1 | ##########################################################################
|
---|
2 | #
|
---|
3 | # jenaTDBBuildproc.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 | # This document processor outputs a document for indexing (should be
|
---|
27 | # implemented by subclass) and storing in the database
|
---|
28 |
|
---|
29 | package jenaTDBBuildproc;
|
---|
30 |
|
---|
31 | use strict;
|
---|
32 | #no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
33 |
|
---|
34 | use util;
|
---|
35 |
|
---|
36 | use extrabuildproc;
|
---|
37 |
|
---|
38 |
|
---|
39 | BEGIN {
|
---|
40 | @jenaTDBBuildproc::ISA = ('extrabuildproc');
|
---|
41 | }
|
---|
42 |
|
---|
43 | sub new()
|
---|
44 | {
|
---|
45 | my $class = shift @_;
|
---|
46 |
|
---|
47 | my $self = new extrabuildproc (@_);
|
---|
48 |
|
---|
49 | return bless $self, $class;
|
---|
50 | }
|
---|
51 |
|
---|
52 |
|
---|
53 | sub textedit {
|
---|
54 | my $self = shift (@_);
|
---|
55 | my ($doc_obj,$file,$mode) = @_;
|
---|
56 |
|
---|
57 | # Code written on the assumption that that jenaTDB does a replace
|
---|
58 | # operation when presented with a docid that already extis.
|
---|
59 | # => don't need to do anything special to distinguish between
|
---|
60 | # a mode of "add" and "update"
|
---|
61 |
|
---|
62 | my $outhandle = $self->{'outhandle'};
|
---|
63 |
|
---|
64 |
|
---|
65 | my $source_dir = $self->{'source_dir'}; # typically the archives dir
|
---|
66 | my $build_dir = $self->{'build_dir'};
|
---|
67 |
|
---|
68 | # full path to adb database
|
---|
69 | my $adb_filename
|
---|
70 | = &util::filename_cat($build_dir, "jenaTDB", "lsh-features.adb");
|
---|
71 |
|
---|
72 | # get doc id
|
---|
73 | my $doc_oid = $doc_obj->get_OID();
|
---|
74 |
|
---|
75 | # map to assoc dir
|
---|
76 | my $top_section = $doc_obj->get_top_section();
|
---|
77 | my $assoc_file
|
---|
78 | = $doc_obj->get_metadata_element ($top_section,"assocfilepath");
|
---|
79 | my $assoc_filename = &util::filename_cat($source_dir,$assoc_file);
|
---|
80 |
|
---|
81 | my $chr12_filename = &util::filename_cat($assoc_filename,"doc.chr12");
|
---|
82 | my $powerlog_filename = &util::filename_cat($assoc_filename,"doc.power");
|
---|
83 |
|
---|
84 | print $outhandle " Inserting tripples for $doc_oid\n";
|
---|
85 |
|
---|
86 | # my $cmd = "jenaTDB -d \"$adb_filename\" -I -k \"$doc_oid\" -f \"$chr12_filename\" -w \"$powerlog_filename\"";
|
---|
87 |
|
---|
88 | # my $status = system($cmd);
|
---|
89 | # if ($status != 0) {
|
---|
90 | # print STDERR "Error: failed to run:\n $cmd\n$!\n";
|
---|
91 | # }
|
---|
92 |
|
---|
93 | }
|
---|
94 |
|
---|
95 | sub text {
|
---|
96 | my $self = shift (@_);
|
---|
97 | my ($doc_obj,$file) = @_;
|
---|
98 |
|
---|
99 | $self->textedit($doc_obj,$file,"add");
|
---|
100 | }
|
---|
101 |
|
---|
102 | sub textreindex
|
---|
103 | {
|
---|
104 | my $self = shift @_;
|
---|
105 | my ($doc_obj,$file) = @_;
|
---|
106 |
|
---|
107 | $self->textedit($doc_obj,$file,"update");
|
---|
108 | }
|
---|
109 |
|
---|
110 | sub textdelete
|
---|
111 | {
|
---|
112 | my $self = shift @_;
|
---|
113 |
|
---|
114 | my ($doc_obj,$file) = @_;
|
---|
115 |
|
---|
116 | print STDERR "Warning: jenaTDB command-line does not currently support delete operation\n";
|
---|
117 |
|
---|
118 | # $self->textedit($doc_obj,$file,"delete");
|
---|
119 | }
|
---|
120 |
|
---|
121 |
|
---|
122 |
|
---|
123 |
|
---|
124 |
|
---|
125 |
|
---|
126 | sub MGtext {
|
---|
127 | my $self = shift (@_);
|
---|
128 | my ($doc_obj) = @_;
|
---|
129 | my $handle = $self->{'output_handle'};
|
---|
130 |
|
---|
131 | # only output this document if it is one to be indexed
|
---|
132 | return if ($doc_obj->get_doc_type() ne "indexed_doc");
|
---|
133 |
|
---|
134 | # see if this document belongs to this subcollection
|
---|
135 | my $indexed_doc = $self->is_subcollection_doc($doc_obj);
|
---|
136 |
|
---|
137 | # this is another document
|
---|
138 | $self->{'num_docs'} += 1;
|
---|
139 |
|
---|
140 | # get the parameters for the output
|
---|
141 | my ($level, $fields) = split (/:/, $self->{'index'});
|
---|
142 | $fields =~ s/\ball\b/Title,Creator,text/;
|
---|
143 | $fields =~ s/\btopall\b/topTitle,topCreator,toptext/;
|
---|
144 |
|
---|
145 | my $doc_section = 0; # just for this document
|
---|
146 | my $text = "";
|
---|
147 | my $text_extra = "";
|
---|
148 |
|
---|
149 | # get the text for this document
|
---|
150 | my $section = $doc_obj->get_top_section();
|
---|
151 | while (defined $section) {
|
---|
152 | # update a few statistics
|
---|
153 | $doc_section++;
|
---|
154 | $self->{'num_sections'} += 1;
|
---|
155 |
|
---|
156 | my $indexed_section = $doc_obj->get_metadata_element($section, "gsdldoctype") || "indexed_section";
|
---|
157 | if (($indexed_doc) && ($indexed_section eq "indexed_section" || $indexed_section eq "indexed_doc")) {
|
---|
158 | $self->{'num_bytes'} += $doc_obj->get_text_length ($section);
|
---|
159 | foreach my $field (split (/,/, $fields)) {
|
---|
160 | # only deal with this field if it doesn't start with top or
|
---|
161 | # this is the first section
|
---|
162 | my $real_field = $field;
|
---|
163 | if (!($real_field =~ s/^top//) || ($doc_section == 1)) {
|
---|
164 | my $new_text = "";
|
---|
165 | if ($level eq "dummy") {
|
---|
166 | # a dummy index is a special case used when no
|
---|
167 | # indexes are specified (since there must always be
|
---|
168 | # at least one index or we can't retrieve the
|
---|
169 | # compressed text) - we add a small amount of text
|
---|
170 | # to these dummy indexes which will never be seen
|
---|
171 | # but will overcome mg's problems with building
|
---|
172 | # empty indexes
|
---|
173 | $new_text = "this is dummy text to stop mg barfing";
|
---|
174 | $self->{'num_processed_bytes'} += length ($new_text);
|
---|
175 |
|
---|
176 | } elsif ($real_field eq "text") {
|
---|
177 | $new_text = $doc_obj->get_text ($section) if $self->{'store_text'};
|
---|
178 | $self->{'num_processed_bytes'} += length ($new_text);
|
---|
179 | $new_text =~ s/[\cB\cC]//g;
|
---|
180 | $self->find_paragraphs($new_text);
|
---|
181 |
|
---|
182 | } else {
|
---|
183 | my $first = 1;
|
---|
184 | $real_field =~ s/^ex\.([^.]+)$/$1/; # remove ex. namespace iff it's the only namespace prefix (will leave ex.dc.* intact)
|
---|
185 | my @section_metadata = @{$doc_obj->get_metadata ($section, $real_field)};
|
---|
186 | if ($level eq "section" && $section ne $doc_obj->get_top_section() && $self->{'indexing_text'} && defined ($self->{'sections_index_document_metadata'})) {
|
---|
187 | if ($self->{'sections_index_document_metadata'} eq "always" || ( scalar(@section_metadata) == 0 && $self->{'sections_index_document_metadata'} eq "unless_section_metadata_exists")) {
|
---|
188 | push (@section_metadata, @{$doc_obj->get_metadata ($doc_obj->get_top_section(), $real_field)});
|
---|
189 | }
|
---|
190 | }
|
---|
191 | foreach my $meta (@section_metadata) {
|
---|
192 | $meta =~ s/[\cB\cC]//g;
|
---|
193 | $self->{'num_processed_bytes'} += length ($meta);
|
---|
194 | $new_text .= "\cC" unless $first;
|
---|
195 | $new_text .= $meta if $self->{'store_text'};
|
---|
196 | $first = 0;
|
---|
197 | }
|
---|
198 | }
|
---|
199 |
|
---|
200 | # filter the text
|
---|
201 | $new_text = $self->filter_text ($field, $new_text);
|
---|
202 |
|
---|
203 | $text .= "$new_text\cC";
|
---|
204 | }
|
---|
205 | }
|
---|
206 | }
|
---|
207 |
|
---|
208 | if ($level eq "document") { $text_extra .= "\cB"; }
|
---|
209 | else { $text .= "\cB"; }
|
---|
210 |
|
---|
211 | $section = $doc_obj->get_next_section($section);
|
---|
212 | }
|
---|
213 |
|
---|
214 | print $handle "$text$text_extra";
|
---|
215 | }
|
---|
216 |
|
---|
217 |
|
---|
218 | 1;
|
---|