1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # GAPlug.pm
|
---|
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 | # Processes GreenstoneArchive XML documents. Note that this plugin does no
|
---|
27 | # syntax checking (though the XML::Parser module tests for
|
---|
28 | # well-formedness). It's assumed that the GreenstoneArchive files conform
|
---|
29 | # to their DTD.
|
---|
30 |
|
---|
31 | package GAPlug;
|
---|
32 |
|
---|
33 | use XMLPlug;
|
---|
34 |
|
---|
35 | sub BEGIN {
|
---|
36 | @ISA = ('XMLPlug');
|
---|
37 | }
|
---|
38 |
|
---|
39 | sub new {
|
---|
40 | my $class = shift (@_);
|
---|
41 | my $self = new XMLPlug ($class, @_);
|
---|
42 |
|
---|
43 | $self->{'section'} = "";
|
---|
44 | $self->{'section_level'} = 0;
|
---|
45 | $self->{'metadata_name'} = "";
|
---|
46 | $self->{'metadata_value'} = "";
|
---|
47 | $self->{'content'} = "";
|
---|
48 |
|
---|
49 | return bless $self, $class;
|
---|
50 | }
|
---|
51 |
|
---|
52 | sub xml_start_document {
|
---|
53 | }
|
---|
54 |
|
---|
55 | sub xml_end_document {
|
---|
56 | }
|
---|
57 |
|
---|
58 | sub xml_doctype {
|
---|
59 | my $self = shift(@_);
|
---|
60 | my ($expat, $name, $sysid, $pubid, $internal) = @_;
|
---|
61 |
|
---|
62 | die "" if ($name ne "GreenstoneArchive");
|
---|
63 | my $outhandle = $self->{'outhandle'};
|
---|
64 | print $outhandle "GAPLug: processing $self->{'file'}\n";
|
---|
65 | }
|
---|
66 |
|
---|
67 |
|
---|
68 | sub xml_start_tag {
|
---|
69 | my $self = shift(@_);
|
---|
70 | my ($expat, $element) = @_;
|
---|
71 |
|
---|
72 | $self->{'element'} = $element;
|
---|
73 | if ($element eq "Section") {
|
---|
74 | if ($self->{'section_level'} == 0) {
|
---|
75 | $self->open_document();
|
---|
76 | } else {
|
---|
77 | my $doc_obj = $self->{'doc_obj'};
|
---|
78 | $self->{'section'} =
|
---|
79 | $doc_obj->insert_section($doc_obj->get_end_child($self->{'section'}));
|
---|
80 | }
|
---|
81 | $self->{'section_level'} ++;
|
---|
82 | }
|
---|
83 | elsif ($element eq "Metadata") {
|
---|
84 | $self->{'metadata_name'} = $_{'name'};
|
---|
85 | }
|
---|
86 | }
|
---|
87 |
|
---|
88 | sub xml_end_tag {
|
---|
89 | my $self = shift(@_);
|
---|
90 | my ($expat, $element) = @_;
|
---|
91 |
|
---|
92 | if ($element eq "Section") {
|
---|
93 | $self->{'section_level'} --;
|
---|
94 | $self->{'section'} = $self->{'doc_obj'}->get_parent_section ($self->{'section'});
|
---|
95 | $self->close_document() if $self->{'section_level'} == 0;
|
---|
96 | }
|
---|
97 | elsif ($element eq "Metadata") {
|
---|
98 | $self->{'doc_obj'}->add_utf8_metadata($self->{'section'}, $self->{'metadata_name'},
|
---|
99 | $self->{'metadata_value'});
|
---|
100 | $self->{'metadata_name'} = "";
|
---|
101 | $self->{'metadata_value'} = "";
|
---|
102 | }
|
---|
103 | elsif ($element eq "Content" && $self->{'content'} ne "") {
|
---|
104 | $self->{'doc_obj'}->add_utf8_text($self->{'section'}, $self->{'content'});
|
---|
105 | $self->{'content'} = "";
|
---|
106 | }
|
---|
107 |
|
---|
108 | $self->{'element'} = "";
|
---|
109 | }
|
---|
110 |
|
---|
111 | sub xml_text {
|
---|
112 | my $self = shift(@_);
|
---|
113 | my ($expat) = @_;
|
---|
114 |
|
---|
115 | if ($self->{'element'} eq "Metadata") {
|
---|
116 | $self->{'metadata_value'} .= $_;
|
---|
117 | }
|
---|
118 | elsif ($self->{'element'} eq "Content") {
|
---|
119 | $self->{'content'} .= $_;
|
---|
120 | }
|
---|
121 | }
|
---|
122 |
|
---|
123 | sub open_document {
|
---|
124 | my $self = shift(@_);
|
---|
125 |
|
---|
126 | # create a new document
|
---|
127 | $self->{'doc_obj'} = new doc ();
|
---|
128 | $self->{'section'} = "";
|
---|
129 | }
|
---|
130 |
|
---|
131 | sub close_document {
|
---|
132 | my $self = shift(@_);
|
---|
133 |
|
---|
134 | # add the associated files
|
---|
135 | my $assoc_files =
|
---|
136 | $self->{'doc_obj'}->get_metadata($self->{'doc_obj'}->get_top_section(), "gsdlassocfile");
|
---|
137 |
|
---|
138 | my $parent_dir = $self->{'filename'};
|
---|
139 | $parent_dir =~ s/[^\\\/]*$//;
|
---|
140 |
|
---|
141 | foreach my $assoc_file_info (@$assoc_files) {
|
---|
142 | my ($assoc_file, $mime_type, $dir) = split (":", $assoc_file_info);
|
---|
143 | my $real_dir = &util::filename_cat($parent_dir, $assoc_file),
|
---|
144 | my $assoc_dir = (defined $dir && $dir ne "")
|
---|
145 | ? &util::filename_cat($dir, $assoc_file) : $assoc_file;
|
---|
146 | $self->{'doc_obj'}->associate_file($real_dir, $assoc_dir, $mime_type);
|
---|
147 | }
|
---|
148 | $self->{'doc_obj'}->delete_metadata($self->{'doc_obj'}->get_top_section(), "gsdlassocfile");
|
---|
149 |
|
---|
150 | # process the document
|
---|
151 | $self->{'processor'}->process($self->{'doc_obj'}, $self->{'file'});
|
---|
152 | }
|
---|
153 |
|
---|
154 |
|
---|
155 | 1;
|
---|