| 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 BaseImporter;
|
|---|
| 31 | use MetadataRead;
|
|---|
| 32 | use CSVFieldSeparator;
|
|---|
| 33 |
|
|---|
| 34 | use strict;
|
|---|
| 35 | no strict 'refs';
|
|---|
| 36 |
|
|---|
| 37 | use extrametautil;
|
|---|
| 38 | use multiread;
|
|---|
| 39 | use util;
|
|---|
| 40 |
|
|---|
| 41 | use Encode;
|
|---|
| 42 | use Text::CSV;
|
|---|
| 43 |
|
|---|
| 44 | # methods with identical signatures take precedence in the order given in the ISA list.
|
|---|
| 45 | sub BEGIN {
|
|---|
| 46 | @MetadataCSVPlugin::ISA = ('MetadataRead', 'BaseImporter', 'CSVFieldSeparator');
|
|---|
| 47 | }
|
|---|
| 48 |
|
|---|
| 49 |
|
|---|
| 50 |
|
|---|
| 51 | my $arguments = [
|
|---|
| 52 | { 'name' => "process_exp",
|
|---|
| 53 | 'desc' => "{BaseImporter.process_exp}",
|
|---|
| 54 | 'type' => "regexp",
|
|---|
| 55 | 'reqd' => "no",
|
|---|
| 56 | 'deft' => &get_default_process_exp() },
|
|---|
| 57 | { 'name' => "filename_formed_from",
|
|---|
| 58 | 'desc' => "Specify in a comma separated list the metadata fields that get combined together to form the filename",
|
|---|
| 59 | 'type' => "string",
|
|---|
| 60 | 'reqd' => "no"
|
|---|
| 61 | },
|
|---|
| 62 | { 'name' => "filename_formed_with_no_spaces",
|
|---|
| 63 | 'desc' => "Set this flag to remove an spaces in the metadata values used to form the filename",
|
|---|
| 64 | 'type' => "flag",
|
|---|
| 65 | 'reqd' => "no"
|
|---|
| 66 | },
|
|---|
| 67 | { 'name' => "filename_formed_from_ext",
|
|---|
| 68 | 'desc' => "The filename extension to add on to the end of the filename_formed_from metadata",
|
|---|
| 69 | 'type' => "string",
|
|---|
| 70 | 'reqd' => "no",
|
|---|
| 71 | 'deft' => ".nul"
|
|---|
| 72 | }
|
|---|
| 73 | ];
|
|---|
| 74 |
|
|---|
| 75 |
|
|---|
| 76 | my $options = { 'name' => "MetadataCSVPlugin",
|
|---|
| 77 | 'desc' => "{MetadataCSVPlugin.desc}",
|
|---|
| 78 | 'abstract' => "no",
|
|---|
| 79 | 'inherits' => "yes",
|
|---|
| 80 | 'args' => $arguments };
|
|---|
| 81 |
|
|---|
| 82 |
|
|---|
| 83 | sub new
|
|---|
| 84 | {
|
|---|
| 85 | my ($class) = shift (@_);
|
|---|
| 86 | my ($pluginlist,$inputargs,$hashArgOptLists) = @_;
|
|---|
| 87 | push(@$pluginlist, $class);
|
|---|
| 88 |
|
|---|
| 89 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
|---|
| 90 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
|---|
| 91 |
|
|---|
| 92 | new CSVFieldSeparator($pluginlist, $inputargs, $hashArgOptLists);
|
|---|
| 93 | my $self = new BaseImporter($pluginlist, $inputargs, $hashArgOptLists);
|
|---|
| 94 |
|
|---|
| 95 | return bless $self, $class;
|
|---|
| 96 | }
|
|---|
| 97 |
|
|---|
| 98 |
|
|---|
| 99 | sub get_default_process_exp
|
|---|
| 100 | {
|
|---|
| 101 | return q^(?i)\.csv$^;
|
|---|
| 102 | }
|
|---|
| 103 |
|
|---|
| 104 | sub file_block_read {
|
|---|
| 105 | my $self = shift (@_);
|
|---|
| 106 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $gli) = @_;
|
|---|
| 107 |
|
|---|
| 108 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
|---|
| 109 |
|
|---|
| 110 | if (!-f $filename_full_path || !$self->can_process_this_file($filename_full_path)) {
|
|---|
| 111 | return undef; # can't recognise
|
|---|
| 112 | }
|
|---|
| 113 |
|
|---|
| 114 | # set this so we know this is a metadata file - needed for incremental
|
|---|
| 115 | # build
|
|---|
| 116 | # if this file changes, then we need to reimport everything
|
|---|
| 117 | $block_hash->{'metadata_files'}->{$filename_full_path} = 1;
|
|---|
| 118 |
|
|---|
| 119 | return 1;
|
|---|
| 120 | }
|
|---|
| 121 |
|
|---|
| 122 | sub metadata_read
|
|---|
| 123 | {
|
|---|
| 124 | my $self = shift (@_);
|
|---|
| 125 | my ($pluginfo, $base_dir, $file, $block_hash,
|
|---|
| 126 | $extrametakeys, $extrametadata, $extrametafile,
|
|---|
| 127 | $processor, $gli, $aux) = @_;
|
|---|
| 128 |
|
|---|
| 129 | # can we process this file??
|
|---|
| 130 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
|---|
| 131 | return undef unless $self->can_process_this_file_for_metadata($filename_full_path);
|
|---|
| 132 |
|
|---|
| 133 | print STDERR "\n<Processing n='$file' p='MetadataCSVPlugin'>\n" if ($gli);
|
|---|
| 134 | print STDERR "MetadataCSVPlugin: processing $file\n" if ($self->{'verbosity'}) > 1;
|
|---|
| 135 |
|
|---|
| 136 | my $outhandle = $self->{'outhandle'};
|
|---|
| 137 | my $failhandle = $self->{'failhandle'};
|
|---|
| 138 |
|
|---|
| 139 | # 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
|
|---|
| 140 | $self->block_raw_filename($block_hash,$filename_full_path);
|
|---|
| 141 |
|
|---|
| 142 |
|
|---|
| 143 | # Read the CSV file to get the metadata
|
|---|
| 144 | my $csv_file_content;
|
|---|
| 145 | open(CSV_FILE, "$filename_full_path");
|
|---|
| 146 | my $csv_file_reader = new multiread();
|
|---|
| 147 | $csv_file_reader->set_handle('MetadataCSVPlugin::CSV_FILE');
|
|---|
| 148 | $csv_file_reader->read_file(\$csv_file_content);
|
|---|
| 149 |
|
|---|
| 150 | # Would be nice if MetadataCSVPlugin was extended to support a minus
|
|---|
| 151 | # option to choose the character encoding the CSV file is in
|
|---|
| 152 | # For now we will assume it is always in UTF8
|
|---|
| 153 | $csv_file_content = decode("utf8",$csv_file_content);
|
|---|
| 154 |
|
|---|
| 155 | close(CSV_FILE);
|
|---|
| 156 |
|
|---|
| 157 | # Split the file into lines and read the first line (contains the metadata names)
|
|---|
| 158 | $csv_file_content =~ s/\r/\n/g; # Handle non-Unix line endings
|
|---|
| 159 | $csv_file_content =~ s/\n+/\n/g;
|
|---|
| 160 |
|
|---|
| 161 | my $separate_char = $self->{'csv_field_separator'};
|
|---|
| 162 |
|
|---|
| 163 | my $md_val_sep = $self->{'metadata_value_separator'};
|
|---|
| 164 | undef $md_val_sep if ($md_val_sep eq "");
|
|---|
| 165 |
|
|---|
| 166 | my @csv_file_lines = split(/\n/, $csv_file_content);
|
|---|
| 167 | my $csv_file_field_line = shift(@csv_file_lines);
|
|---|
| 168 |
|
|---|
| 169 | if ($separate_char =~ m/^auto$/i) {
|
|---|
| 170 | $separate_char = $self->resolve_auto($csv_file_field_line,$self->{'plugin_type'});
|
|---|
| 171 | }
|
|---|
| 172 |
|
|---|
| 173 | my $csv = Text::CSV->new();
|
|---|
| 174 | $csv->sep_char($separate_char);
|
|---|
| 175 |
|
|---|
| 176 | my @csv_file_fields = undef;
|
|---|
| 177 | if ($csv->parse($csv_file_field_line)) {
|
|---|
| 178 | @csv_file_fields = $csv->fields;
|
|---|
| 179 | }
|
|---|
| 180 | else {
|
|---|
| 181 | $self->print_error($outhandle, $failhandle, $gli, $filename_full_path, "Error: Badly formatted CSV header line: $csv_file_field_line");
|
|---|
| 182 | return -1;
|
|---|
| 183 | }
|
|---|
| 184 |
|
|---|
| 185 | my $found_filename_field = 0;
|
|---|
| 186 | for (my $i = 0; $i < scalar(@csv_file_fields); $i++) {
|
|---|
| 187 | # Remove any spaces from the field names, and surrounding quotes too
|
|---|
| 188 | $csv_file_fields[$i] =~ s/ //g;
|
|---|
| 189 | $csv_file_fields[$i] =~ s/^"//;
|
|---|
| 190 | $csv_file_fields[$i] =~ s/"$//;
|
|---|
| 191 |
|
|---|
| 192 | if ($csv_file_fields[$i] eq "Filename") {
|
|---|
| 193 | $found_filename_field = 1;
|
|---|
| 194 | }
|
|---|
| 195 | }
|
|---|
| 196 |
|
|---|
| 197 | if (!defined $self->{'filename_formed_from'}) {
|
|---|
| 198 | if (!$found_filename_field) {
|
|---|
| 199 | $self->print_error($outhandle, $failhandle, $gli, $filename_full_path, "No Filename field in CSV file");
|
|---|
| 200 | return -1; # error
|
|---|
| 201 | }
|
|---|
| 202 | }
|
|---|
| 203 |
|
|---|
| 204 | # Read each line of the file and assign the metadata appropriately
|
|---|
| 205 | foreach my $csv_line (@csv_file_lines) {
|
|---|
| 206 | # Ignore lines containing only whitespace
|
|---|
| 207 | next if ($csv_line =~ /^\s*$/);
|
|---|
| 208 | my $orig_csv_line = $csv_line;
|
|---|
| 209 |
|
|---|
| 210 | # Build a hash of metadata name to metadata value for this line
|
|---|
| 211 | my %csv_line_metadata;
|
|---|
| 212 |
|
|---|
| 213 | if ($csv->parse($csv_line)) {
|
|---|
| 214 | my @md_vals = $csv->fields;
|
|---|
| 215 | my $md_vals_len = scalar(@md_vals);
|
|---|
| 216 |
|
|---|
| 217 | for (my $i=0; $i<$md_vals_len; $i++) {
|
|---|
| 218 | my $md_val = $md_vals[$i];
|
|---|
| 219 | # Only bother with non-empty values
|
|---|
| 220 | if ($md_val ne "" && defined($csv_file_fields[$i])) {
|
|---|
| 221 |
|
|---|
| 222 | my $md_name = $csv_file_fields[$i];
|
|---|
| 223 |
|
|---|
| 224 | if (!defined $md_name) {
|
|---|
| 225 | $csv_line_metadata{$md_name} = [];
|
|---|
| 226 | }
|
|---|
| 227 |
|
|---|
| 228 | if (defined $md_val_sep) {
|
|---|
| 229 |
|
|---|
| 230 | my @within_md_vals = split(/${md_val_sep}/,$md_val);
|
|---|
| 231 | #push (@{$csv_line_metadata{$md_name}}, @within_md_vals);
|
|---|
| 232 |
|
|---|
| 233 | # protect square brackets in metadata values by hex entity encoding them
|
|---|
| 234 | # As unescaped square bracket chars in metadata
|
|---|
| 235 | # have special meaning in GS' Java runtime code
|
|---|
| 236 | my @escaped_within_md_vals = ();
|
|---|
| 237 | for my $meta_value (@within_md_vals) {
|
|---|
| 238 | $meta_value =~ s/\[/&\#091;/g;
|
|---|
| 239 | $meta_value =~ s/\]/&\#093;/g;
|
|---|
| 240 | push(@escaped_within_md_vals, $meta_value);
|
|---|
| 241 | }
|
|---|
| 242 | push (@{$csv_line_metadata{$md_name}}, @escaped_within_md_vals);
|
|---|
| 243 |
|
|---|
| 244 | # foreach my $within_md_val (@within_md_vals) {
|
|---|
| 245 | # push (@{$csv_line_metadata{$md_name}}, $within_md_val);
|
|---|
| 246 | # }
|
|---|
| 247 | }
|
|---|
| 248 | else {
|
|---|
| 249 | #push (@{$csv_line_metadata{$md_name}}, $md_val);
|
|---|
| 250 | # protect square brackets in metadata values by hex entity encoding them
|
|---|
| 251 | my $escaped_metadata_value = $md_val;
|
|---|
| 252 | $escaped_metadata_value =~ s/\[/&\#091;/g;
|
|---|
| 253 | $escaped_metadata_value =~ s/\]/&\#093;/g;
|
|---|
| 254 | push (@{$csv_line_metadata{$md_name}}, $escaped_metadata_value);
|
|---|
| 255 | }
|
|---|
| 256 | }
|
|---|
| 257 | }
|
|---|
| 258 | }
|
|---|
| 259 | else {
|
|---|
| 260 | $self->print_error($outhandle, $failhandle, $gli, $filename_full_path, "Badly formatted CSV line: $csv_line");
|
|---|
| 261 | last;
|
|---|
| 262 | }
|
|---|
| 263 |
|
|---|
| 264 | # We can't associate any metadata without knowing the file to associate it with
|
|---|
| 265 | if (defined $self->{'filename_formed_from'}) {
|
|---|
| 266 | my @filename_formed_from_metanames = split(",",$self->{'filename_formed_from'});
|
|---|
| 267 |
|
|---|
| 268 | my $filename_metavalue = "";
|
|---|
| 269 | foreach my $metaname (@filename_formed_from_metanames) {
|
|---|
| 270 | my $metavalue = shift(@{$csv_line_metadata{$metaname}});
|
|---|
| 271 | if (defined $self->{'filename_formed_with_no_spaces'}) {
|
|---|
| 272 | $metavalue =~ s/\s+//g;
|
|---|
| 273 | }
|
|---|
| 274 |
|
|---|
| 275 |
|
|---|
| 276 | $filename_metavalue .= $metavalue if defined $metavalue;
|
|---|
| 277 | }
|
|---|
| 278 |
|
|---|
| 279 | if (defined $self->{'filename_formed_from_ext'}) {
|
|---|
| 280 | $filename_metavalue .= $self->{'filename_formed_from_ext'};
|
|---|
| 281 | }
|
|---|
| 282 |
|
|---|
| 283 | if ($filename_metavalue ne "") {
|
|---|
| 284 | # print STDERR "**** Setting up Metadata Filename match on: $filename_metavalue\n";
|
|---|
| 285 |
|
|---|
| 286 | $csv_line_metadata{"Filename"} = [ $filename_metavalue ];
|
|---|
| 287 | }
|
|---|
| 288 | }
|
|---|
| 289 |
|
|---|
| 290 | my $csv_line_filename_array = $csv_line_metadata{"Filename"};
|
|---|
| 291 | if (!defined $csv_line_filename_array) {
|
|---|
| 292 | $self->print_error($outhandle, $failhandle, $gli, $filename_full_path, "No Filename metadata in CSV line: $orig_csv_line");
|
|---|
| 293 | next;
|
|---|
| 294 | }
|
|---|
| 295 | my $csv_line_filename = shift(@$csv_line_filename_array);
|
|---|
| 296 | delete $csv_line_metadata{"Filename"};
|
|---|
| 297 |
|
|---|
| 298 | my $csv_line_section_array = $csv_line_metadata{"Section"};
|
|---|
| 299 | my $section_suffix = "";
|
|---|
| 300 | if (defined $csv_line_section_array) {
|
|---|
| 301 | my $section_value = shift(@$csv_line_section_array);
|
|---|
| 302 | if ($section_value =~ /[\d.]+/m){
|
|---|
| 303 | my $section_suffix = "///Section/" . $section_value;
|
|---|
| 304 | foreach my $metaname (keys %csv_line_metadata) {
|
|---|
| 305 | my $new_name = $metaname . $section_suffix;
|
|---|
| 306 | $csv_line_metadata{$new_name} = delete $csv_line_metadata{$metaname};
|
|---|
| 307 | }
|
|---|
| 308 | } else{
|
|---|
| 309 | unshift(@$csv_line_section_array, $section_value);
|
|---|
| 310 | }
|
|---|
| 311 | }
|
|---|
| 312 |
|
|---|
| 313 | # For CSV files to include line-breaks these are typically stored as '\n'
|
|---|
| 314 | # Unescape them here so they present more nicely in the browser
|
|---|
| 315 | #
|
|---|
| 316 | # foreach my $metaname (keys %csv_line_metadata) {
|
|---|
| 317 | # my $csv_line_metavalue_array = $csv_line_metadata{$metaname};
|
|---|
| 318 | #
|
|---|
| 319 | # map { $_ =~ s/\\n/\n/g } @$csv_line_metavalue_array;
|
|---|
| 320 | # }
|
|---|
| 321 |
|
|---|
| 322 |
|
|---|
| 323 | # Associate the metadata now
|
|---|
| 324 | # Indexing into the extrameta data structures requires the filename's style of slashes to be in URL format
|
|---|
| 325 | # Then need to convert the filename to a regex, no longer to protect windows directory chars \, but for
|
|---|
| 326 | # protecting special characters like brackets in the filepath such as "C:\Program Files (x86)\Greenstone".
|
|---|
| 327 | $csv_line_filename = &util::filepath_to_url_format($csv_line_filename);
|
|---|
| 328 | $csv_line_filename = &util::filename_to_regex($csv_line_filename);
|
|---|
| 329 |
|
|---|
| 330 | if (defined &extrametautil::getmetadata($extrametadata, $csv_line_filename)) { # merge with existing meta
|
|---|
| 331 |
|
|---|
| 332 | my $file_metadata_table = &extrametautil::getmetadata($extrametadata, $csv_line_filename);
|
|---|
| 333 |
|
|---|
| 334 | foreach my $metaname (keys %csv_line_metadata) {
|
|---|
| 335 | # will create new entry if one does not already exist
|
|---|
| 336 | push(@{$file_metadata_table->{$metaname}}, @{$csv_line_metadata{$metaname}});
|
|---|
| 337 | }
|
|---|
| 338 |
|
|---|
| 339 | # no need to push $file on to $extrametakeys as it is already in the list
|
|---|
| 340 | } else { # add as new meta
|
|---|
| 341 |
|
|---|
| 342 | &extrametautil::setmetadata($extrametadata, $csv_line_filename, \%csv_line_metadata);
|
|---|
| 343 | &extrametautil::addmetakey($extrametakeys, $csv_line_filename);
|
|---|
| 344 | }
|
|---|
| 345 | # record which file the metadata came from
|
|---|
| 346 | if (!defined &extrametautil::getmetafile($extrametafile, $csv_line_filename)) {
|
|---|
| 347 | &extrametautil::setmetafile($extrametafile, $csv_line_filename, {});
|
|---|
| 348 | }
|
|---|
| 349 | # maps the file to full path
|
|---|
| 350 | &extrametautil::setmetafile_for_named_file($extrametafile, $csv_line_filename, $file, $filename_full_path);
|
|---|
| 351 | }
|
|---|
| 352 | }
|
|---|
| 353 |
|
|---|
| 354 | sub print_error
|
|---|
| 355 | {
|
|---|
| 356 |
|
|---|
| 357 | my $self = shift(@_);
|
|---|
| 358 | my ($outhandle, $failhandle, $gli, $file, $error) = @_;
|
|---|
| 359 |
|
|---|
| 360 | print $outhandle "MetadataCSVPlugin Error: $file: $error\n";
|
|---|
| 361 | print $failhandle "MetadataCSVPlugin Error: $file: $error\n";
|
|---|
| 362 | print STDERR "<ProcessingError n='$file' r='$error'/>\n" if ($gli);
|
|---|
| 363 | }
|
|---|
| 364 | 1;
|
|---|
