root/main/trunk/greenstone2/perllib/plugins/MetadataCSVPlugin.pm @ 24971

Revision 24971, 7.7 KB (checked in by ak19, 8 years ago)

1. Introduced the util::filepath_to_url_format subroutine which will be used to convert filenames to URL style filenames to match the slashes used in the filename regex-es in extrameta keys used to index into extrameta data structures. 2. Fixed bug on windows where metadata.xml specifies filenames as regex with backslash in front of the file extension's period mark: DirectoryPlugin? needed to unregex the filepath before calling fileparse on it, else the escaping backslash would interfere with perl's fileparse routine (only on windows, since backslash also represents a dirsep here). 3. Updated all those perl plugins where the new util::filepath_to_url_format needs to be called so that they use URL style filenames (thereafter regexed) to index into the extrameta data structures.

  • Property svn:keywords set to Author Date Id Revision
Line 
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
27package MetadataCSVPlugin;
28
29
30use BasePlugin;
31use MetadataRead;
32
33use strict;
34no strict 'refs';
35
36use extrametautil;
37use multiread;
38use util;
39
40use Encode;
41
42# methods with identical signatures take precedence in the order given in the ISA list.
43sub BEGIN {
44    @MetadataCSVPlugin::ISA = ('MetadataRead', 'BasePlugin');
45}
46
47
48my $arguments = [
49      { 'name' => "process_exp",
50    'desc' => "{BasePlugin.process_exp}",
51    'type' => "regexp",
52    'reqd' => "no",
53    'deft' => &get_default_process_exp() }
54
55];
56
57
58my $options = { 'name'     => "MetadataCSVPlugin",
59        'desc'     => "{MetadataCSVPlugin.desc}",
60        'abstract' => "no",
61        'inherits' => "yes",
62        'args'     => $arguments };
63
64
65sub new
66{
67    my ($class) = shift (@_);
68    my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
69    push(@$pluginlist, $class);
70
71    push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
72    push(@{$hashArgOptLists->{"OptList"}},$options);
73
74    my $self = new BasePlugin($pluginlist, $inputargs, $hashArgOptLists);
75
76    return bless $self, $class;
77}
78
79
80sub get_default_process_exp
81{
82    return q^(?i)\.csv$^;
83}
84
85sub file_block_read {
86    my $self = shift (@_);
87    my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $gli) = @_;
88
89    my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
90
91    if (!-f $filename_full_path || !$self->can_process_this_file($filename_full_path)) {
92    return undef; # can't recognise
93    }
94
95    # set this so we know this is a metadata file - needed for incremental
96    # build
97    # if this file changes, then we need to reimport everything
98    $block_hash->{'metadata_files'}->{$filename_full_path} = 1;
99
100    return 1;
101}
102
103sub metadata_read
104{
105    my $self = shift (@_);
106    my ($pluginfo, $base_dir, $file, $block_hash,
107    $extrametakeys, $extrametadata, $extrametafile,
108    $processor, $gli, $aux) = @_;
109
110    # Read metadata from CSV files
111    my $filename = &util::filename_cat($base_dir, $file);
112    if ($filename !~ /\.csv$/ || !-f $filename) {
113    return undef;
114    }
115    print STDERR "\n<Processing n='$file' p='MetadataCSVPlugin'>\n" if ($gli);
116    print STDERR "MetadataCSVPlugin: processing $file\n" if ($self->{'verbosity'}) > 1;
117
118    my $outhandle = $self->{'outhandle'};
119    my $failhandle = $self->{'failhandle'};
120
121    # add the file to the block list so that it won't be processed in read, as we will do all we can with it here
122    &util::block_filename($block_hash,$filename);
123
124
125    # Read the CSV file to get the metadata
126    my $csv_file_content;
127    open(CSV_FILE, "$filename");
128    my $csv_file_reader = new multiread();
129    $csv_file_reader->set_handle('MetadataCSVPlugin::CSV_FILE');
130    $csv_file_reader->read_file(\$csv_file_content);
131
132    # Would be nice if MetadataCSVPlugin was extended to support a minus
133    # option to choose the character encoding the CSV file is in
134    # For now we will assume it is always in UTF8
135    $csv_file_content = decode("utf8",$csv_file_content);
136
137    close(CSV_FILE);
138
139    # Split the file into lines and read the first line (contains the metadata names)
140    $csv_file_content =~ s/\r/\n/g;  # Handle non-Unix line endings
141    $csv_file_content =~ s/\n+/\n/g;
142    my @csv_file_lines = split(/\n/, $csv_file_content);
143    my $csv_file_field_line = shift(@csv_file_lines);
144    my @csv_file_fields = split(/\,/, $csv_file_field_line);
145    my $found_filename_field = 0;
146    for (my $i = 0; $i < scalar(@csv_file_fields); $i++) {
147    # Remove any spaces from the field names
148    $csv_file_fields[$i] =~ s/ //g;
149    if ($csv_file_fields[$i] eq "Filename") {
150        $found_filename_field = 1;
151    }
152    }
153
154    if (!$found_filename_field) {
155    $self->print_error($outhandle, $failhandle, $gli, $filename, "No Filename field in CSV file");
156    return -1; # error
157    }
158    # Read each line of the file and assign the metadata appropriately
159    foreach my $csv_line (@csv_file_lines) {
160    # Ignore lines containing only whitespace
161    next if ($csv_line =~ /^\s*$/);
162    my $orig_csv_line = $csv_line;
163    # Build a hash of metadata name to metadata value for this line
164    my %csv_line_metadata;
165    my $i = 0;
166    $csv_line .= ",";  # To make the regular expressions simpler
167    while ($csv_line ne "") {
168        # Metadata values containing commas are quoted
169        if ($csv_line =~ s/^\"(.*?)\"\,//) {
170        # Only bother with non-empty values
171        if ($1 ne "" && defined($csv_file_fields[$i])) {
172            if (!defined $csv_line_metadata{$csv_file_fields[$i]}) {
173            $csv_line_metadata{$csv_file_fields[$i]} = [];
174            }
175            push (@{$csv_line_metadata{$csv_file_fields[$i]}}, $1);
176        }
177        }
178        # Normal comma-separated case
179        elsif ($csv_line =~ s/^(.*?)\,//) {
180        # Only bother with non-empty values
181        if ($1 ne "" && defined($csv_file_fields[$i])) {
182            if (!defined $csv_line_metadata{$csv_file_fields[$i]}) {
183            $csv_line_metadata{$csv_file_fields[$i]} = [];
184            }
185            push (@{$csv_line_metadata{$csv_file_fields[$i]}}, $1);
186        }
187        }
188        # The line must be formatted incorrectly
189        else {
190        $self->print_error($outhandle, $failhandle, $gli, $filename, "Badly formatted CSV line: $csv_line");
191        last;
192        }
193
194        $i++;
195    }
196
197    # We can't associate any metadata without knowing the file to associate it with
198    my $csv_line_filename_array = $csv_line_metadata{"Filename"};
199    if (!defined $csv_line_filename_array) {
200        $self->print_error($outhandle, $failhandle, $gli, $filename, "No Filename metadata in CSV line: $orig_csv_line");
201        next;
202    }
203    my $csv_line_filename = shift(@$csv_line_filename_array);
204    delete $csv_line_metadata{"Filename"};
205
206
207    # Associate the metadata now
208    # Indexing into the extrameta data structures requires the filename's style of slashes to be in URL format
209    # Then need to convert the filename to a regex, no longer to protect windows directory chars \, but for
210    # protecting special characters like brackets in the filepath such as "C:\Program Files (x86)\Greenstone".
211    $csv_line_filename = &util::filepath_to_url_format($csv_line_filename);
212    $csv_line_filename = &util::filename_to_regex($csv_line_filename);
213
214    &extrametautil::setmetadata($extrametadata, $csv_line_filename, \%csv_line_metadata);
215    &extrametautil::addmetakey($extrametakeys, $csv_line_filename);
216    # record which file the metadata came from
217    if (!defined &extrametautil::getmetafile($extrametafile, $csv_line_filename)) {
218        &extrametautil::setmetafile($extrametafile, $csv_line_filename, {});
219    }
220    # maps the file to full path
221    &extrametautil::setmetafile_for_named_file($extrametafile, $csv_line_filename, $file, $filename);
222    }
223}
224
225sub print_error
226{
227
228    my $self = shift(@_);
229    my ($outhandle, $failhandle, $gli, $file, $error) = @_;
230
231    print $outhandle "MetadataCSVPlugin Error: $file: $error\n";
232    print $failhandle "MetadataCSVPlugin Error: $file: $error\n";
233    print STDERR "<ProcessingError n='$file' r='$error'/>\n" if ($gli);
234}
2351;
Note: See TracBrowser for help on using the browser.