source: main/trunk/greenstone2/perllib/classify/AZSectionList.pm@ 23116

Last change on this file since 23116 was 23116, checked in by kjdon, 14 years ago

for incremental build, classifiers are not really done incrementally. Previously, we reconstructed all the docs from the database, and classified them, then processed any new/edited/deleted docs, updating the classifier as necessary. Now, we process all new/updated docs, then reconstruct the docs from the database, but only classify those not changed/deleted. This means that we are only ever adding docs to a classifier, never updating or deleting. I have removed edit_mode and all code handling deleting stuff from the classifier.

  • Property svn:executable set to *
  • Property svn:keywords set to Author Date Id Revision
File size: 3.5 KB
Line 
1###########################################################################
2#
3# AZSectionList.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# classifier plugin for sorting sections alphabetically
27
28# this is very similar to AZList except it sorts by
29# section level metadata (excluding the top level)
30# instead of just top level metadata
31
32# the only change is to the classify() subroutine which
33# must now iterate through each section, adding each
34# to the classification
35
36# 12/05/02 Added usage datastructure - John Thompson
37
38package AZSectionList;
39
40use AZList;
41use sorttools;
42
43use strict;
44no strict 'refs'; # allow filehandles to be variables and viceversa
45
46sub BEGIN {
47 @AZSectionList::ISA = ('AZList');
48}
49
50my $arguments = [
51 ];
52my $options = { 'name' => "AZSectionList",
53 'desc' => "{AZSectionList.desc}",
54 'abstract' => "no",
55 'inherits' => "yes" };
56
57
58sub new {
59 my ($class) = shift (@_);
60 my ($classifierslist,$inputargs,$hashArgOptLists) = @_;
61 push(@$classifierslist, $class);
62
63 push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
64push(@{$hashArgOptLists->{"OptList"}},$options);
65
66 my $self = new AZList($classifierslist, $inputargs, $hashArgOptLists);
67
68 return bless $self, $class;
69}
70
71sub classify {
72 my $self = shift (@_);
73 my ($doc_obj) = @_;
74
75 my $doc_OID = $doc_obj->get_OID();
76 my $thissection = $doc_obj->get_next_section ($doc_obj->get_top_section());
77
78 while (defined $thissection) {
79 $self->classify_section ($thissection, $doc_obj);
80 $thissection = $doc_obj->get_next_section ($thissection);
81 }
82}
83
84sub classify_section {
85 my $self = shift (@_);
86 my ($section, $doc_obj) = @_;
87
88 my $doc_OID = $doc_obj->get_OID();
89
90 my $metavalue;
91 my $metaname;
92
93 if (!defined $self->{'meta_list'}) {
94 # just in case
95 return;
96 }
97
98 # find the first available metadata
99 foreach my $m (@{$self->{'meta_list'}}) {
100 $metavalue = $doc_obj->get_metadata_element($section, $m);
101 $metaname = $m;
102 last if defined $metavalue;
103 }
104
105 # if this section doesn't contain the metadata element we're
106 # sorting by we won't include it in this classification
107
108 if (defined $metavalue && $metavalue ne "") {
109 if ($self->{'removeprefix'}) {
110 $metavalue =~ s/^$self->{'removeprefix'}//;
111 }
112
113 $metavalue = &sorttools::format_metadata_for_sorting($metaname, $metavalue, $doc_obj) unless $self->{'no_metadata_formatting'};
114 if (defined $self->{'list'}->{"$doc_OID.$section"}) {
115 my $outhandle = $self->{'outhandle'};
116 print $outhandle "WARNING: AZSectionList::classify called multiple times " .
117 "for $doc_OID.$section\n";
118 }
119 $self->{'list'}->{"$doc_OID.$section"} = $metavalue;
120 }
121}
122
123
1241;
Note: See TracBrowser for help on using the repository browser.