| 1 | ###########################################################################
|
|---|
| 2 | #
|
|---|
| 3 | # UnknownPlugin.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 | # UnknownPlugin - 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 UnknownPlugin -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 | # UnknownPlugin 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 UnknownPlugin;
|
|---|
| 57 |
|
|---|
| 58 | use BasePlugin;
|
|---|
| 59 |
|
|---|
| 60 | use strict;
|
|---|
| 61 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
|---|
| 62 |
|
|---|
| 63 | sub BEGIN {
|
|---|
| 64 | @UnknownPlugin::ISA = ('BasePlugin');
|
|---|
| 65 | }
|
|---|
| 66 |
|
|---|
| 67 | my $arguments =
|
|---|
| 68 | [ { 'name' => "assoc_field",
|
|---|
| 69 | 'desc' => "{UnknownPlugin.assoc_field}",
|
|---|
| 70 | 'type' => "string",
|
|---|
| 71 | 'deft' => "",
|
|---|
| 72 | 'reqd' => "no" },
|
|---|
| 73 | { 'name' => "file_format",
|
|---|
| 74 | 'desc' => "{UnknownPlugin.file_format}",
|
|---|
| 75 | 'type' => "string",
|
|---|
| 76 | 'deft' => "",
|
|---|
| 77 | 'reqd' => "no" },
|
|---|
| 78 | { 'name' => "mime_type",
|
|---|
| 79 | 'desc' => "{UnknownPlugin.mime_type}",
|
|---|
| 80 | 'type' => "string",
|
|---|
| 81 | 'deft' => "",
|
|---|
| 82 | 'reqd' => "no" },
|
|---|
| 83 | { 'name' => "srcicon",
|
|---|
| 84 | 'desc' => "{UnknownPlugin.srcicon}",
|
|---|
| 85 | 'type' => "string",
|
|---|
| 86 | 'deft' => "iconunknown",
|
|---|
| 87 | 'reqd' => "no" },
|
|---|
| 88 | { 'name' => "process_extension",
|
|---|
| 89 | 'desc' => "{UnknownPlugin.process_extension}",
|
|---|
| 90 | 'type' => "string",
|
|---|
| 91 | 'deft' => "",
|
|---|
| 92 | 'reqd' => "no" } ];
|
|---|
| 93 |
|
|---|
| 94 | my $options = { 'name' => "UnknownPlugin",
|
|---|
| 95 | 'desc' => "{UnknownPlugin.desc}",
|
|---|
| 96 | 'abstract' => "no",
|
|---|
| 97 | 'inherits' => "yes",
|
|---|
| 98 | 'args' => $arguments };
|
|---|
| 99 |
|
|---|
| 100 |
|
|---|
| 101 | sub new {
|
|---|
| 102 | my ($class) = shift (@_);
|
|---|
| 103 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
|---|
| 104 | push(@$pluginlist, $class);
|
|---|
| 105 |
|
|---|
| 106 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
|---|
| 107 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
|---|
| 108 |
|
|---|
| 109 | my $self = new BasePlugin($pluginlist, $inputargs, $hashArgOptLists);
|
|---|
| 110 |
|
|---|
| 111 | # "-process_extension" is a simpler alternative to -process_exp for non-regexp people
|
|---|
| 112 | if (!$self->{'process_exp'} && $self->{'process_extension'}) {
|
|---|
| 113 | $self->{'process_exp'} = "\\." . $self->{'process_extension'} . "\$";
|
|---|
| 114 | }
|
|---|
| 115 |
|
|---|
| 116 | return bless $self, $class;
|
|---|
| 117 | }
|
|---|
| 118 |
|
|---|
| 119 |
|
|---|
| 120 | sub process {
|
|---|
| 121 | my $self = shift (@_);
|
|---|
| 122 | my ($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
|---|
| 123 |
|
|---|
| 124 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
|---|
| 125 | my $outhandle = $self->{'outhandle'};
|
|---|
| 126 | my $verbosity = $self->{'verbosity'};
|
|---|
| 127 |
|
|---|
| 128 | # check the filename is okay - do we need this??
|
|---|
| 129 | if ($filename_full_path eq "" || $filename_no_path eq "") {
|
|---|
| 130 | print $outhandle "UnknownPlugin: couldn't process \"$filename_no_path\"\n";
|
|---|
| 131 | return undef;
|
|---|
| 132 | }
|
|---|
| 133 |
|
|---|
| 134 | # Add the file as an associated file ...
|
|---|
| 135 | my $section = $doc_obj->get_top_section();
|
|---|
| 136 | my $file_format = $self->{'file_format'} || "unknown";
|
|---|
| 137 | my $mime_type = $self->{'mime_type'} || "unknown/unknown";
|
|---|
| 138 | my $assoc_field = $self->{'assoc_field'} || "unknown_file";
|
|---|
| 139 |
|
|---|
| 140 | # The assocfilename is the url-encoded version of the utf8 filename
|
|---|
| 141 | my $assoc_file = $doc_obj->get_assocfile_from_sourcefile();
|
|---|
| 142 |
|
|---|
| 143 | $doc_obj->associate_file($filename_full_path, $assoc_file, $mime_type, $section);
|
|---|
| 144 | $doc_obj->add_metadata ($section, "FileFormat", $file_format);
|
|---|
| 145 | $doc_obj->add_metadata ($section, "MimeType", $mime_type);
|
|---|
| 146 | $doc_obj->add_utf8_metadata ($section, $assoc_field, $doc_obj->get_source()); # Source metadata is already in utf8
|
|---|
| 147 | # srclink_file is now deprecated because of the "_" in the metadataname. Use srclinkFile
|
|---|
| 148 | $doc_obj->add_metadata ($section, "srclink_file", $doc_obj->get_sourcefile());
|
|---|
| 149 | $doc_obj->add_metadata ($section, "srclinkFile", $doc_obj->get_sourcefile());
|
|---|
| 150 | $doc_obj->add_metadata ($section, "srcicon", "_".$self->{'srcicon'}."_");
|
|---|
| 151 |
|
|---|
| 152 | # we have no text - add dummy text and NoText metadata
|
|---|
| 153 | $self->add_dummy_text($doc_obj, $section);
|
|---|
| 154 |
|
|---|
| 155 | return 1;
|
|---|
| 156 | }
|
|---|
| 157 |
|
|---|
| 158 |
|
|---|
| 159 | 1;
|
|---|
| 160 |
|
|---|
| 161 |
|
|---|
| 162 |
|
|---|
| 163 |
|
|---|
| 164 |
|
|---|
| 165 |
|
|---|
| 166 |
|
|---|
| 167 |
|
|---|
| 168 |
|
|---|
| 169 |
|
|---|
| 170 |
|
|---|
