1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # UnknownPlug.pm -- Plugin for files you know about but Greenstone doesn't
|
---|
4 | #
|
---|
5 | # A component of the Greenstone digital library software from the New
|
---|
6 | # Zealand Digital Library Project at the University of Waikato, New
|
---|
7 | # Zealand.
|
---|
8 | #
|
---|
9 | # Copyright (C) 2001 Gordon W. Paynter
|
---|
10 | # Copyright (C) 2001 New Zealand Digital Library Project
|
---|
11 | #
|
---|
12 | # This program is free software; you can redistribute it and/or modify
|
---|
13 | # it under the terms of the GNU General Public License as published by
|
---|
14 | # the Free Software Foundation; either version 2 of the License, or
|
---|
15 | # (at your option) any later version.
|
---|
16 | #
|
---|
17 | # This program is distributed in the hope that it will be useful, but
|
---|
18 | # WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
19 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
---|
20 | # General Public License for more details.
|
---|
21 | #
|
---|
22 | # You should have received a copy of the GNU General Public License
|
---|
23 | # along with this program; if not, write to the Free Software
|
---|
24 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
25 | #
|
---|
26 | ###########################################################################
|
---|
27 |
|
---|
28 | # UnknownPlug - a plugin for unknown files
|
---|
29 |
|
---|
30 | # This is a simple Plugin for importing files in formats that
|
---|
31 | # Greenstone doesn't know anything about. A fictional document will
|
---|
32 | # be created for every such file, and the file itself will be passed
|
---|
33 | # to Greenstone as the "associated file" of the document.
|
---|
34 |
|
---|
35 | # Here's an example where it is useful: I have a collection of
|
---|
36 | # pictures that include a couple of quicktime movie files with names
|
---|
37 | # like DCP_0163.MOV. Rather than write a new plugin for quicktime
|
---|
38 | # movies, I add this line to the collection configuration file:
|
---|
39 |
|
---|
40 | # plugin UnknownPlug -process_exp "*.MOV" -assoc_field "movie"
|
---|
41 |
|
---|
42 | # A document is created for each movie, with the associated movie
|
---|
43 | # file's name in the "movie" metadata field. In the collection's
|
---|
44 | # format strings, I use the {If} macro to output different text for
|
---|
45 | # each type of file, like this:
|
---|
46 |
|
---|
47 | # {If}{[movie],<HTML for displaying movie>}{If}{[Image],<HTML for displaying image>}
|
---|
48 |
|
---|
49 | # You can also add extra metadata, such as the Title, Subject, and
|
---|
50 | # Duration, with metadata.xml files and RecPlug. (If you want to use
|
---|
51 | # UnknownPlug with more than one type of file, you will have to add
|
---|
52 | # some sort of distinguishing metadata in this way.)
|
---|
53 |
|
---|
54 |
|
---|
55 |
|
---|
56 | package UnknownPlug;
|
---|
57 |
|
---|
58 | use BasPlug;
|
---|
59 | use parsargv;
|
---|
60 |
|
---|
61 |
|
---|
62 | sub BEGIN {
|
---|
63 | @ISA = ('BasPlug');
|
---|
64 | }
|
---|
65 |
|
---|
66 | sub print_usage {
|
---|
67 | print STDERR "\n usage: plugin UnknownPlug [options]\n\n";
|
---|
68 | print STDERR " options:\n";
|
---|
69 | print STDERR " -assoc_field\t Name of the metadata field that will hold\n";
|
---|
70 | print STDERR "\t\t the associated file's name.\n";
|
---|
71 | print STDERR "\n";
|
---|
72 | print STDERR " -file_type\t Mime type of the file (e.g. image/gif)\n";
|
---|
73 | print STDERR "\n";
|
---|
74 | }
|
---|
75 |
|
---|
76 | sub new {
|
---|
77 | my ($class) = @_;
|
---|
78 | my $self = new BasPlug ($class, @_);
|
---|
79 |
|
---|
80 | if (!parsargv::parse(\@_,
|
---|
81 | q^assoc_field/.*/^, \$self->{'assoc_field'},
|
---|
82 | q^file_type/.*/^, \$self->{'file_type'},
|
---|
83 | "allow_extra_options")) {
|
---|
84 | print STDERR "\nIncorrect options passed to UnknownPlug, check your collect.cfg configuration file\n";
|
---|
85 | &print_usage();
|
---|
86 | die "\n";
|
---|
87 | }
|
---|
88 |
|
---|
89 |
|
---|
90 | return bless $self, $class;
|
---|
91 | }
|
---|
92 |
|
---|
93 | sub get_default_process_exp {
|
---|
94 | return '';
|
---|
95 | }
|
---|
96 |
|
---|
97 |
|
---|
98 | # Associate the unknown file with the new document
|
---|
99 |
|
---|
100 | sub associate_unknown_file {
|
---|
101 | my $self = shift (@_);
|
---|
102 | my $filename = shift (@_); # filename with full path
|
---|
103 | my $file = shift (@_); # filename without path
|
---|
104 | my $doc_obj = shift (@_);
|
---|
105 |
|
---|
106 | my $verbosity = $self->{'verbosity'};
|
---|
107 | my $outhandle = $self->{'outhandle'};
|
---|
108 |
|
---|
109 | # check the filename is okay
|
---|
110 | return 0 if ($file eq "" || $filename eq "");
|
---|
111 |
|
---|
112 | if ($filename =~ m/ /) {
|
---|
113 | print $outhandle "UnknownPlug: \"$filename\" contains a space. choking.\n";
|
---|
114 | return undef;
|
---|
115 | }
|
---|
116 |
|
---|
117 | # Add the file as an associated file ...
|
---|
118 | my $section = $doc_obj->get_top_section();
|
---|
119 | my $file_type = $self->{'file_type'} || "unknown/unknown";
|
---|
120 | my $assoc_field = $self->{'assoc_field'} || "unknown_file";
|
---|
121 |
|
---|
122 | $doc_obj->associate_file($filename, $file, $mime_type, $section);
|
---|
123 | $doc_obj->add_metadata ($section, $assoc_field, $file);
|
---|
124 |
|
---|
125 | return 1;
|
---|
126 | }
|
---|
127 |
|
---|
128 |
|
---|
129 |
|
---|
130 | # The UnknownPlug read() function. This function does all the right
|
---|
131 | # things to make general options work for a given plugin. UnknownPlug
|
---|
132 | # overrides read() because there is no need to read the actual text of
|
---|
133 | # the file in, because the contents of the file is not text...
|
---|
134 | #
|
---|
135 | #
|
---|
136 | # Return number of files processed, undef if can't process
|
---|
137 | #
|
---|
138 | # Note that $base_dir might be "" and that $file might include directories
|
---|
139 |
|
---|
140 | sub read {
|
---|
141 | my $self = shift (@_);
|
---|
142 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs) = @_;
|
---|
143 |
|
---|
144 | my $outhandle = $self->{'outhandle'};
|
---|
145 |
|
---|
146 | # Make sure we're processing the correct file
|
---|
147 | my $filename = &util::filename_cat($base_dir, $file);
|
---|
148 | return 0 if $self->{'block_exp'} ne "" && $filename =~ /$self->{'block_exp'}/;
|
---|
149 | if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) {
|
---|
150 | return undef;
|
---|
151 | }
|
---|
152 | print $outhandle "UnknownPlug processing \"$filename\"\n"
|
---|
153 | if $self->{'verbosity'} > 1;
|
---|
154 |
|
---|
155 | #if there's a leading directory name, eat it...
|
---|
156 | $file =~ s/^.*[\/\\]//;
|
---|
157 |
|
---|
158 | # create a new document
|
---|
159 | my $doc_obj = new doc ($filename, "indexed_doc");
|
---|
160 | $doc_obj->set_OIDtype ($processor->{'OIDtype'});
|
---|
161 |
|
---|
162 | # associate the file with the document
|
---|
163 | if (associate_unknown_file($self, $filename, $file, $doc_obj) != 1)
|
---|
164 | {
|
---|
165 | print "UnknownPlug: couldn't process \"$filename\"\n";
|
---|
166 | return 0;
|
---|
167 | }
|
---|
168 |
|
---|
169 | #create an empty text string so we don't break downstream plugins
|
---|
170 | my $text = "Dummy text inserted by UnknownPlug...";
|
---|
171 |
|
---|
172 | # include any metadata passed in from previous plugins
|
---|
173 | my $section = $doc_obj->get_top_section();
|
---|
174 | $self->extra_metadata ($doc_obj, $section, $metadata);
|
---|
175 |
|
---|
176 | # do plugin specific processing of doc_obj
|
---|
177 | return undef unless defined ($self->process (\$text, $pluginfo, $base_dir,
|
---|
178 | $file, $metadata, $doc_obj));
|
---|
179 |
|
---|
180 | # do any automatic metadata extraction
|
---|
181 | $self->auto_extract_metadata ($doc_obj);
|
---|
182 |
|
---|
183 | # add an OID
|
---|
184 | $doc_obj->set_OID();
|
---|
185 | $doc_obj->add_text($section, $text);
|
---|
186 |
|
---|
187 | # process the document
|
---|
188 | $processor->process($doc_obj);
|
---|
189 |
|
---|
190 | return 1;
|
---|
191 | }
|
---|
192 |
|
---|
193 |
|
---|
194 | # UnknownPlug processing of doc_obj. In practice we don't need to do
|
---|
195 | # anything here because the read function takes care of everything.
|
---|
196 |
|
---|
197 | sub process {
|
---|
198 | my $self = shift (@_);
|
---|
199 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
|
---|
200 | my $outhandle = $self->{'outhandle'};
|
---|
201 |
|
---|
202 | return 1;
|
---|
203 | }
|
---|
204 |
|
---|
205 |
|
---|
206 | 1;
|
---|
207 |
|
---|
208 |
|
---|
209 |
|
---|
210 |
|
---|
211 |
|
---|
212 |
|
---|
213 |
|
---|
214 |
|
---|
215 |
|
---|
216 |
|
---|
217 |
|
---|