1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # SplitTextFile.pm - a plugin for splitting input files into segments that
|
---|
4 | # will then be individually processed.
|
---|
5 | #
|
---|
6 | #
|
---|
7 | # Copyright 2000 Gordon W. Paynter ([email protected])
|
---|
8 | # Copyright 2000 The New Zealand Digital Library Project
|
---|
9 | #
|
---|
10 | # A component of the Greenstone digital library software
|
---|
11 | # from the New Zealand Digital Library Project at the
|
---|
12 | # University of Waikato, New Zealand.
|
---|
13 | #
|
---|
14 | # This program is free software; you can redistribute it and/or modify
|
---|
15 | # it under the terms of the GNU General Public License as published by
|
---|
16 | # the Free Software Foundation; either version 2 of the License, or
|
---|
17 | # (at your option) any later version.
|
---|
18 | #
|
---|
19 | # This program is distributed in the hope that it will be useful,
|
---|
20 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
21 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
22 | # GNU General Public License for more details.
|
---|
23 | #
|
---|
24 | # You should have received a copy of the GNU General Public License
|
---|
25 | # along with this program; if not, write to the Free Software
|
---|
26 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
27 | #
|
---|
28 | ###########################################################################
|
---|
29 |
|
---|
30 |
|
---|
31 | # SplitTextFile is a plugin for splitting input files into segments that will
|
---|
32 | # then be individually processed.
|
---|
33 |
|
---|
34 | # This plugin should not be called directly. Instead, if you need to
|
---|
35 | # process input files that contain several documents, you should write a
|
---|
36 | # plugin with a process function that will handle one of those documents
|
---|
37 | # and have it inherit from SplitTextFile. See ReferPlug for an example.
|
---|
38 |
|
---|
39 |
|
---|
40 | package SplitTextFile;
|
---|
41 |
|
---|
42 | use ReadTextFile;
|
---|
43 | use gsprintf 'gsprintf';
|
---|
44 | use util;
|
---|
45 |
|
---|
46 | use strict;
|
---|
47 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
48 |
|
---|
49 | # SplitTextFile is a sub-class of ReadTextFile
|
---|
50 | sub BEGIN {
|
---|
51 | @SplitTextFile::ISA = ('ReadTextFile');
|
---|
52 | }
|
---|
53 |
|
---|
54 |
|
---|
55 | my $arguments =
|
---|
56 | [ { 'name' => "split_exp",
|
---|
57 | 'desc' => "{SplitTextFile.split_exp}",
|
---|
58 | 'type' => "regexp",
|
---|
59 | #'deft' => &get_default_split_exp(),
|
---|
60 | 'deft' => "",
|
---|
61 | 'reqd' => "no" } ];
|
---|
62 |
|
---|
63 | my $options = { 'name' => "SplitTextFile",
|
---|
64 | 'desc' => "{SplitTextFile.desc}",
|
---|
65 | 'abstract' => "yes",
|
---|
66 | 'inherits' => "yes",
|
---|
67 | 'args' => $arguments };
|
---|
68 |
|
---|
69 |
|
---|
70 | sub new {
|
---|
71 | my ($class) = shift (@_);
|
---|
72 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
---|
73 | push(@$pluginlist, $class);
|
---|
74 |
|
---|
75 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
76 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
77 |
|
---|
78 | my $self = new ReadTextFile($pluginlist, $inputargs, $hashArgOptLists);
|
---|
79 |
|
---|
80 | $self->{'textcat_store'} = {};
|
---|
81 | $self->{'metapass_srcdoc'} = {}; # which segments have valid metadata_srcdoc
|
---|
82 | return bless $self, $class;
|
---|
83 | }
|
---|
84 |
|
---|
85 | sub init {
|
---|
86 | my $self = shift (@_);
|
---|
87 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
88 |
|
---|
89 | $self->ReadTextFile::init($verbosity, $outhandle, $failhandle);
|
---|
90 |
|
---|
91 | # why is this is init and not in new??
|
---|
92 | if ((!defined $self->{'process_exp'}) || ($self->{'process_exp'} eq "")) {
|
---|
93 |
|
---|
94 | $self->{'process_exp'} = $self->get_default_process_exp ();
|
---|
95 | if ($self->{'process_exp'} eq "") {
|
---|
96 | warn ref($self) . " Warning: plugin has no process_exp\n";
|
---|
97 | }
|
---|
98 | }
|
---|
99 |
|
---|
100 |
|
---|
101 | # set split_exp to default unless explicitly set
|
---|
102 | if (!$self->{'split_exp'}) {
|
---|
103 | $self->{'split_exp'} = $self->get_default_split_exp ();
|
---|
104 | }
|
---|
105 |
|
---|
106 | }
|
---|
107 |
|
---|
108 | # This plugin recurs over the segments it finds
|
---|
109 | sub is_recursive {
|
---|
110 | return 1;
|
---|
111 | }
|
---|
112 |
|
---|
113 | # By default, we split the input text at blank lines
|
---|
114 | sub get_default_split_exp {
|
---|
115 | return q^\n\s*\n^;
|
---|
116 | }
|
---|
117 |
|
---|
118 | sub metadata_read {
|
---|
119 | my $self = shift (@_);
|
---|
120 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $extrametakeys, $extrametadata, $processor, $maxdocs, $gli) = @_;
|
---|
121 |
|
---|
122 | # returns 1 if matches process_exp, and has done blocking in the meantime
|
---|
123 | my $matched = $self->SUPER::metadata_read($pluginfo, $base_dir, $file,
|
---|
124 | $block_hash,
|
---|
125 | $metadata, $extrametakeys,
|
---|
126 | $extrametadata, $processor,
|
---|
127 | $maxdocs, $gli);
|
---|
128 | my $split_matched = undef;
|
---|
129 |
|
---|
130 | if ($matched) {
|
---|
131 |
|
---|
132 | my $outhandle = $self->{'outhandle'};
|
---|
133 | my $filename = &util::filename_cat($base_dir, $file);
|
---|
134 |
|
---|
135 | my $plugin_name = ref ($self);
|
---|
136 | $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
|
---|
137 |
|
---|
138 | $self->{'metapass_srcdoc'}->{$file} = {};
|
---|
139 |
|
---|
140 | # Do encoding stuff
|
---|
141 | my ($language, $encoding) = $self->textcat_get_language_encoding ($filename);
|
---|
142 | my $le_rec = { 'language' => $language, 'encoding' => $encoding };
|
---|
143 | $self->{'textcat_store'}->{$file} = $le_rec;
|
---|
144 |
|
---|
145 | # Read in file ($text will be in utf8)
|
---|
146 | my $text = "";
|
---|
147 | $self->read_file ($filename, $encoding, $language, \$text);
|
---|
148 |
|
---|
149 |
|
---|
150 | if ($text !~ /\w/) {
|
---|
151 | gsprintf($outhandle, "$plugin_name: {ReadTextFile.file_has_no_text}\n",
|
---|
152 | $file)
|
---|
153 | if $self->{'verbosity'};
|
---|
154 |
|
---|
155 | my $failhandle = $self->{'failhandle'};
|
---|
156 | print $failhandle "$file: " . ref($self) . ": file contains no text\n";
|
---|
157 | $self->{'num_not_processed'} ++;
|
---|
158 |
|
---|
159 | $self->{'textcat_store'}->{$file} = undef;
|
---|
160 |
|
---|
161 | return 0;
|
---|
162 | }
|
---|
163 |
|
---|
164 |
|
---|
165 | # Split the text into several smaller segments
|
---|
166 | my $split_exp = $self->{'split_exp'};
|
---|
167 | my @tmp = split(/$split_exp/i, $text);
|
---|
168 | my @segments =();
|
---|
169 | ## get rid of empty segments
|
---|
170 | foreach my $seg (@tmp){
|
---|
171 | if ($seg ne ""){
|
---|
172 | push @segments, $seg;
|
---|
173 | }
|
---|
174 | }
|
---|
175 |
|
---|
176 | print $outhandle "SplitTextFile found " . (scalar @segments) . " documents in $filename\n"
|
---|
177 | if $self->{'verbosity'};
|
---|
178 |
|
---|
179 | $self->{'split_segments'}->{$file} = \@segments;
|
---|
180 | $split_matched = scalar(@segments);
|
---|
181 | }
|
---|
182 |
|
---|
183 | return $split_matched;
|
---|
184 | }
|
---|
185 |
|
---|
186 |
|
---|
187 |
|
---|
188 | # The read function opens a file and splits it into parts.
|
---|
189 | # Each part is sent to the process function
|
---|
190 | #
|
---|
191 | # Returns: Number of document objects created (or undef if it fails)
|
---|
192 |
|
---|
193 | sub read {
|
---|
194 | my $self = shift (@_);
|
---|
195 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
196 | my $outhandle = $self->{'outhandle'};
|
---|
197 | my $verbosity = $self->{'verbosity'};
|
---|
198 |
|
---|
199 | # can we process this file??
|
---|
200 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
201 | return undef unless $self->can_process_this_file($filename_full_path);
|
---|
202 |
|
---|
203 | $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
|
---|
204 |
|
---|
205 | my $le_rec = $self->{'textcat_store'}->{$file};
|
---|
206 | if (!defined $le_rec) {
|
---|
207 | # means no text was found;
|
---|
208 | return 0; # not processed but no point in passing it on
|
---|
209 | }
|
---|
210 |
|
---|
211 | print STDERR "<Processing n='$file' p='$self->{'plugin_type'}'>\n" if ($gli);
|
---|
212 | print $outhandle "$self->{'plugin_type'} processing $file\n"
|
---|
213 | if $self->{'verbosity'} > 1;
|
---|
214 |
|
---|
215 | my $language = $le_rec->{'language'};
|
---|
216 | my $encoding = $le_rec->{'encoding'};
|
---|
217 | $self->{'textcat_store'}->{$file} = undef;
|
---|
218 |
|
---|
219 | my $segments = $self->{'split_segments'}->{$file};
|
---|
220 | $self->{'split_segments'}->{$file} = undef;
|
---|
221 |
|
---|
222 | # Process each segment in turn
|
---|
223 | my ($count, $segment, $segtext, $status, $id);
|
---|
224 | $segment = 0;
|
---|
225 | $count = 0;
|
---|
226 | foreach $segtext (@$segments) {
|
---|
227 | $segment++;
|
---|
228 |
|
---|
229 | if (defined $self->{'metapass_srcdoc'}->{$file}->{$segment}) {
|
---|
230 | # metadata is attached to a srcdoc
|
---|
231 | next;
|
---|
232 | }
|
---|
233 |
|
---|
234 | # create a new document
|
---|
235 | my $doc_obj = new doc ($filename_full_path, "indexed_doc");
|
---|
236 | $doc_obj->set_OIDtype ($processor->{'OIDtype'}, $processor->{'OIDmetadata'});
|
---|
237 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Language", $language);
|
---|
238 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Encoding", $encoding);
|
---|
239 | my ($filemeta) = $file =~ /([^\\\/]+)$/;
|
---|
240 | $self->set_Source_metadata($doc_obj, $filemeta, $encoding);
|
---|
241 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "SourceSegment", "$segment");
|
---|
242 | if ($self->{'cover_image'}) {
|
---|
243 | $self->associate_cover_image($doc_obj, $filename_full_path);
|
---|
244 | }
|
---|
245 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Plugin", "$self->{'plugin_type'}");
|
---|
246 | #$doc_obj->add_metadata($doc_obj->get_top_section(), "FileFormat", "Split");
|
---|
247 |
|
---|
248 | # Calculate a "base" document ID.
|
---|
249 | if (!defined $id) {
|
---|
250 | $doc_obj->set_OID();
|
---|
251 | $id = $doc_obj->get_OID();
|
---|
252 | }
|
---|
253 |
|
---|
254 | # include any metadata passed in from previous plugins
|
---|
255 | # note that this metadata is associated with the top level section
|
---|
256 | $self->extra_metadata ($doc_obj, $doc_obj->get_top_section(), $metadata);
|
---|
257 |
|
---|
258 | # do plugin specific processing of doc_obj
|
---|
259 | print $outhandle "segment $segment\n" if ($self->{'verbosity'});
|
---|
260 | $status = $self->process (\$segtext, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli);
|
---|
261 | if (!defined $status) {
|
---|
262 | print $outhandle "WARNING - no plugin could process segment $segment of $file\n"
|
---|
263 | if ($verbosity >= 2);
|
---|
264 | next;
|
---|
265 | }
|
---|
266 | # If the plugin returned 0, it threw away this part
|
---|
267 | if ($status == 0) {
|
---|
268 | next;
|
---|
269 | }
|
---|
270 | $count += $status;
|
---|
271 |
|
---|
272 | # do any automatic metadata extraction
|
---|
273 | $self->auto_extract_metadata ($doc_obj);
|
---|
274 |
|
---|
275 | # add an OID
|
---|
276 | $self->set_OID($doc_obj, $id, $segment);
|
---|
277 |
|
---|
278 | # process the document
|
---|
279 | $processor->process($doc_obj);
|
---|
280 |
|
---|
281 | $self->{'num_processed'} ++;
|
---|
282 | }
|
---|
283 |
|
---|
284 | delete $self->{'metapass_srcdoc'}->{$file};
|
---|
285 |
|
---|
286 | # Return number of document objects produced
|
---|
287 | return $count;
|
---|
288 | }
|
---|
289 |
|
---|
290 | sub set_OID {
|
---|
291 | my $self = shift (@_);
|
---|
292 | my ($doc_obj, $id, $segment_number) = @_;
|
---|
293 |
|
---|
294 | $doc_obj->set_OID($id . "s" . $segment_number);
|
---|
295 | }
|
---|
296 |
|
---|
297 | 1;
|
---|