mdtype' => $metadata_group,
|
---|
863 | 'contains' => [] );
|
---|
864 |
|
---|
865 | #@OIDs = $self->sort_leaf_items(\@OIDs);
|
---|
866 | # If there are metadata elements remaining, recursively apply the process
|
---|
867 | if (@metadata_groups > 0) {
|
---|
868 | my $next_metadata_group = $metadata_groups[0];
|
---|
869 | $child_classifier_node{'childtype'} = $self->{$next_metadata_group . ".list_type"};
|
---|
870 |
|
---|
871 | # separate metadata into those that below in the next/sub-metadata_group
|
---|
872 | # and those that below at the current level's metadata_group
|
---|
873 |
|
---|
874 | my $OID_to_metadata_values_hash_ref = $self->{$next_metadata_group . ".list"};
|
---|
875 | my @current_level_OIDs = ();
|
---|
876 | my @next_level_OIDs = ();
|
---|
877 | foreach my $OID (@OIDs)
|
---|
878 | {
|
---|
879 | if ($OID_to_metadata_values_hash_ref->{$OID}) {
|
---|
880 | push(@next_level_OIDs, $OID);
|
---|
881 | } else {
|
---|
882 | push(@current_level_OIDs, $OID);
|
---|
883 | }
|
---|
884 | }
|
---|
885 | # recursively process those docs belonging to the sub-metadata_group
|
---|
886 | $self->add_level(\@metadata_groups, \@next_level_OIDs, \%child_classifier_node);
|
---|
887 |
|
---|
888 | # For those docs that don't belong in the sub/next_metadata_group, but which belong
|
---|
889 | # at this level, just add the documents as children of this list at the current level
|
---|
890 | @current_level_OIDs = $self->sort_leaf_items(\@current_level_OIDs);
|
---|
891 | foreach my $current_level_OID (@current_level_OIDs) {
|
---|
892 | my $offset = $self->metadata_offset($metadata_group, $current_level_OID, $metadata_value);
|
---|
893 | push(@{$child_classifier_node{'contains'}}, { 'OID' => $current_level_OID , 'offset' => $offset });
|
---|
894 | }
|
---|
895 | }
|
---|
896 | # Otherwise just add the documents as children of this list
|
---|
897 | else {
|
---|
898 | @OIDs = $self->sort_leaf_items(\@OIDs);
|
---|
899 | foreach my $OID (@OIDs) {
|
---|
900 | my $offset = $self->metadata_offset($metadata_group, $OID, $metadata_value);
|
---|
901 | push(@{$child_classifier_node{'contains'}}, { 'OID' => $OID , 'offset' => $offset });
|
---|
902 | }
|
---|
903 |
|
---|
904 | }
|
---|
905 |
|
---|
906 | # Add the sublist to the list
|
---|
907 | push(@{$classifier_node->{'contains'}}, \%child_classifier_node);
|
---|
908 | }
|
---|
909 | }
|
---|
910 | }
|
---|
911 |
|
---|
912 | sub metadata_offset
|
---|
913 | {
|
---|
914 | my $self = shift(@_);
|
---|
915 | my $metadata_group = shift(@_);
|
---|
916 | my $OID = shift(@_);
|
---|
917 | my $metadata_value = shift(@_);
|
---|
918 |
|
---|
919 | my $OID_to_metadata_values_hash_ref = $self->{$metadata_group . ".list"};
|
---|
920 | my @metadata_values = @{$OID_to_metadata_values_hash_ref->{$OID}};
|
---|
921 | for (my $i = 0; $i < scalar(@metadata_values); $i++) {
|
---|
922 | if ($metadata_value eq $metadata_values[$i]) {
|
---|
923 | return $i;
|
---|
924 | }
|
---|
925 | }
|
---|
926 |
|
---|
927 | return 0;
|
---|
928 | }
|
---|
929 |
|
---|
930 | sub sort_leaf_items
|
---|
931 | {
|
---|
932 | my $self = shift(@_);
|
---|
933 | my @OIDs = @{shift(@_)};
|
---|
934 | # my $classifier_node = shift(@_);
|
---|
935 |
|
---|
936 | # Sort leaf nodes and add to list
|
---|
937 | my @sort_leaf_nodes_using_metadata_groups = @{$self->{'sort_leaf_nodes_using_metadata_groups'}};
|
---|
938 | foreach my $sort_leaf_nodes_usingmetaelem (reverse @sort_leaf_nodes_using_metadata_groups) {
|
---|
939 | my $OID_to_metadata_values_hash_ref = $self->{$sort_leaf_nodes_usingmetaelem . ".list"};
|
---|
940 | # Force a stable sort (Perl 5.6's sort isn't stable)
|
---|
941 | # !! The [0] bits aren't ideal (multiple metadata values) !!
|
---|
942 | @OIDs = @OIDs[ sort {
|
---|
943 | if (defined($OID_to_metadata_values_hash_ref->{$OIDs[$a]}) && defined($OID_to_metadata_values_hash_ref->{$OIDs[$b]}))
|
---|
944 | {
|
---|
945 | $OID_to_metadata_values_hash_ref->{$OIDs[$a]}[0] cmp $OID_to_metadata_values_hash_ref->{$OIDs[$b]}[0];
|
---|
946 | }
|
---|
947 | else
|
---|
948 | {
|
---|
949 | $a <=> $b;
|
---|
950 | }
|
---|
951 | } 0..$#OIDs ];
|
---|
952 | }
|
---|
953 | if ($self->{'reverse_sort_leaf_nodes'}) {
|
---|
954 | #print STDERR "reversing\n";
|
---|
955 | return reverse @OIDs;
|
---|
956 | }
|
---|
957 | return @OIDs;
|
---|
958 | }
|
---|
959 |
|
---|
960 |
|
---|
961 |
|
---|
962 | sub sort_metadata_values_array
|
---|
963 | {
|
---|
964 | my $self = shift(@_);
|
---|
965 | my @metadata_values = @_;
|
---|
966 |
|
---|
967 | if ($self->{'unicode_collator'}) {
|
---|
968 | return $self->{'unicode_collator'}->sort(@metadata_values);
|
---|
969 | }
|
---|
970 | else {
|
---|
971 | return sort { $self->alpha_numeric_cmp($a,$b) }(@metadata_values);
|
---|
972 | }
|
---|
973 | }
|
---|
974 |
|
---|
975 | # $a and $b args automatically passed in and shouldn't be declared
|
---|
976 | sub alpha_numeric_cmp
|
---|
977 | {
|
---|
978 | my $self = shift (@_);
|
---|
979 | my ($aStr, $bStr) = @_;
|
---|
980 | if ($aStr =~ m/^(\d+(\.\d+)?)/)
|
---|
981 | {
|
---|
982 | my $val_a = $1;
|
---|
983 | if ($bStr =~ m/^(\d+(\.\d+)?)/)
|
---|
984 | {
|
---|
985 | my $val_b = $1;
|
---|
986 | if ($val_a != $val_b)
|
---|
987 | {
|
---|
988 | return ($val_a <=> $val_b);
|
---|
989 | }
|
---|
990 | }
|
---|
991 | }
|
---|
992 |
|
---|
993 | return ($aStr cmp $bStr);
|
---|
994 | }
|
---|
995 |
|
---|
996 |
|
---|
997 | sub get_metadata_value_display {
|
---|
998 | my $self = shift(@_);
|
---|
999 | my ($metadata_group, $metadata_value) = @_;
|
---|
1000 | return $metadata_value if $self->{'standardize_capitalization'};
|
---|
1001 | my $actual_values_hash = $self->{$metadata_group . ".actualvalues"}->{$metadata_value};
|
---|
1002 | my $display_value ="";
|
---|
1003 | my $max_count=0;
|
---|
1004 | foreach my $v (keys %$actual_values_hash) {
|
---|
1005 | if ($actual_values_hash->{$v} > $max_count) {
|
---|
1006 | $display_value = $v;
|
---|
1007 | $max_count = $actual_values_hash->{$v};
|
---|
1008 | }
|
---|
1009 | }
|
---|
1010 | return $display_value;
|
---|
1011 | }
|
---|
1012 | 1;
|
---|