1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # ReadXMLFile.pm -- base class for XML plugins
|
---|
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) 2001 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 | package ReadXMLFile;
|
---|
27 |
|
---|
28 | use BasePlugin;
|
---|
29 | use doc;
|
---|
30 | use strict;
|
---|
31 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
32 |
|
---|
33 | sub BEGIN {
|
---|
34 | @ReadXMLFile::ISA = ('BasePlugin');
|
---|
35 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib/cpan");
|
---|
36 | }
|
---|
37 |
|
---|
38 | use XMLParser;
|
---|
39 |
|
---|
40 | my $arguments =
|
---|
41 | [ { 'name' => "process_exp",
|
---|
42 | 'desc' => "{BasePlugin.process_exp}",
|
---|
43 | 'type' => "regexp",
|
---|
44 | 'deft' => &get_default_process_exp(),
|
---|
45 | 'reqd' => "no" },
|
---|
46 | { 'name' => "xslt",
|
---|
47 | 'desc' => "{ReadXMLFile.xslt}",
|
---|
48 | 'type' => "string",
|
---|
49 | 'deft' => "",
|
---|
50 | 'reqd' => "no" } ];
|
---|
51 |
|
---|
52 | my $options = { 'name' => "ReadXMLFile",
|
---|
53 | 'desc' => "{ReadXMLFile.desc}",
|
---|
54 | 'abstract' => "yes",
|
---|
55 | 'inherits' => "yes",
|
---|
56 | 'args' => $arguments };
|
---|
57 |
|
---|
58 | sub new {
|
---|
59 | my ($class) = shift (@_);
|
---|
60 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
---|
61 | push(@$pluginlist, $class);
|
---|
62 |
|
---|
63 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
64 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
65 |
|
---|
66 | my $self = new BasePlugin($pluginlist, $inputargs, $hashArgOptLists);
|
---|
67 |
|
---|
68 | if ($self->{'info_only'}) {
|
---|
69 | # don't worry about creating the XML parser as all we want is the
|
---|
70 | # list of plugin options
|
---|
71 | return bless $self, $class;
|
---|
72 | }
|
---|
73 |
|
---|
74 | my $parser = new XML::Parser('Style' => 'Stream',
|
---|
75 | 'Pkg' => 'ReadXMLFile',
|
---|
76 | 'PluginObj' => $self,
|
---|
77 | 'Handlers' => {'Char' => \&Char,
|
---|
78 | 'XMLDecl' => \&XMLDecl,
|
---|
79 | 'Entity' => \&Entity,
|
---|
80 | 'Doctype' => \&Doctype,
|
---|
81 | 'Default' => \&Default,
|
---|
82 | });
|
---|
83 |
|
---|
84 | $self->{'parser'} = $parser;
|
---|
85 |
|
---|
86 | return bless $self, $class;
|
---|
87 | }
|
---|
88 |
|
---|
89 | # the inheriting class must implement this method to tell whether to parse this doc type
|
---|
90 | sub get_doctype {
|
---|
91 | my $self = shift(@_);
|
---|
92 | die "$self The inheriting class must implement get_doctype method";
|
---|
93 | }
|
---|
94 |
|
---|
95 |
|
---|
96 | sub apply_xslt
|
---|
97 | {
|
---|
98 | my $self = shift @_;
|
---|
99 | my ($xslt,$filename) = @_;
|
---|
100 |
|
---|
101 | my $outhandle = $self->{'outhandle'};
|
---|
102 |
|
---|
103 | my $xslt_filename = $xslt;
|
---|
104 |
|
---|
105 | if (! -e $xslt_filename) {
|
---|
106 | # Look in main site directory
|
---|
107 | my $gsdlhome = $ENV{'GSDLHOME'};
|
---|
108 | $xslt_filename = &util::filename_cat($gsdlhome,$xslt);
|
---|
109 | }
|
---|
110 |
|
---|
111 | if (! -e $xslt_filename) {
|
---|
112 | # Look in collection directory
|
---|
113 | my $coldir = $ENV{'GSDLCOLLECTDIR'};
|
---|
114 | $xslt_filename = &util::filename_cat($coldir,$xslt);
|
---|
115 | }
|
---|
116 |
|
---|
117 | if (! -e $xslt_filename) {
|
---|
118 | print $outhandle "Warning: Unable to find XSLT $xslt\n";
|
---|
119 | if (open(XMLIN,"<$filename")) {
|
---|
120 |
|
---|
121 | my $untransformed_xml = "";
|
---|
122 | while (defined (my $line = <XMLIN>)) {
|
---|
123 |
|
---|
124 | $untransformed_xml .= $line;
|
---|
125 | }
|
---|
126 | close(XMLIN);
|
---|
127 |
|
---|
128 | return $untransformed_xml;
|
---|
129 | }
|
---|
130 | else {
|
---|
131 | print $outhandle "Error: Unable to open file $filename\n";
|
---|
132 | print $outhandle " $!\n";
|
---|
133 | return "";
|
---|
134 | }
|
---|
135 |
|
---|
136 | }
|
---|
137 |
|
---|
138 | my $bin_java = &util::filename_cat($ENV{'GSDLHOME'},"bin","java");
|
---|
139 | my $jar_filename = &util::filename_cat($bin_java,"xalan.jar");
|
---|
140 | my $xslt_base_cmd = "java -jar $jar_filename";
|
---|
141 | my $xslt_cmd = "$xslt_base_cmd -IN \"$filename\" -XSL \"$xslt_filename\"";
|
---|
142 |
|
---|
143 | my $transformed_xml = "";
|
---|
144 |
|
---|
145 | if (open(XSLT_IN,"$xslt_cmd |")) {
|
---|
146 | while (defined (my $line = <XSLT_IN>)) {
|
---|
147 |
|
---|
148 | $transformed_xml .= $line;
|
---|
149 | }
|
---|
150 | close(XSLT_IN);
|
---|
151 | }
|
---|
152 | else {
|
---|
153 | print $outhandle "Error: Unable to run command $xslt_cmd\n";
|
---|
154 | print $outhandle " $!\n";
|
---|
155 | }
|
---|
156 |
|
---|
157 | return $transformed_xml;
|
---|
158 |
|
---|
159 | }
|
---|
160 |
|
---|
161 | sub check_doctype {
|
---|
162 | my $self = shift (@_);
|
---|
163 |
|
---|
164 | my ($filename) = @_;
|
---|
165 |
|
---|
166 | if (open(XMLIN,"<$filename")) {
|
---|
167 | my $doctype = $self->get_doctype();
|
---|
168 | ## check whether the doctype has the same name as the root element tag
|
---|
169 | while (defined (my $line = <XMLIN>)) {
|
---|
170 | ## find the root element
|
---|
171 | if ($line =~ /<([\w\d:]+)[\s>]/){
|
---|
172 | my $root = $1;
|
---|
173 | if ($root !~ $doctype){
|
---|
174 | close(XMLIN);
|
---|
175 | return 0;
|
---|
176 | }
|
---|
177 | else {
|
---|
178 | close(XMLIN);
|
---|
179 | return 1;
|
---|
180 | }
|
---|
181 | }
|
---|
182 | }
|
---|
183 | close(XMLIN);
|
---|
184 | }
|
---|
185 |
|
---|
186 | return undef; # haven't found a valid line
|
---|
187 |
|
---|
188 | }
|
---|
189 |
|
---|
190 | # because we are not just using process_exp to determine whether to process or not, we need to implement this too, so that a file can be passed down if we are not actually processing it
|
---|
191 | sub metadata_read {
|
---|
192 | my $self = shift (@_);
|
---|
193 |
|
---|
194 | my ($pluginfo, $base_dir, $file, $metadata, $extrametakeys, $extrametadata, $processor, $maxdocs, $gli) = @_;
|
---|
195 |
|
---|
196 | my $result = $self->SUPER::metadata_read($pluginfo, $base_dir, $file, $metadata, $extrametakeys, $extrametadata, $processor, $maxdocs, $gli);
|
---|
197 |
|
---|
198 | if (defined $result) {
|
---|
199 | # we think we are processing this, but check that we actually are
|
---|
200 | my $filename = $self->get_full_filename($base_dir, $file);
|
---|
201 |
|
---|
202 | if ($self->check_doctype($filename)) {
|
---|
203 | return $result;
|
---|
204 | }
|
---|
205 | }
|
---|
206 | return undef;
|
---|
207 | }
|
---|
208 |
|
---|
209 | # we need to implement read cos we are not just using process_exp to determine
|
---|
210 | # whether to process this or not.
|
---|
211 | sub read {
|
---|
212 | my $self = shift (@_);
|
---|
213 |
|
---|
214 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
215 |
|
---|
216 | # Make sure we're processing the correct file, do blocking etc
|
---|
217 | my ($block_status,$filename_full_path) = $self->read_block(@_);
|
---|
218 | return $block_status if ((!defined $block_status) || ($block_status==0));
|
---|
219 |
|
---|
220 | ## check the doctype to see whether we really want to process the file
|
---|
221 | if (!$self->check_doctype($filename_full_path)) {
|
---|
222 | # this file is not for us
|
---|
223 | return undef;
|
---|
224 | }
|
---|
225 |
|
---|
226 | $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
|
---|
227 | $self->{'base_dir'} = $base_dir;
|
---|
228 | $self->{'file'} = $file;
|
---|
229 | $self->{'filename'} = $filename_full_path;
|
---|
230 | $self->{'processor'} = $processor;
|
---|
231 | $self->{'metadata'} = $metadata;
|
---|
232 |
|
---|
233 | eval {
|
---|
234 | my $xslt = $self->{'xslt'};
|
---|
235 | if (defined $xslt && ($xslt ne "")) {
|
---|
236 | # perform xslt
|
---|
237 | my $transformed_xml = $self->apply_xslt($xslt,$filename_full_path);
|
---|
238 |
|
---|
239 | # feed transformed file (now in memory as string) into XML parser
|
---|
240 | $self->{'parser'}->parse($transformed_xml);
|
---|
241 | }
|
---|
242 | else {
|
---|
243 | $self->{'parser'}->parsefile($filename_full_path);
|
---|
244 | }
|
---|
245 | };
|
---|
246 |
|
---|
247 | if ($@) {
|
---|
248 |
|
---|
249 | # parsefile may either croak somewhere in XML::Parser (e.g. because
|
---|
250 | # the document is not well formed) or die somewhere in ReadXMLFile or a
|
---|
251 | # derived plugin (e.g. because we're attempting to process a
|
---|
252 | # document whose DOCTYPE is not meant for this plugin). For the
|
---|
253 | # first case we'll print a warning and continue, for the second
|
---|
254 | # we'll just continue quietly
|
---|
255 |
|
---|
256 | print STDERR "**** Error is: $@\n";
|
---|
257 |
|
---|
258 | my ($msg) = $@ =~ /Carp::croak\(\'(.*?)\'\)/;
|
---|
259 | if (defined $msg) {
|
---|
260 | my $outhandle = $self->{'outhandle'};
|
---|
261 | my $plugin_name = ref ($self);
|
---|
262 | print $outhandle "$plugin_name failed to process $file ($msg)\n";
|
---|
263 | }
|
---|
264 |
|
---|
265 | # reset ourself for the next document
|
---|
266 | $self->{'section_level'}=0;
|
---|
267 | print STDERR "<ProcessingError n='$file'>\n" if ($gli);
|
---|
268 | return -1; # error during processing
|
---|
269 | }
|
---|
270 |
|
---|
271 |
|
---|
272 | return 1; # processed the file
|
---|
273 | }
|
---|
274 |
|
---|
275 |
|
---|
276 | sub get_default_process_exp {
|
---|
277 | my $self = shift (@_);
|
---|
278 |
|
---|
279 | return q^(?i)\.xml$^;
|
---|
280 | }
|
---|
281 |
|
---|
282 | sub StartDocument {$_[0]->{'PluginObj'}->xml_start_document(@_);}
|
---|
283 | sub XMLDecl {$_[0]->{'PluginObj'}->xml_xmldecl(@_);}
|
---|
284 | sub Entity {$_[0]->{'PluginObj'}->xml_entity(@_);}
|
---|
285 | sub Doctype {$_[0]->{'PluginObj'}->xml_doctype(@_);}
|
---|
286 | sub StartTag {$_[0]->{'PluginObj'}->xml_start_tag(@_);}
|
---|
287 | sub EndTag {$_[0]->{'PluginObj'}->xml_end_tag(@_);}
|
---|
288 | sub Text {$_[0]->{'PluginObj'}->xml_text(@_);}
|
---|
289 | sub PI {$_[0]->{'PluginObj'}->xml_pi(@_);}
|
---|
290 | sub EndDocument {$_[0]->{'PluginObj'}->xml_end_document(@_);}
|
---|
291 | sub Default {$_[0]->{'PluginObj'}->xml_default(@_);}
|
---|
292 |
|
---|
293 | # This Char function overrides the one in XML::Parser::Stream to overcome a
|
---|
294 | # problem where $expat->{Text} is treated as the return value, slowing
|
---|
295 | # things down significantly in some cases.
|
---|
296 | sub Char {
|
---|
297 | use bytes; # Necessary to prevent encoding issues with XML::Parser 2.31+
|
---|
298 | $_[0]->{'Text'} .= $_[1];
|
---|
299 | return undef;
|
---|
300 | }
|
---|
301 |
|
---|
302 | # Called at the beginning of the XML document.
|
---|
303 | sub xml_start_document {
|
---|
304 | my $self = shift(@_);
|
---|
305 | my ($expat) = @_;
|
---|
306 |
|
---|
307 | $self->open_document();
|
---|
308 | }
|
---|
309 |
|
---|
310 | # Called for XML declarations
|
---|
311 | sub xml_xmldecl {
|
---|
312 | my $self = shift(@_);
|
---|
313 | my ($expat, $version, $encoding, $standalone) = @_;
|
---|
314 | }
|
---|
315 |
|
---|
316 | # Called for XML entities
|
---|
317 | sub xml_entity {
|
---|
318 | my $self = shift(@_);
|
---|
319 | my ($expat, $name, $val, $sysid, $pubid, $ndata) = @_;
|
---|
320 | }
|
---|
321 |
|
---|
322 | # Called for DOCTYPE declarations - use die to bail out if this doctype
|
---|
323 | # is not meant for this plugin
|
---|
324 | sub xml_doctype {
|
---|
325 | my $self = shift(@_);
|
---|
326 |
|
---|
327 | my ($expat, $name, $sysid, $pubid, $internal) = @_;
|
---|
328 | die "ReadXMLFile Cannot process XML document with DOCTYPE of $name";
|
---|
329 | }
|
---|
330 |
|
---|
331 |
|
---|
332 | # Called for every start tag. The $_ variable will contain a copy of the
|
---|
333 | # tag and the %_ variable will contain the element's attributes.
|
---|
334 | sub xml_start_tag {
|
---|
335 | my $self = shift(@_);
|
---|
336 | my ($expat, $element) = @_;
|
---|
337 | }
|
---|
338 |
|
---|
339 | # Called for every end tag. The $_ variable will contain a copy of the tag.
|
---|
340 | sub xml_end_tag {
|
---|
341 | my $self = shift(@_);
|
---|
342 | my ($expat, $element) = @_;
|
---|
343 | }
|
---|
344 |
|
---|
345 | # Called just before start or end tags with accumulated non-markup text in
|
---|
346 | # the $_ variable.
|
---|
347 | sub xml_text {
|
---|
348 | my $self = shift(@_);
|
---|
349 | my ($expat) = @_;
|
---|
350 | }
|
---|
351 |
|
---|
352 | # Called for processing instructions. The $_ variable will contain a copy
|
---|
353 | # of the pi.
|
---|
354 | sub xml_pi {
|
---|
355 | my $self = shift(@_);
|
---|
356 | my ($expat, $target, $data) = @_;
|
---|
357 | }
|
---|
358 |
|
---|
359 | # Called at the end of the XML document.
|
---|
360 | sub xml_end_document {
|
---|
361 | my $self = shift(@_);
|
---|
362 | my ($expat) = @_;
|
---|
363 |
|
---|
364 | $self->close_document();
|
---|
365 | }
|
---|
366 |
|
---|
367 | # Called for any characters not handled by the above functions.
|
---|
368 | sub xml_default {
|
---|
369 | my $self = shift(@_);
|
---|
370 | my ($expat, $text) = @_;
|
---|
371 | }
|
---|
372 |
|
---|
373 | sub open_document {
|
---|
374 | my $self = shift(@_);
|
---|
375 |
|
---|
376 | # create a new document
|
---|
377 | $self->{'doc_obj'} = new doc ($self->{'filename'}, "indexed_doc");
|
---|
378 | $self->{'doc_obj'}->set_OIDtype ($self->{'processor'}->{'OIDtype'}, $self->{'processor'}->{'OIDmetadata'});
|
---|
379 | $self->{'doc_obj'}->add_utf8_metadata($self->{'doc_obj'}->get_top_section(), "Plugin", "$self->{'plugin_type'}");
|
---|
380 |
|
---|
381 | # do we want other auto metadata here (see BasePlugin.read_into_doc_obj)
|
---|
382 | }
|
---|
383 |
|
---|
384 | sub close_document {
|
---|
385 | my $self = shift(@_);
|
---|
386 | my $doc_obj = $self->{'doc_obj'};
|
---|
387 |
|
---|
388 | # do we want other auto stuff here, see BasePlugin.read_into_doc_obj
|
---|
389 |
|
---|
390 | # include any metadata passed in from previous plugins
|
---|
391 | # note that this metadata is associated with the top level section
|
---|
392 | $self->extra_metadata ($doc_obj,
|
---|
393 | $doc_obj->get_top_section(),
|
---|
394 | $self->{'metadata'});
|
---|
395 |
|
---|
396 | # do any automatic metadata extraction
|
---|
397 | $self->auto_extract_metadata ($doc_obj);
|
---|
398 |
|
---|
399 | # add an OID
|
---|
400 | $self->add_OID();
|
---|
401 |
|
---|
402 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Plugin", "$self->{'plugin_type'}");
|
---|
403 | $doc_obj->add_metadata($doc_obj->get_top_section(), "FileFormat", "XML");
|
---|
404 |
|
---|
405 | # process the document
|
---|
406 | $self->{'processor'}->process($doc_obj);
|
---|
407 |
|
---|
408 | $self->{'num_processed'} ++;
|
---|
409 | undef $self->{'doc_obj'};
|
---|
410 | undef $doc_obj; # is this the same as above??
|
---|
411 | }
|
---|
412 |
|
---|
413 | 1;
|
---|
414 |
|
---|
415 |
|
---|
416 |
|
---|
417 |
|
---|