1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # MetadataCSVPlugin.pm -- A plugin for metadata in comma-separated value format
|
---|
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 2006 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 | package MetadataCSVPlugin;
|
---|
28 |
|
---|
29 |
|
---|
30 | use BasePlugin;
|
---|
31 | use strict;
|
---|
32 |
|
---|
33 | use multiread;
|
---|
34 |
|
---|
35 |
|
---|
36 | sub BEGIN {
|
---|
37 | @MetadataCSVPlugin::ISA = ('BasePlugin');
|
---|
38 | }
|
---|
39 |
|
---|
40 |
|
---|
41 | my $arguments = [
|
---|
42 | { 'name' => "process_exp",
|
---|
43 | 'desc' => "{BasePlugin.process_exp}",
|
---|
44 | 'type' => "regexp",
|
---|
45 | 'reqd' => "no",
|
---|
46 | 'deft' => &get_default_process_exp() }
|
---|
47 |
|
---|
48 | ];
|
---|
49 |
|
---|
50 |
|
---|
51 | my $options = { 'name' => "MetadataCSVPlugin",
|
---|
52 | 'desc' => "{MetadataCSVPlugin.desc}",
|
---|
53 | 'abstract' => "no",
|
---|
54 | 'inherits' => "yes",
|
---|
55 | 'args' => $arguments };
|
---|
56 |
|
---|
57 |
|
---|
58 | sub new
|
---|
59 | {
|
---|
60 | my ($class) = shift (@_);
|
---|
61 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
---|
62 | push(@$pluginlist, $class);
|
---|
63 |
|
---|
64 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
65 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
66 |
|
---|
67 | my $self = new BasePlugin($pluginlist, $inputargs, $hashArgOptLists);
|
---|
68 |
|
---|
69 | return bless $self, $class;
|
---|
70 | }
|
---|
71 |
|
---|
72 |
|
---|
73 | sub get_default_process_exp
|
---|
74 | {
|
---|
75 | return q^(?i)\.csv$^;
|
---|
76 | }
|
---|
77 |
|
---|
78 | sub file_block_read {
|
---|
79 | my $self = shift (@_);
|
---|
80 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $gli) = @_;
|
---|
81 |
|
---|
82 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
83 |
|
---|
84 | if (!-f $filename_full_path || !$self->can_process_this_file($filename_full_path)) {
|
---|
85 | return undef; # can't recognise
|
---|
86 | }
|
---|
87 |
|
---|
88 | # set this so we know this is a metadata file - needed for incremental
|
---|
89 | # build
|
---|
90 | # if this file changes, then we need to reimport everything
|
---|
91 | $block_hash->{'metadata_files'}->{$filename_full_path} = 1;
|
---|
92 |
|
---|
93 | return 1;
|
---|
94 | }
|
---|
95 |
|
---|
96 | sub metadata_read
|
---|
97 | {
|
---|
98 | my $self = shift (@_);
|
---|
99 | my ($pluginfo, $base_dir, $file, $block_hash,
|
---|
100 | $extrametakeys, $extrametadata, $extrametafile,
|
---|
101 | $processor, $maxdocs, $gli) = @_;
|
---|
102 |
|
---|
103 | # Read metadata from CSV files
|
---|
104 | my $filename = &util::filename_cat($base_dir, $file);
|
---|
105 | if ($filename !~ /\.csv$/ || !-f $filename) {
|
---|
106 | return undef;
|
---|
107 | }
|
---|
108 | print STDERR "\n<Processing n='$file' p='MetadataCSVPlugin'>\n" if ($gli);
|
---|
109 | print STDERR "MetadataCSVPlugin: processing $file\n" if ($self->{'verbosity'}) > 1;
|
---|
110 |
|
---|
111 | # add the file to the block list so that it won't be processed in read, as we will do all we can with it here
|
---|
112 | $block_hash->{'file_blocks'}->{$filename} = 1;
|
---|
113 |
|
---|
114 | # Read the CSV file to get the metadata
|
---|
115 | my $csv_file_content;
|
---|
116 | open(CSV_FILE, "$filename");
|
---|
117 | my $csv_file_reader = new multiread();
|
---|
118 | $csv_file_reader->set_handle('MetadataCSVPlugin::CSV_FILE');
|
---|
119 | $csv_file_reader->read_file(\$csv_file_content);
|
---|
120 | close(CSV_FILE);
|
---|
121 |
|
---|
122 | # Split the file into lines and read the first line (contains the metadata names)
|
---|
123 | $csv_file_content =~ s/\r/\n/g; # Handle non-Unix line endings
|
---|
124 | $csv_file_content =~ s/\n+/\n/g;
|
---|
125 | my @csv_file_lines = split(/\n/, $csv_file_content);
|
---|
126 | my $csv_file_field_line = shift(@csv_file_lines);
|
---|
127 | my @csv_file_fields = split(/\,/, $csv_file_field_line);
|
---|
128 | my $found_filename_field = 0;
|
---|
129 | for (my $i = 0; $i < scalar(@csv_file_fields); $i++) {
|
---|
130 | # Remove any spaces from the field names
|
---|
131 | $csv_file_fields[$i] =~ s/ //g;
|
---|
132 | if ($csv_file_fields[$i] eq "Filename") {
|
---|
133 | $found_filename_field = 1;
|
---|
134 | }
|
---|
135 | }
|
---|
136 |
|
---|
137 | if (!$found_filename_field) {
|
---|
138 | print STDERR "MetadataCSVPlugin Error: No Filename field in CSV file: $filename\n";
|
---|
139 | return -1; # error
|
---|
140 | }
|
---|
141 | # Read each line of the file and assign the metadata appropriately
|
---|
142 | foreach my $csv_line (@csv_file_lines) {
|
---|
143 | # Ignore lines containing only whitespace
|
---|
144 | next if ($csv_line =~ /^\s*$/);
|
---|
145 | my $orig_csv_line = $csv_line;
|
---|
146 | # Build a hash of metadata name to metadata value for this line
|
---|
147 | my %csv_line_metadata;
|
---|
148 | my $i = 0;
|
---|
149 | $csv_line .= ","; # To make the regular expressions simpler
|
---|
150 | while ($csv_line ne "") {
|
---|
151 | # Metadata values containing commas are quoted
|
---|
152 | if ($csv_line =~ s/^\"(.*?)\"\,//) {
|
---|
153 | # Only bother with non-empty values
|
---|
154 | if ($1 ne "" && defined($csv_file_fields[$i])) {
|
---|
155 | if (!defined $csv_line_metadata{$csv_file_fields[$i]}) {
|
---|
156 | $csv_line_metadata{$csv_file_fields[$i]} = [];
|
---|
157 | }
|
---|
158 | push (@{$csv_line_metadata{$csv_file_fields[$i]}}, $1);
|
---|
159 | }
|
---|
160 | }
|
---|
161 | # Normal comma-separated case
|
---|
162 | elsif ($csv_line =~ s/^(.*?)\,//) {
|
---|
163 | # Only bother with non-empty values
|
---|
164 | if ($1 ne "" && defined($csv_file_fields[$i])) {
|
---|
165 | if (!defined $csv_line_metadata{$csv_file_fields[$i]}) {
|
---|
166 | $csv_line_metadata{$csv_file_fields[$i]} = [];
|
---|
167 | }
|
---|
168 | push (@{$csv_line_metadata{$csv_file_fields[$i]}}, $1);
|
---|
169 | }
|
---|
170 | }
|
---|
171 | # The line must be formatted incorrectly
|
---|
172 | else {
|
---|
173 | print STDERR "MetadataCSVPlugin Error: Badly formatted CSV line: $csv_line.\n";
|
---|
174 | last;
|
---|
175 | }
|
---|
176 |
|
---|
177 | $i++;
|
---|
178 | }
|
---|
179 |
|
---|
180 | # We can't associate any metadata without knowing the file to associate it with
|
---|
181 | my $csv_line_filename_array = $csv_line_metadata{"Filename"};
|
---|
182 | if (!defined $csv_line_filename_array) {
|
---|
183 | print STDERR "MetadataCSVPlugin Error: No Filename metadata in CSV line: $orig_csv_line\n";
|
---|
184 | next;
|
---|
185 | }
|
---|
186 | my $csv_line_filename = shift(@$csv_line_filename_array);
|
---|
187 | delete $csv_line_metadata{"Filename"};
|
---|
188 |
|
---|
189 |
|
---|
190 | # Associate the metadata now
|
---|
191 | $csv_line_filename = &util::filename_to_regex($csv_line_filename);
|
---|
192 |
|
---|
193 | $extrametadata->{$csv_line_filename} = \%csv_line_metadata;
|
---|
194 | push(@$extrametakeys, $csv_line_filename);
|
---|
195 | # record which file the metadata came from
|
---|
196 | if (!defined $extrametafile->{$csv_line_filename}) {
|
---|
197 | $extrametafile->{$csv_line_filename} = {};
|
---|
198 | }
|
---|
199 | # maps the file to full path
|
---|
200 | $extrametafile->{$csv_line_filename}->{$file} = $filename;
|
---|
201 | }
|
---|
202 | }
|
---|
203 |
|
---|
204 |
|
---|
205 | 1;
|
---|