root/main/trunk/greenstone2/perllib/plugins/EmbeddedMetadataPlugin.pm @ 23212

Revision 23212, 6.0 KB (checked in by kjdon, 9 years ago)

metadata_read no longer takes maxdocs args - metadata_read must process all docs, so that whatever few are actually processed by read will get their metadata

Line 
1###########################################################################
2#
3# EmbeddedMetadataPlugin.pm -- A plugin for EXIF
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 2007 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
27
28package EmbeddedMetadataPlugin;
29
30use BasePlugin;
31
32use Image::ExifTool qw(:Public);
33use strict;
34
35no strict 'refs'; # allow filehandles to be variables and viceversa
36
37
38sub BEGIN
39{
40  @EmbeddedMetadataPlugin::ISA = ('BasePlugin');
41}
42
43
44
45my $arguments =
46    [ { 'name' => "metadata_field_separator",
47    'desc' => "{HTMLPlugin.metadata_field_separator}",
48    'type' => "string",
49    'deft' => "" }
50
51];
52
53
54my $options = { 'name'     => "EmbeddedMetadataPlugin",
55        'desc'     => "{EmbeddedMetadataPlugin.desc}",
56        'abstract' => "no",
57        'inherits' => "yes",
58        'args'     => $arguments };
59
60sub new()
61{
62    my ($class) = shift (@_);
63    my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
64    push(@$pluginlist, $class);
65
66    if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
67    if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
68
69    my $self = new BasePlugin($pluginlist, $inputargs, $hashArgOptLists);
70
71
72    # Create a new Image::ExifTool object
73    my $exifTool = new Image::ExifTool;
74    $exifTool->Options(Duplicates => 0);
75    $exifTool->Options(PrintConv => 0);
76    $exifTool->Options(Unknown => 1);
77    $exifTool->Options('Verbose');
78    $self->{'exiftool'} = $exifTool;
79
80
81    return bless $self, $class;
82}
83
84
85# Need to think some more about this
86sub get_default_process_exp()
87{
88    return ".*";
89    #q^(?i)\.(wma|wmv|jpe?g|gif)$^;
90}
91
92
93# This plugin doesn't block any files
94#sub get_default_block_exp()
95#{
96#    return '';
97#}
98
99
100sub extractEmbeddedMetadata()
101{
102    my $self = shift(@_);
103    my ($file, $filename, $extrametadata, $extrametakeys) = @_;
104 
105    my %exif_metadata = ();
106
107    my $verbosity = $self->{'verbosity'};
108    my $outhandle = $self->{'outhandle'};
109
110    my $metadata_count = 0;
111   
112    my $separator = $self->{'metadata_field_separator'};
113    if ($separator eq "") {
114    undef $separator;
115    }
116
117    my @group_list = Image::ExifTool::GetAllGroups(0);
118    foreach my $group (@group_list)
119      {
120##    print STDERR "**** group = $group\n";
121
122        # Extract meta information from an image
123        $self->{'exiftool'}->Options(Group0 => [$group]);
124        $self->{'exiftool'}->ExtractInfo($filename);
125
126        # Get list of tags in the order they were found in the file
127        my @tag_list = $self->{'exiftool'}->GetFoundTags('File');
128        foreach my $tag (@tag_list)
129          {
130###       print STDERR "**** tag = $tag\n";
131
132            # Strip any numbering suffix
133            $tag =~ s/^([^\s]+)\s.*$/$1/i;
134            my $value = $self->{'exiftool'}->GetValue($tag);
135            if (defined $value && $value =~ /[a-z0-9]+/i) {
136
137        my $field = "ex.$group.$tag";
138
139        if (!defined $exif_metadata{$field})
140        {
141            $exif_metadata{$field} = [];
142        }
143
144        my $metadata_done = 0;
145        if (ref $value eq 'SCALAR') {
146           
147                    if ($$value =~ /^Binary data/) {
148           
149                        $value = "($$value)";
150            }
151                    else {
152           
153                        my $len = length($$value);
154                        $value = "(Binary data $len bytes)";
155            }
156        }
157        elsif (ref $value eq 'ARRAY') {
158            $metadata_done = 1;
159            foreach my $v (@$value) {
160            push (@{$exif_metadata{$field}}, $self->gsSafe($v));
161            ++$metadata_count;
162            }
163        }
164        else {
165            if (defined $separator) {
166            my @vs = split($separator, $value);
167            $metadata_done = 1;
168            foreach my $v (@vs) {
169                if ($v =~ /\S/) {
170                push (@{$exif_metadata{$field}}, $self->gsSafe($v));
171                ++$metadata_count;
172                }
173            }
174            }
175        }
176        if (!$metadata_done) {
177            push (@{$exif_metadata{$field}}, $self->gsSafe($value));
178            ++$metadata_count;
179        }
180        }
181          }
182      }
183
184
185    if ($metadata_count > 0) {
186    print $outhandle " Extracted $metadata_count pieces of metadata from $filename EXIF block\n";
187    }
188
189    # Protect windows directory chars \
190    $file = &util::filename_to_regex($file);
191   
192    # Associate the metadata now
193
194    $extrametadata->{$file} = \%exif_metadata;
195    push(@$extrametakeys, $file);
196
197}
198
199
200sub metadata_read()
201{
202  my $self = shift (@_);
203  my ($pluginfo, $base_dir, $file, $block_hash,
204      $extrametakeys, $extrametadata, $extrametafile,
205      $processor, $gli, $aux) = @_;
206 
207  my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
208 
209  # we don't want to process directories
210  if (!-f $filename_full_path || !$self->can_process_this_file($filename_full_path)) {
211      return undef;
212  }
213  print STDERR "\n<Processing n='$file' p='EmbeddedMetadataPlugin'>\n" if ($gli);
214  print STDERR "EmbeddedMetadataPlugin: processing $file\n" if ($self->{'verbosity'}) > 1;
215
216
217  $self->extractEmbeddedMetadata($filename_no_path,$filename_full_path,
218                 $extrametadata,$extrametakeys);
219
220   
221  return undef;
222}
223
224
225sub process()
226{
227    # not used
228    return undef;
229}
230
231sub gsSafe()
232  {
233    my $self = shift(@_);
234    my ($text) = @_;
235    # Replace dangerous characters
236    $text =~ s/\(/&#40;/g;
237    $text =~ s/\)/&#41;/g;
238    $text =~ s/,/&#44;/g;
239    $text =~ s/\</&#60;/g;
240    $text =~ s/\</&#62;/g;
241    $text =~ s/\[/&#91;/g;
242    $text =~ s/\]/&#93;/g;
243    # Done
244    return $text;
245  }
246
2471;
Note: See TracBrowser for help on using the browser.