root/gsdl/trunk/perllib/plugins/OAIPlugin.pm @ 17066

Revision 17066, 13.1 KB (checked in by ak19, 11 years ago)

OAIPlugin now works again: 1. needs to inherit from ReadTextFile? as well in order to have access to the method textcat_get_language_encoding; 2. if dc is not specified as a prefix in each metadata element tag, looks inside the containing/wrapper element to see if that is set to dc and if so, uses that as the prefix.

  • Property svn:keywords set to Author Date Id Revision
Line 
1###########################################################################
2#
3# OAIPlug.pm -- basic Open Archives Initiative (OAI) plugin
4#
5# A component of the Greenstone digital library software
6# from the New Zealand Digital Library Project at the
7# University of Waikato, New Zealand.
8#
9# Copyright (C) 1999 New Zealand Digital Library Project
10#
11# This program is free software; you can redistribute it and/or modify
12# it under the terms of the GNU General Public License as published by
13# the Free Software Foundation; either version 2 of the License, or
14# (at your option) any later version.
15#
16# This program is distributed in the hope that it will be useful,
17# but WITHOUT ANY WARRANTY; without even the implied warranty of
18# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
19# GNU General Public License for more details.
20#
21# You should have received a copy of the GNU General Public License
22# along with this program; if not, write to the Free Software
23# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
24#
25###########################################################################
26
27package OAIPlugin;
28
29use unicode;
30use util;
31
32use strict;
33no strict 'refs'; # allow filehandles to be variables and viceversa
34
35use ReadXMLFile;
36use ReadTextFile; # needed for subroutine textcat_get_language_encoding
37
38sub BEGIN {
39    @OAIPlugin::ISA = ('ReadXMLFile', 'ReadTextFile');
40}
41
42
43my $arguments =
44    [ { 'name' => "process_exp",
45    'desc' => "{BasePlugin.process_exp}",
46    'type' => "regexp",
47    'reqd' => "no",
48    'deft' => &get_default_process_exp() },
49      { 'name' => "input_encoding", # needed by subroutine textcat_get_language_encoding
50    'desc' => "{ReadTextFile.input_encoding}",
51    'type' => "enum",
52    'list' => $ReadTextFile::encoding_plus_auto_list,
53    'reqd' => "no" ,
54    'deft' => "auto" }
55      ];
56
57my $options = { 'name'     => "OAIPlugin",
58        'desc'     => "{OAIPlugin.desc}",
59        'abstract' => "no",
60        'inherits' => "yes",
61        'args'     => $arguments };
62
63
64sub new {
65    my ($class) = shift (@_);
66    my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
67    push(@$pluginlist, $class);
68
69    push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
70    push(@{$hashArgOptLists->{"OptList"}},$options);
71
72    my $self = new ReadXMLFile($pluginlist, $inputargs, $hashArgOptLists);
73
74    return bless $self, $class;
75}
76
77sub get_default_process_exp {
78    my $self = shift (@_);
79
80    return q^(?i)(\.oai)$^;
81}
82
83sub get_doctype {
84    my $self = shift(@_);
85   
86    return "OAI-PMH";
87}
88
89sub xml_start_document {
90    my $self = shift (@_);
91    $self->{'in_metadata_node'} = 0;
92    $self->{'rawxml'} = "";
93}
94
95sub xml_end_document {
96}
97
98sub xml_doctype {
99    my $self = shift(@_);
100
101    my ($expat, $name, $sysid, $pubid, $internal) = @_;
102
103    ##die "" if ($name !~ /^OAI-PMH$/);
104
105    my $outhandle = $self->{'outhandle'};
106    print $outhandle "OAIPlugin: processing $self->{'file'}\n" if $self->{'verbosity'} > 1;
107    print STDERR "<Processing n='$self->{'file'}' p='OAIPlugin'>\n" if $self->{'gli'};
108
109}
110
111
112sub xml_start_tag {
113    my $self = shift(@_);
114    my ($expat,$element) = @_;
115
116    my %attr_hash = %_;
117
118    my $attr = "";
119    map { $attr .= " $_=$attr_hash{$_}"; } keys %attr_hash;
120
121    $self->{'rawxml'} .= "<$element$attr>";
122
123    if ($element eq "metadata") {
124    $self->{'in_metadata_node'} = 1;
125    $self->{'metadata_xml'} = "";
126    }
127
128    if ($self->{'in_metadata_node'}) {
129    $self->{'metadata_xml'} .= "<$element$attr>";
130    }
131}
132
133sub xml_end_tag {
134    my $self = shift(@_);
135    my ($expat, $element) = @_;
136
137    $self->{'rawxml'} .= "</$element>";
138
139    if ($self->{'in_metadata_node'}) {
140    $self->{'metadata_xml'} .= "</$element>";
141    }
142
143    if ($element eq "metadata") {
144    my $textref = \$self->{'metadata_xml'};
145    my $metadata = $self->{'metadata'};
146    $self->extract_oai_metadata($textref,$metadata);
147
148    $self->{'in_metadata_node'} = 0;   
149    }
150
151
152}
153
154sub xml_text {
155    my $self = shift(@_);
156    my ($expat) = @_;
157
158    $self->{'rawxml'} .= $_;
159
160    if ($self->{'in_metadata_node'}) {
161    $self->{'metadata_xml'} .= $_;
162    }
163}
164
165
166
167
168sub read {
169    my $self = shift (@_); 
170 
171    my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
172
173    my $outhandle = $self->{'outhandle'};
174
175    my $filename = $file;
176    $filename = &util::filename_cat ($base_dir, $file) if $base_dir =~ /\w/;
177
178    return 0 if ((-d $filename) && ($filename =~ m/srcdocs$/));
179
180    if ($self->SUPER::read(@_)) {
181
182    # Do encoding stuff
183    my ($language, $encoding) = $self->textcat_get_language_encoding ($filename);
184   
185    my $url_array = $metadata->{'URL'};
186    my $num_urls = (defined $url_array) ? scalar(@$url_array) : 0;
187   
188    my $srcdoc_exists = 0;
189    my $srcdoc_pos = 0;
190    my $filename_dir = &util::filename_head($filename);
191   
192    for (my $i=0; $i<$num_urls; $i++) {
193       
194        if ($url_array->[$i] !~ m/^(http|ftp):/) {
195       
196        my $src_filename = &util::filename_cat($filename_dir, $url_array->[$i]);
197       
198        if (-e $src_filename) {
199            $srcdoc_pos = $i;
200            $srcdoc_exists = 1;
201        }
202        }
203    }
204   
205    if ($srcdoc_exists)
206    {
207        print $outhandle "OAIPlugin: passing metadata on to $url_array->[0]\n"
208        if ($self->{'verbosity'}>1);
209       
210       
211        # Make pretty print metadata table stick with src filename
212        my $ppmd_table = $self->{'ppmd_table'};
213        $metadata->{'prettymd'} = [ $ppmd_table ];
214        $self->{'ppmd_table'} = undef;
215       
216        return &plugin::read ($pluginfo, $filename_dir, $url_array->[0],
217                  $block_hash, $metadata, $processor, $maxdocs,
218                  $total_count, $gli);
219    }
220    else
221    {
222        # create a new document
223        my $doc_obj = new doc ($filename, "indexed_doc");
224        my $top_section = $doc_obj->get_top_section;
225        my $plugin_type = $self->{'plugin_type'};
226       
227        $doc_obj->add_utf8_metadata($top_section, "Language", $language);
228        $doc_obj->add_utf8_metadata($top_section, "Encoding", $encoding);
229        $doc_obj->add_utf8_metadata($top_section, "Plugin", $plugin_type);
230        $doc_obj->add_metadata($top_section, "FileFormat", "OAI");
231        $doc_obj->add_metadata($top_section, "FileSize", (-s $filename));
232       
233        # include any metadata passed in from previous plugins
234        # note that this metadata is associated with the top level section
235        $self->extra_metadata ($doc_obj, $doc_obj->get_top_section(), $metadata);
236       
237        # do plugin specific processing of doc_obj
238        my $textref = \$self->{'rawxml'};
239        unless (defined ($self->process($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj))) {
240        print STDERR "<ProcessingError n='$file'>\n" if ($gli);
241        return -1;
242        }
243       
244        # do any automatic metadata extraction
245        $self->auto_extract_metadata ($doc_obj);
246       
247        # add an OID
248        $self->add_OID($doc_obj);
249       
250        my $ppmd_table = $self->{'ppmd_table'};
251        $doc_obj->add_utf8_metadata($top_section,"prettymd",$ppmd_table);
252        $self->{'ppmd_table'} = undef;
253       
254        # process the document
255        $processor->process($doc_obj);
256       
257        $self->{'num_processed'} ++;
258       
259        return 1; # processed the file
260    }
261    }
262    else {
263    return undef;
264    }
265}
266
267
268# do plugin specific processing of doc_obj
269sub process {
270    my $self = shift (@_);
271    my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
272    my $outhandle = $self->{'outhandle'};
273
274    print STDERR "<Processing n='$file' p='OAIPlugin'>\n" if ($gli);
275    print $outhandle "OAIPlugin: processing $file\n"
276    if $self->{'verbosity'} > 1;
277
278    my $cursection = $doc_obj->get_top_section();
279
280##    $self->extract_metadata ($textref, $metadata, $doc_obj, $cursection);
281
282    # add text to document object
283
284#    $$textref =~ s/<(.*?)>/$1 /g;
285    $$textref =~ s/</&lt;/g;
286    $$textref =~ s/>/&gt;/g;
287    $$textref =~ s/\[/&#91;/g;
288    $$textref =~ s/\]/&#93;/g;
289
290##    print STDERR "*** adding text: $$textref\n";
291   
292    $doc_obj->add_utf8_text($cursection, $$textref);
293
294    return 1;
295}
296
297
298# Improvement is to merge this with newer version in MetadataPass
299
300sub open_prettyprint_metadata_table
301{
302    my $self = shift(@_);
303
304    my $att   = "width=100% cellspacing=2";
305    my $style = "style=\'border-bottom: 4px solid #000080\'";
306
307    $self->{'ppmd_table'} = "\n<table $att $style>";
308}
309
310sub add_prettyprint_metadata_line
311{
312    my $self = shift(@_);
313    my ($metaname, $metavalue_utf8) = @_;
314
315###    $metavalue_utf8 =~ s/hdl\.handle\.net/mcgonagall.cs.waikato.ac.nz:8080\/dspace\/handle/;
316    $metavalue_utf8 = &util::hyperlink_text($metavalue_utf8);
317
318    $self->{'ppmd_table'} .= "  <tr bgcolor=#b5d3cd>\n";
319    $self->{'ppmd_table'} .= "    <td width=30%>\n";
320    $self->{'ppmd_table'} .= "      $metaname\n";
321    $self->{'ppmd_table'} .= "    </td>\n";
322    $self->{'ppmd_table'} .= "    <td>\n";
323    $self->{'ppmd_table'} .= "      $metavalue_utf8\n";
324    $self->{'ppmd_table'} .= "    </td>\n";
325    $self->{'ppmd_table'} .= "  </tr>\n";
326
327}
328
329sub close_prettyprint_metadata_table
330{
331    my $self = shift(@_);
332
333    $self->{'ppmd_table'} .= "</table>\n";
334}
335
336
337sub remap_dcterms_metadata
338{
339    my $self = shift(@_);
340
341    my ($metaname) = @_;
342
343    my $dcterm_mapping = {
344    "alternative" => "dc.title",
345    "tableOfContents" => "dc.description",
346    "abstract" => "dc.description",
347    "created" => "dc.date",
348    "valid" => "dc.date",
349    "available" => "dc.date",
350    "issued" => "dc.date",
351    "modified" => "dc.date",
352    "dateAccepted" => "dc.date",
353    "dateCopyrighted" => "dc.date",
354    "dateSubmitted" => "dc.date",
355    "extent" => "dc.format",
356    "medium" => "dc.format",
357    "isVersionOf" => "dc.relation",
358    "hasVersion" => "dc.relation",
359    "isReplacedBy" => "dc.relation",
360    "replaces" => "dc.relation",
361    "isRequiredBy" => "dc.relation",
362    "requires" => "dc.relation",
363    "isPartOf" => "dc.relation",
364    "hasPart" => "dc.relation",
365    "isReferencedBy" => "dc.relation",
366    "references" => "dc.relation",
367    "isFormatOf" => "dc.relation",
368    "hasFormat" => "dc.relation",
369    "conformsTo" => "dc.relation",
370    "spatial" => "dc.coverage",
371    "temporal" => "dc.coverage",
372    "audience" => "dc.any",
373    "accrualMethod" => "dc.any",
374    "accrualPeriodicity" => "dc.any",
375    "accrualPolicy" => "dc.any",
376    "instructionalMethod" => "dc.any",
377    "provenance" => "dc.any",
378    "rightsHolder" => "dc.any",
379    "mediator" => "audience",
380    "educationLevel" => "audience",
381    "accessRights" => "dc.rights",
382    "license" => "dc.rights",
383    "bibliographicCitation" => "dc.identifier"
384    };
385
386    my ($prefix,$name) = ($metaname =~ m/^(.*?)\.(.*?)$/);
387   
388    if ($prefix eq "dcterms")
389    {
390    if (defined $dcterm_mapping->{$name})
391    {
392        return $dcterm_mapping->{$name}."^".$name;
393    }
394
395    }
396    return $metaname; # didn't get a match, return param passed in unchanged
397}
398
399
400sub extract_oai_metadata {
401    my $self = shift (@_);
402    my ($textref, $metadata) = @_;
403    my $outhandle = $self->{'outhandle'};
404
405    # Only handles DC metadata
406
407    $self->open_prettyprint_metadata_table();
408
409    if ($$textref =~ m/<metadata\s*>(.*?)<\/metadata\s*>/s)
410    {
411    my $metadata_text = $1;
412
413    # locate and remove outermost tag (ignoring any attribute information in top-level tag)
414    my ($wrapper_metadata_xml,$inner_metadata_text) = ($metadata_text =~ m/<([^ >]+).*?>(.*?)<\/\1>/s);
415
416    # split tag into namespace and tag name
417    my($namespace,$top_level_prefix) = ($wrapper_metadata_xml =~ m/^(.*?):(.*?)$/);
418
419    # sometimes, the dc namespace is not specified as the prefix in each element (like <dc:title>)
420    # but is rather defined in the wrapper element containing the various dc meta elements,
421    # like <dc><title></title><creator></creator></dc>.
422    # In such a case, we use this wrapper element as the top_level_prefix
423    if(!defined $top_level_prefix && defined $wrapper_metadata_xml && $wrapper_metadata_xml =~ m/dc$/) {
424        $top_level_prefix = $wrapper_metadata_xml;
425    }
426
427    if ($top_level_prefix !~ m/dc$/) {
428        print $outhandle "Warning: OAIPlugin currently only designed for Dublin Core (or variant) metadata\n";
429        print $outhandle "         This recorded metadata section '$top_level_prefix' does not appear to match.\n";
430        print $outhandle "         Metadata assumed to be in form: <prefix:tag>value</prefix:tag> and will be converted\n";
431        print $outhandle "         into Greenstone metadata as prefix.tag = value\n";
432    }
433
434    while ($inner_metadata_text =~ m/<([^ >]+).*?>(.*?)<\/\1>(.*)/s)
435    {
436        # if URL given for document as identifier metadata, store it ...
437        # $doc_obj->add_utf8_metadata($cursection, "URL", $web_url);
438
439        my $metaname = $1;
440        my $metavalue = $2;
441        $inner_metadata_text = $3;
442
443#       print STDERR "*** metaname = $metaname\n";
444#       print STDERR "*** metavalue = $metavalue\n";
445
446        # $metaname =~ s/^(dc:)?(.)/\u$2/; # strip of optional prefix and uppercase first letter
447        $metaname =~ s/:/\./;
448        if ($metaname !~ m/\./)
449        {
450        $metaname = "$top_level_prefix.$metaname";
451#       print STDERR "*** metaname = $metaname\tmetavalue = $metavalue\n";
452        }
453        $metaname =~ s/\.(.)/\.\u$1/;
454
455        $metaname = $self->remap_dcterms_metadata($metaname);
456
457        $metavalue =~ s/\[/&#91;/g;
458        $metavalue =~ s/\]/&#93;/g;
459
460
461#       if ($metaname eq "Identifier")
462#       {
463#       # name clashes with GSDL reserved metadata name for hash id
464#       $metaname = "URL";
465#       }
466
467        if (defined $metadata->{$metaname})
468        {
469        push(@{$metadata->{$metaname}},$metavalue);
470
471        }
472        else
473        {
474        $metadata->{$metaname} = [ $metavalue ];
475        }
476
477        $self->add_prettyprint_metadata_line($metaname, $metavalue);
478       
479    }
480    }
481
482    $self->close_prettyprint_metadata_table();
483}
484
485## we know from the file extension, so doesn't need to check the doctype
486sub check_doctype {
487
488    return 1;
489}
490
4911;
Note: See TracBrowser for help on using the browser.