[537] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
[31492] | 3 | # BaseImporter.pm -- base class for all the import plugins
|
---|
[537] | 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 | #
|
---|
[9413] | 8 | # Copyright (C) 1999-2005 New Zealand Digital Library Project
|
---|
[537] | 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 | ###########################################################################
|
---|
[4] | 25 |
|
---|
[31492] | 26 | package BaseImporter;
|
---|
[2219] | 27 |
|
---|
[10254] | 28 | use strict;
|
---|
| 29 | no strict 'subs';
|
---|
| 30 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
[9413] | 31 |
|
---|
[8892] | 32 | use File::Basename;
|
---|
[23335] | 33 | use Encode;
|
---|
[23832] | 34 | use Unicode::Normalize 'normalize';
|
---|
[8892] | 35 |
|
---|
[1870] | 36 | use encodings;
|
---|
[11389] | 37 | use unicode;
|
---|
[1242] | 38 | use doc;
|
---|
[2751] | 39 | use ghtml;
|
---|
[9413] | 40 | use gsprintf 'gsprintf';
|
---|
[27306] | 41 | use util;
|
---|
| 42 | use FileUtils;
|
---|
[4] | 43 |
|
---|
[31492] | 44 | use CommonUtil;
|
---|
[10218] | 45 |
|
---|
[15868] | 46 | BEGIN {
|
---|
[31492] | 47 | @BaseImporter::ISA = ( 'CommonUtil' );
|
---|
[15868] | 48 | }
|
---|
[5681] | 49 |
|
---|
[18320] | 50 | # the different methods that can be applied when renaming
|
---|
| 51 | # imported documents and their associated files
|
---|
| 52 | our $file_rename_method_list =
|
---|
| 53 | [ { 'name' => "url",
|
---|
[31492] | 54 | 'desc' => "{BaseImporter.rename_method.url}" },
|
---|
[18320] | 55 | { 'name' => "base64",
|
---|
[31492] | 56 | 'desc' => "{BaseImporter.rename_method.base64}" },
|
---|
[18404] | 57 | { 'name' => "none",
|
---|
[31492] | 58 | 'desc' => "{BaseImporter.rename_method.none}",
|
---|
[18398] | 59 | 'hiddengli' => "yes" } ];
|
---|
[18320] | 60 |
|
---|
[31457] | 61 | # here went encoding list stuff
|
---|
[3540] | 62 |
|
---|
[16698] | 63 | our $oidtype_list =
|
---|
[17026] | 64 | [ { 'name' => "auto",
|
---|
[31492] | 65 | 'desc' => "{BaseImporter.OIDtype.auto}" },
|
---|
[17026] | 66 | { 'name' => "hash",
|
---|
[16698] | 67 | 'desc' => "{import.OIDtype.hash}" },
|
---|
[24290] | 68 | { 'name' => "hash_on_ga_xml",
|
---|
| 69 | 'desc' => "{import.OIDtype.hash_on_ga_xml}" },
|
---|
[26536] | 70 | { 'name' => "hash_on_full_filename",
|
---|
| 71 | 'desc' => "{import.OIDtype.hash_on_full_filename}" },
|
---|
[16698] | 72 | { 'name' => "assigned",
|
---|
| 73 | 'desc' => "{import.OIDtype.assigned}" },
|
---|
| 74 | { 'name' => "incremental",
|
---|
| 75 | 'desc' => "{import.OIDtype.incremental}" },
|
---|
[26221] | 76 | { 'name' => "filename",
|
---|
| 77 | 'desc' => "{import.OIDtype.filename}" },
|
---|
[16698] | 78 | { 'name' => "dirname",
|
---|
[26536] | 79 | 'desc' => "{import.OIDtype.dirname}" },
|
---|
| 80 | { 'name' => "full_filename",
|
---|
| 81 | 'desc' => "{import.OIDtype.full_filename}" } ];
|
---|
[16698] | 82 |
|
---|
[4873] | 83 | my $arguments =
|
---|
[3540] | 84 | [ { 'name' => "process_exp",
|
---|
[31492] | 85 | 'desc' => "{BaseImporter.process_exp}",
|
---|
[6408] | 86 | 'type' => "regexp",
|
---|
[3540] | 87 | 'deft' => "",
|
---|
| 88 | 'reqd' => "no" },
|
---|
[31476] | 89 | { 'name' => "store_original_file",
|
---|
[31492] | 90 | 'desc' => "{BaseImporter.store_original_file}",
|
---|
[22215] | 91 | 'type' => "flag",
|
---|
| 92 | 'reqd' => "no" },
|
---|
[8892] | 93 | { 'name' => "associate_ext",
|
---|
[31492] | 94 | 'desc' => "{BaseImporter.associate_ext}",
|
---|
[8892] | 95 | 'type' => "string",
|
---|
| 96 | 'reqd' => "no" },
|
---|
[11122] | 97 | { 'name' => "associate_tail_re",
|
---|
[31492] | 98 | 'desc' => "{BaseImporter.associate_tail_re}",
|
---|
[11122] | 99 | 'type' => "string",
|
---|
| 100 | 'reqd' => "no" },
|
---|
[16698] | 101 | { 'name' => "OIDtype",
|
---|
| 102 | 'desc' => "{import.OIDtype}",
|
---|
| 103 | 'type' => "enum",
|
---|
| 104 | 'list' => $oidtype_list,
|
---|
[16847] | 105 | # leave default empty so we can tell if its been set or not - if not set will use option from import.pl
|
---|
[17026] | 106 | 'deft' => "auto",
|
---|
[18591] | 107 | 'reqd' => "no" },
|
---|
[16698] | 108 | { 'name' => "OIDmetadata",
|
---|
| 109 | 'desc' => "{import.OIDmetadata}",
|
---|
| 110 | 'type' => "metadata",
|
---|
| 111 | 'deft' => "dc.Identifier",
|
---|
[18591] | 112 | 'reqd' => "no" },
|
---|
[16698] | 113 | # { 'name' => "use_as_doc_identifier",
|
---|
[31492] | 114 | # 'desc' => "{BaseImporter.use_as_doc_identifier}",
|
---|
[16698] | 115 | # 'type' => "string",
|
---|
| 116 | # 'reqd' => "no" ,
|
---|
| 117 | # 'deft' => "" } ,
|
---|
[18320] | 118 | { 'name' => "no_cover_image",
|
---|
[31492] | 119 | 'desc' => "{BaseImporter.no_cover_image}",
|
---|
[3540] | 120 | 'type' => "flag",
|
---|
| 121 | 'reqd' => "no" },
|
---|
[31493] | 122 | { 'name' => "file_rename_method",
|
---|
[31492] | 123 | 'desc' => "{BaseImporter.file_rename_method}",
|
---|
[18320] | 124 | 'type' => "enum",
|
---|
| 125 | 'deft' => &get_default_file_rename_method(), # by default rename imported files and assoc files using this encoding
|
---|
| 126 | 'list' => $file_rename_method_list,
|
---|
| 127 | 'reqd' => "no"
|
---|
[23457] | 128 | }
|
---|
[15868] | 129 |
|
---|
| 130 | ];
|
---|
[3540] | 131 |
|
---|
[9398] | 132 |
|
---|
[31492] | 133 | my $options = { 'name' => "BaseImporter",
|
---|
| 134 | 'desc' => "{BaseImporter.desc}",
|
---|
[6408] | 135 | 'abstract' => "yes",
|
---|
[31457] | 136 | 'inherits' => "yes",
|
---|
[4750] | 137 | 'args' => $arguments };
|
---|
[3540] | 138 |
|
---|
[4] | 139 | sub new {
|
---|
[10218] | 140 |
|
---|
[15868] | 141 | my ($class) = shift (@_);
|
---|
[16698] | 142 | my ($pluginlist,$inputargs,$hashArgOptLists,$auxiliary) = @_;
|
---|
[10218] | 143 | push(@$pluginlist, $class);
|
---|
[9398] | 144 |
|
---|
[15868] | 145 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
| 146 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
[10218] | 147 |
|
---|
[31492] | 148 | my $self = new CommonUtil($pluginlist, $inputargs, $hashArgOptLists,$auxiliary);
|
---|
[16390] | 149 |
|
---|
| 150 | if ($self->{'info_only'}) {
|
---|
| 151 | # don't worry about any options etc
|
---|
| 152 | return bless $self, $class;
|
---|
| 153 | }
|
---|
| 154 |
|
---|
[15868] | 155 | my $plugin_name = (defined $pluginlist->[0]) ? $pluginlist->[0] : $class;
|
---|
| 156 | $self->{'plugin_type'} = $plugin_name;
|
---|
[10579] | 157 |
|
---|
[24404] | 158 | # remove ex. from OIDmetadata iff it's the only namespace prefix
|
---|
| 159 | $self->{'OIDmetadata'} =~ s/^ex\.([^.]+)$/$1/ if defined $self->{'OIDmetadata'};
|
---|
[2785] | 160 | $self->{'num_processed'} = 0;
|
---|
| 161 | $self->{'num_not_processed'} = 0;
|
---|
| 162 | $self->{'num_blocked'} = 0;
|
---|
| 163 | $self->{'num_archives'} = 0;
|
---|
[8678] | 164 | $self->{'cover_image'} = 1; # cover image is on by default
|
---|
[10218] | 165 | $self->{'cover_image'} = 0 if ($self->{'no_cover_image'});
|
---|
[19222] | 166 | $self->{'can_process_directories'} = 0;
|
---|
[10579] | 167 | #$self->{'option_list'} = $hashArgOptLists->{"OptList"};
|
---|
[3540] | 168 |
|
---|
[8892] | 169 | my $associate_ext = $self->{'associate_ext'};
|
---|
| 170 | if ((defined $associate_ext) && ($associate_ext ne "")) {
|
---|
[9351] | 171 |
|
---|
[11122] | 172 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 173 | if ((defined $associate_tail_re) && ($associate_tail_re ne "")) {
|
---|
| 174 | my $outhandle = $self->{'outhandle'};
|
---|
| 175 | print $outhandle "Warning: can only specify 'associate_ext' or 'associate_tail_re'\n";
|
---|
| 176 | print $outhandle " defaulting to 'associate_tail_re'\n";
|
---|
[8892] | 177 | }
|
---|
[11122] | 178 | else {
|
---|
| 179 | my @exts = split(/,/,$associate_ext);
|
---|
[8892] | 180 |
|
---|
[11122] | 181 | my @exts_bracketed = map { $_ = "(?:\\.$_)" } @exts;
|
---|
| 182 | my $associate_tail_re = join("|",@exts_bracketed);
|
---|
| 183 | $self->{'associate_tail_re'} = $associate_tail_re;
|
---|
| 184 | }
|
---|
| 185 |
|
---|
| 186 | delete $self->{'associate_ext'};
|
---|
[8892] | 187 | }
|
---|
| 188 |
|
---|
[15868] | 189 | return bless $self, $class;
|
---|
[11089] | 190 |
|
---|
[4] | 191 | }
|
---|
| 192 |
|
---|
[16821] | 193 | sub merge_inheritance
|
---|
| 194 | {
|
---|
| 195 | my $self = {};
|
---|
| 196 | my @child_selfs = @_;
|
---|
| 197 |
|
---|
| 198 | foreach my $child_self (@child_selfs) {
|
---|
| 199 | foreach my $key (keys %$child_self) {
|
---|
| 200 | if (defined $self->{$key}) {
|
---|
| 201 | if ($self->{$key} ne $child_self->{$key}) {
|
---|
| 202 | # print STDERR "Warning: Conflicting value in multiple inheritance for '$key'\n";
|
---|
| 203 | # print STDERR "Existing stored value = $self->{$key}\n";
|
---|
| 204 | # print STDERR "New (child) value = $child_self->{$key}\n";
|
---|
| 205 | # print STDERR "Keeping existing value\n";
|
---|
| 206 | # Existing value seems to be option specified in collect.cfg
|
---|
| 207 |
|
---|
| 208 | ### $self->{$key} = $child_self->{$key};
|
---|
| 209 |
|
---|
| 210 | }
|
---|
| 211 | else {
|
---|
| 212 | ## print STDERR "****Info: Value $self->{$key} for $key already defined through multiple inheritance as the same value\n";
|
---|
| 213 | }
|
---|
| 214 |
|
---|
| 215 | }
|
---|
| 216 | else {
|
---|
| 217 | $self->{$key} = $child_self->{$key};
|
---|
| 218 | }
|
---|
| 219 | }
|
---|
| 220 | }
|
---|
| 221 |
|
---|
| 222 | return $self;
|
---|
| 223 | }
|
---|
| 224 |
|
---|
[31492] | 225 | # initialize BaseImporter options
|
---|
| 226 | # if init() is overridden in a sub-class, remember to call BaseImporter::init()
|
---|
[1242] | 227 | sub init {
|
---|
| 228 | my $self = shift (@_);
|
---|
[2785] | 229 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
[16390] | 230 |
|
---|
[31457] | 231 | $self->SUPER::init(@_);
|
---|
| 232 |
|
---|
[1242] | 233 | # set process_exp and block_exp to defaults unless they were
|
---|
| 234 | # explicitly set
|
---|
[1244] | 235 |
|
---|
| 236 | if ((!$self->is_recursive()) and
|
---|
[1242] | 237 | (!defined $self->{'process_exp'}) || ($self->{'process_exp'} eq "")) {
|
---|
[1244] | 238 |
|
---|
[1242] | 239 | $self->{'process_exp'} = $self->get_default_process_exp ();
|
---|
| 240 | if ($self->{'process_exp'} eq "") {
|
---|
[1244] | 241 | warn ref($self) . " Warning: Non-recursive plugin has no process_exp\n";
|
---|
[1242] | 242 | }
|
---|
| 243 | }
|
---|
| 244 |
|
---|
| 245 | if ((!defined $self->{'block_exp'}) || ($self->{'block_exp'} eq "")) {
|
---|
| 246 | $self->{'block_exp'} = $self->get_default_block_exp ();
|
---|
| 247 | }
|
---|
[11089] | 248 |
|
---|
[1242] | 249 | }
|
---|
| 250 |
|
---|
[839] | 251 | sub begin {
|
---|
| 252 | my $self = shift (@_);
|
---|
| 253 | my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
|
---|
[16821] | 254 |
|
---|
[17026] | 255 | if ($self->{'OIDtype'} eq "auto") {
|
---|
| 256 | # hasn't been set in the plugin, use the processor values
|
---|
| 257 | $self->{'OIDtype'} = $processor->{'OIDtype'};
|
---|
| 258 | $self->{'OIDmetadata'} = $processor->{'OIDmetadata'};
|
---|
| 259 | }
|
---|
| 260 | if ($self->{'OIDtype'} eq "hash") {
|
---|
| 261 | # should we hash on the file or on the doc xml??
|
---|
| 262 | $self->{'OIDtype'} = $self->get_oid_hash_type();
|
---|
| 263 | if ($self->{'OIDtype'} !~ /^(hash_on_file|hash_on_ga_xml)$/) {
|
---|
| 264 | $self->{'OIDtype'} = "hash_on_file";
|
---|
| 265 | }
|
---|
| 266 | }
|
---|
[839] | 267 | }
|
---|
| 268 |
|
---|
[21308] | 269 | # This is called once if removeold is set with import.pl. Most plugins will do
|
---|
| 270 | # nothing but if a plugin does any stuff outside of creating doc obj, then
|
---|
| 271 | # it may need to clear something.
|
---|
| 272 | sub remove_all {
|
---|
[21286] | 273 | my $self = shift (@_);
|
---|
| 274 | my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
|
---|
| 275 | }
|
---|
| 276 |
|
---|
[21308] | 277 | # This is called per document for docs that have been deleted from the
|
---|
| 278 | # collection. Most plugins will do nothing
|
---|
| 279 | # but if a plugin does any stuff outside of creating doc obj, then it may need
|
---|
| 280 | # to clear something.
|
---|
| 281 | sub remove_one {
|
---|
| 282 | my $self = shift (@_);
|
---|
| 283 |
|
---|
[21315] | 284 | my ($file, $oids, $archivedir) = @_;
|
---|
| 285 | return 0 if $self->can_process_this_file($file);
|
---|
| 286 | return undef;
|
---|
[21308] | 287 | }
|
---|
| 288 |
|
---|
[839] | 289 | sub end {
|
---|
[10155] | 290 | # potentially called at the end of each plugin pass
|
---|
| 291 | # import.pl only has one plugin pass, but buildcol.pl has multiple ones
|
---|
| 292 |
|
---|
[15868] | 293 | my ($self) = shift (@_);
|
---|
[839] | 294 | }
|
---|
| 295 |
|
---|
[10155] | 296 | sub deinit {
|
---|
| 297 | # called only once, after all plugin passes have been done
|
---|
| 298 |
|
---|
| 299 | my ($self) = @_;
|
---|
| 300 | }
|
---|
| 301 |
|
---|
[17026] | 302 | # default hashing type is to hash on the original file (or converted file)
|
---|
| 303 | # override this to return hash_on_ga_xml for filetypes where hashing on the
|
---|
| 304 | # file is no good eg video
|
---|
| 305 | sub get_oid_hash_type {
|
---|
| 306 |
|
---|
| 307 | my $self = shift (@_);
|
---|
| 308 |
|
---|
| 309 | return "hash_on_file";
|
---|
| 310 | }
|
---|
| 311 |
|
---|
[15868] | 312 |
|
---|
[1242] | 313 | # this function should be overridden to return 1
|
---|
| 314 | # in recursive plugins
|
---|
[4] | 315 | sub is_recursive {
|
---|
| 316 | my $self = shift (@_);
|
---|
| 317 |
|
---|
[1242] | 318 | return 0;
|
---|
[4] | 319 | }
|
---|
| 320 |
|
---|
[1242] | 321 | sub get_default_block_exp {
|
---|
| 322 | my $self = shift (@_);
|
---|
| 323 |
|
---|
| 324 | return "";
|
---|
| 325 | }
|
---|
| 326 |
|
---|
| 327 | sub get_default_process_exp {
|
---|
| 328 | my $self = shift (@_);
|
---|
| 329 |
|
---|
| 330 | return "";
|
---|
| 331 | }
|
---|
| 332 |
|
---|
[23419] | 333 |
|
---|
[18320] | 334 | # rename imported files and assoc files using URL encoding by default
|
---|
| 335 | # as this will work for most plugins and give more legible filenames
|
---|
| 336 | sub get_default_file_rename_method() {
|
---|
| 337 | my $self = shift (@_);
|
---|
| 338 | return "url";
|
---|
| 339 | }
|
---|
| 340 |
|
---|
[18398] | 341 | # returns this plugin's active (possibly user-selected) file_rename_method
|
---|
| 342 | sub get_file_rename_method() {
|
---|
| 343 | my $self = shift (@_);
|
---|
| 344 | my $rename_method = $self->{'file_rename_method'};
|
---|
| 345 | if($rename_method) {
|
---|
| 346 | return $rename_method;
|
---|
| 347 | } else {
|
---|
| 348 | return $self->get_default_file_rename_method();
|
---|
| 349 | }
|
---|
| 350 | }
|
---|
| 351 |
|
---|
[16390] | 352 | # default implementation is to do nothing
|
---|
| 353 | sub store_block_files {
|
---|
| 354 |
|
---|
[9067] | 355 | my $self =shift (@_);
|
---|
[16390] | 356 | my ($filename_full_path, $block_hash) = @_;
|
---|
| 357 |
|
---|
[9067] | 358 | }
|
---|
| 359 |
|
---|
[16390] | 360 | # put files to block into hash
|
---|
| 361 | sub use_block_expressions {
|
---|
| 362 |
|
---|
| 363 | my $self =shift (@_);
|
---|
| 364 | my ($filename_full_path, $block_hash) = @_;
|
---|
| 365 |
|
---|
[23363] | 366 | $filename_full_path = &util::upgrade_if_dos_filename($filename_full_path);
|
---|
| 367 |
|
---|
[16390] | 368 | if ($self->{'block_exp'} ne "" && $filename_full_path =~ /$self->{'block_exp'}/) {
|
---|
[31476] | 369 | $self->block_filename($block_hash,$filename_full_path);
|
---|
[16390] | 370 | }
|
---|
| 371 |
|
---|
| 372 | }
|
---|
| 373 |
|
---|
[9067] | 374 | #default implementation is to block a file with same name as this, but extension jpg or JPG, if cover_images is on.
|
---|
| 375 | sub block_cover_image
|
---|
| 376 | {
|
---|
[10833] | 377 | my $self =shift;
|
---|
[16390] | 378 | my ($filename, $block_hash) = @_;
|
---|
[10833] | 379 |
|
---|
[23363] | 380 | $filename = &util::upgrade_if_dos_filename($filename);
|
---|
| 381 |
|
---|
[9067] | 382 | if ($self->{'cover_image'}) {
|
---|
| 383 | my $coverfile = $filename;
|
---|
| 384 | $coverfile =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
[23759] | 385 |
|
---|
[25743] | 386 | #if there is no file extension, coverfile will be the same as filename
|
---|
| 387 | return if $coverfile eq $filename;
|
---|
| 388 |
|
---|
[27306] | 389 | if (!&FileUtils::fileExists($coverfile)) {
|
---|
[9067] | 390 | $coverfile =~ s/jpg$/JPG/;
|
---|
| 391 | }
|
---|
[27306] | 392 | if (&FileUtils::fileExists($coverfile)) {
|
---|
[31476] | 393 | $self->block_filename($block_hash,$coverfile);
|
---|
[11089] | 394 | }
|
---|
[9067] | 395 | }
|
---|
| 396 |
|
---|
| 397 | return;
|
---|
| 398 | }
|
---|
[11122] | 399 |
|
---|
| 400 |
|
---|
[16390] | 401 | # discover all the files that should be blocked by this plugin
|
---|
| 402 | # check the args ...
|
---|
| 403 | sub file_block_read {
|
---|
[11122] | 404 |
|
---|
[8510] | 405 | my $self = shift (@_);
|
---|
[16390] | 406 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $gli) = @_;
|
---|
[8892] | 407 | # Keep track of filenames with same root but different extensions
|
---|
[11122] | 408 | # Used to support -associate_ext and the more generalised
|
---|
| 409 | # -associate_tail_re
|
---|
[16390] | 410 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
[8892] | 411 |
|
---|
[18441] | 412 | if (!-d $filename_full_path) {
|
---|
| 413 | $block_hash->{'all_files'}->{$file} = 1;
|
---|
| 414 | }
|
---|
| 415 |
|
---|
[11122] | 416 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 417 | if ((defined $associate_tail_re) && ($associate_tail_re ne "")) {
|
---|
| 418 | my ($file_prefix,$file_ext)
|
---|
[16390] | 419 | = &util::get_prefix_and_tail_by_regex($filename_full_path,$associate_tail_re);
|
---|
[8892] | 420 | if ((defined $file_prefix) && (defined $file_ext)) {
|
---|
[16390] | 421 | my $shared_fileroot = $block_hash->{'shared_fileroot'};
|
---|
[8892] | 422 | if (!defined $shared_fileroot->{$file_prefix}) {
|
---|
[11122] | 423 | my $file_prefix_rec = { 'tie_to' => undef,
|
---|
| 424 | 'exts' => {} };
|
---|
[8892] | 425 | $shared_fileroot->{$file_prefix} = $file_prefix_rec;
|
---|
| 426 | }
|
---|
| 427 |
|
---|
| 428 | my $file_prefix_rec = $shared_fileroot->{$file_prefix};
|
---|
| 429 |
|
---|
[30358] | 430 | if ($self->can_process_this_file($filename_full_path) && $file_ext !~ m/.\./) {
|
---|
[8892] | 431 | # This is the document the others should be tied to
|
---|
| 432 | $file_prefix_rec->{'tie_to'} = $file_ext;
|
---|
| 433 | }
|
---|
| 434 | else {
|
---|
[11122] | 435 | if ($file_ext =~ m/$associate_tail_re$/) {
|
---|
[16390] | 436 | # this file should be associated to the main one
|
---|
[9351] | 437 | $file_prefix_rec->{'exts'}->{$file_ext} = 1;
|
---|
| 438 | }
|
---|
[8892] | 439 | }
|
---|
[11122] | 440 |
|
---|
[8892] | 441 | }
|
---|
| 442 | }
|
---|
[11122] | 443 |
|
---|
[16390] | 444 | # check block expressions
|
---|
| 445 | $self->use_block_expressions($filename_full_path, $block_hash) unless $self->{'no_blocking'};
|
---|
| 446 |
|
---|
[9067] | 447 | # now check whether we are actually processing this
|
---|
[16390] | 448 | if (!-f $filename_full_path || !$self->can_process_this_file($filename_full_path)) {
|
---|
[9067] | 449 | return undef; # can't recognise
|
---|
| 450 | }
|
---|
[23457] | 451 |
|
---|
[16852] | 452 | # if we have a block_exp, then this overrides the normal 'smart' blocking
|
---|
| 453 | $self->store_block_files($filename_full_path, $block_hash) unless ($self->{'no_blocking'} || $self->{'block_exp'} ne "");
|
---|
[8892] | 454 |
|
---|
[11089] | 455 | # block the cover image if there is one
|
---|
| 456 | if ($self->{'cover_image'}) {
|
---|
[16852] | 457 | $self->block_cover_image($filename_full_path, $block_hash);
|
---|
[11089] | 458 | }
|
---|
[23457] | 459 |
|
---|
[9067] | 460 | return 1;
|
---|
[8510] | 461 | }
|
---|
| 462 |
|
---|
[16390] | 463 | # plugins that rely on more than process_exp (eg XML plugins) can override this method
|
---|
| 464 | sub can_process_this_file {
|
---|
| 465 | my $self = shift(@_);
|
---|
| 466 | my ($filename) = @_;
|
---|
[8892] | 467 |
|
---|
[19222] | 468 | if (-d $filename && !$self->{'can_process_directories'}) {
|
---|
| 469 | return 0;
|
---|
| 470 | }
|
---|
[22427] | 471 |
|
---|
[16390] | 472 | if ($self->{'process_exp'} ne "" && $filename =~ /$self->{'process_exp'}/) {
|
---|
| 473 | return 1;
|
---|
[8892] | 474 | }
|
---|
| 475 | return 0;
|
---|
[10280] | 476 |
|
---|
| 477 | }
|
---|
| 478 |
|
---|
[24403] | 479 | # Even if a plugin can extract metadata in its metadata_read pass,
|
---|
| 480 | # make the default return 'undef' so processing of the file continues
|
---|
| 481 | # down the pipeline, so other plugins can also have the opportunity to
|
---|
| 482 | # locate metadata and set it up in the extrametakeys variables that
|
---|
| 483 | # are passed around.
|
---|
| 484 |
|
---|
| 485 | sub can_process_this_file_for_metadata {
|
---|
| 486 | my $self = shift(@_);
|
---|
| 487 |
|
---|
| 488 | return undef;
|
---|
| 489 | }
|
---|
| 490 |
|
---|
| 491 |
|
---|
[10280] | 492 |
|
---|
[23335] | 493 | # Notionally written to be called once for each document, it is however safe to
|
---|
| 494 | # call multiple times (as in the case of ImagePlugin) which calls this later on
|
---|
| 495 | # after the original image has potentially been converted to a *new* source image
|
---|
| 496 | # format (e.g. TIFF to PNG)
|
---|
| 497 |
|
---|
[15868] | 498 | sub set_Source_metadata {
|
---|
| 499 | my $self = shift (@_);
|
---|
[23461] | 500 | my ($doc_obj, $raw_filename, $filename_encoding, $section) = @_;
|
---|
| 501 |
|
---|
[23335] | 502 | # 1. Sets the filename (Source) for display encoded as Unicode if possible,
|
---|
| 503 | # and (as a fallback) using %xx if not for non-ascii chars
|
---|
| 504 | # 2. Sets the url ref (SourceFile) to the URL encoded version
|
---|
| 505 | # of filename for generated files
|
---|
[23352] | 506 |
|
---|
| 507 | my ($unused_full_rf, $raw_file) = &util::get_full_filenames("", $raw_filename);
|
---|
[23335] | 508 |
|
---|
[23461] | 509 | my $this_section = (defined $section)? $section : $doc_obj->get_top_section();
|
---|
[23352] | 510 |
|
---|
| 511 | my $octet_file = $raw_file;
|
---|
| 512 |
|
---|
| 513 | # UTF-8 version of filename
|
---|
[23387] | 514 | # if ((defined $ENV{"DEBUG_UNICODE"}) && ($ENV{"DEBUG_UNICODE"})) {
|
---|
| 515 | # print STDERR "**** Setting Source Metadata given: $octet_file\n";
|
---|
| 516 | # }
|
---|
[14961] | 517 |
|
---|
[23352] | 518 | # Deal with (on Windows) raw filenames that are in their
|
---|
| 519 | # abbreviated DOS form
|
---|
| 520 |
|
---|
[28375] | 521 | if (($ENV{'GSDLOS'} =~ m/^windows$/i) && ($^O ne "cygwin")) {
|
---|
[23352] | 522 | if ((defined $filename_encoding) && ($filename_encoding eq "unicode")) {
|
---|
| 523 | if (-e $raw_filename) {
|
---|
| 524 | my $unicode_filename = Win32::GetLongPathName($raw_filename);
|
---|
| 525 |
|
---|
| 526 | my $unused_full_uf;
|
---|
| 527 | ($unused_full_uf, $octet_file) = &util::get_full_filenames("", $unicode_filename);
|
---|
| 528 | }
|
---|
[23347] | 529 | }
|
---|
[23352] | 530 | }
|
---|
[23335] | 531 |
|
---|
| 532 | my $url_encoded_filename;
|
---|
[23364] | 533 | if ((defined $filename_encoding) && ($filename_encoding ne "ascii")) {
|
---|
[23352] | 534 | # => Generate a pretty print version of filename that is mapped to Unicode
|
---|
| 535 |
|
---|
| 536 | # Use filename_encoding to map raw filename to a Perl unicode-aware string
|
---|
| 537 | $url_encoded_filename = decode($filename_encoding,$octet_file);
|
---|
[23335] | 538 | }
|
---|
| 539 | else {
|
---|
[23352] | 540 | # otherwise generate %xx encoded version of filename for char > 127
|
---|
| 541 | $url_encoded_filename = &unicode::raw_filename_to_url_encoded($octet_file);
|
---|
[23335] | 542 | }
|
---|
[18320] | 543 |
|
---|
[23387] | 544 | # if ((defined $ENV{"DEBUG_UNICODE"}) && ($ENV{"DEBUG_UNICODE"})) {
|
---|
| 545 | # print STDERR "****** saving Source as: $url_encoded_filename\n";
|
---|
| 546 | # }
|
---|
[23759] | 547 |
|
---|
| 548 | # In the case of converted files and (generalized) exploded documents, there
|
---|
| 549 | # will already be a source filename => store as OrigSource before overriding
|
---|
| 550 | my $orig_source = $doc_obj->get_metadata_element ($this_section, "Source");
|
---|
| 551 | if ((defined $orig_source) && ($orig_source !~ m/^\s*$/)) {
|
---|
| 552 | $doc_obj->set_utf8_metadata_element($this_section, "OrigSource", $orig_source);
|
---|
| 553 | }
|
---|
| 554 |
|
---|
[16919] | 555 | # Source is the UTF8 display name - not necessarily the name of the file on the system
|
---|
[23829] | 556 | if ($ENV{'GSDLOS'} =~ m/^darwin$/i) {
|
---|
| 557 | # on Darwin want all display strings to be in composed form, then can search on that
|
---|
| 558 | $url_encoded_filename = normalize('C', $url_encoded_filename); # Normalisation Form 'C' (composition)
|
---|
| 559 | }
|
---|
[31438] | 560 | # set_utf8_metadata actually sets perl unicode aware strings. not utf8
|
---|
[23461] | 561 | $doc_obj->set_utf8_metadata_element($this_section, "Source", $url_encoded_filename);
|
---|
[23759] | 562 |
|
---|
[23352] | 563 |
|
---|
[23335] | 564 | my $renamed_raw_file = &util::rename_file($raw_file, $self->{'file_rename_method'});
|
---|
[18320] | 565 | # If using URL encoding, then SourceFile is the url-reference to url-encoded
|
---|
[23335] | 566 | # renamed_raw_url: it's a url that refers to the actual file on the system
|
---|
[31438] | 567 | # this call just replaces % with %25
|
---|
[23335] | 568 | my $renamed_raw_url = &unicode::filename_to_url($renamed_raw_file);
|
---|
[18320] | 569 |
|
---|
[23461] | 570 | $doc_obj->set_utf8_metadata_element($this_section, "SourceFile",
|
---|
[23335] | 571 | $renamed_raw_url);
|
---|
| 572 |
|
---|
[23387] | 573 | # if ((defined $ENV{"DEBUG_UNICODE"}) && ($ENV{"DEBUG_UNICODE"})) {
|
---|
| 574 | # print STDERR "****** saving SourceFile as: $renamed_raw_url\n";
|
---|
| 575 | # }
|
---|
[14961] | 576 | }
|
---|
[23457] | 577 |
|
---|
[17026] | 578 | # this should be called by all plugins to set the oid of the doc obj, rather
|
---|
| 579 | # than calling doc_obj->set_OID directly
|
---|
[15868] | 580 | sub add_OID {
|
---|
[15018] | 581 | my $self = shift (@_);
|
---|
[26893] | 582 | my ($doc_obj, $force) = @_;
|
---|
[15018] | 583 |
|
---|
[26893] | 584 | # don't add one if there is one already set, unless we are forced to do so
|
---|
| 585 | return unless ($doc_obj->get_OID() =~ /^NULL$/ || $force);
|
---|
[17026] | 586 | $doc_obj->set_OIDtype($self->{'OIDtype'}, $self->{'OIDmetadata'});
|
---|
[15018] | 587 |
|
---|
[17026] | 588 | # see if there is a plugin specific set_OID function
|
---|
| 589 | if (defined ($self->can('set_OID'))) {
|
---|
| 590 | $self->set_OID(@_); # pass through doc_obj and any extra arguments
|
---|
[15018] | 591 | }
|
---|
[17026] | 592 | else {
|
---|
[15018] | 593 | # use the default set_OID() in doc.pm
|
---|
| 594 | $doc_obj->set_OID();
|
---|
| 595 | }
|
---|
[17026] | 596 |
|
---|
[15018] | 597 | }
|
---|
[23457] | 598 |
|
---|
[31492] | 599 | # The BaseImporter read_into_doc_obj() function. This function does all the
|
---|
[15868] | 600 | # right things to make general options work for a given plugin. It doesn't do anything with the file other than setting reads in
|
---|
[10280] | 601 | # a file and sets up a slew of metadata all saved in doc_obj, which
|
---|
| 602 | # it then returns as part of a tuple (process_status,doc_obj)
|
---|
| 603 | #
|
---|
| 604 | # Much of this functionality used to reside in read, but it was broken
|
---|
| 605 | # down into a supporting routine to make the code more flexible.
|
---|
| 606 | #
|
---|
| 607 | # recursive plugins (e.g. RecPlug) and specialized plugins like those
|
---|
| 608 | # capable of processing many documents within a single file (e.g.
|
---|
| 609 | # GMLPlug) will normally want to implement their own version of
|
---|
| 610 | # read_into_doc_obj()
|
---|
| 611 | #
|
---|
| 612 | # Note that $base_dir might be "" and that $file might
|
---|
| 613 | # include directories
|
---|
[15868] | 614 |
|
---|
| 615 | # currently blocking has been done before it gets here - does this affect secondary plugin stuff??
|
---|
[10280] | 616 | sub read_into_doc_obj {
|
---|
| 617 | my $self = shift (@_);
|
---|
[16390] | 618 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
[10280] | 619 |
|
---|
[15868] | 620 | my $outhandle = $self->{'outhandle'};
|
---|
[10280] | 621 |
|
---|
[15868] | 622 | # should we move this to read? What about secondary plugins?
|
---|
[23759] | 623 | my $pp_file = &util::prettyprint_file($base_dir,$file,$gli);
|
---|
[15868] | 624 | print STDERR "<Processing n='$file' p='$self->{'plugin_type'}'>\n" if ($gli);
|
---|
[23363] | 625 | print $outhandle "$self->{'plugin_type'} processing $pp_file\n"
|
---|
[23457] | 626 | if $self->{'verbosity'} > 1;
|
---|
[10280] | 627 |
|
---|
[16390] | 628 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
[23457] | 629 |
|
---|
[1242] | 630 | # create a new document
|
---|
[18320] | 631 | my $doc_obj = new doc ($filename_full_path, "indexed_doc", $self->{'file_rename_method'});
|
---|
[14961] | 632 | my $top_section = $doc_obj->get_top_section();
|
---|
| 633 |
|
---|
| 634 | $doc_obj->add_utf8_metadata($top_section, "Plugin", "$self->{'plugin_type'}");
|
---|
[15868] | 635 | $doc_obj->add_utf8_metadata($top_section, "FileSize", (-s $filename_full_path));
|
---|
[23457] | 636 |
|
---|
[18469] | 637 |
|
---|
[23352] | 638 | my $plugin_filename_encoding = $self->{'filename_encoding'};
|
---|
[23347] | 639 | my $filename_encoding = $self->deduce_filename_encoding($file,$metadata,$plugin_filename_encoding);
|
---|
[23472] | 640 | $self->set_Source_metadata($doc_obj,$filename_full_path,$filename_encoding,$top_section);
|
---|
[8166] | 641 |
|
---|
[15868] | 642 | # plugin specific stuff - what args do we need here??
|
---|
| 643 | unless (defined ($self->process($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli))) {
|
---|
| 644 | print STDERR "<ProcessingError n='$file'>\n" if ($gli);
|
---|
| 645 | return -1;
|
---|
[2816] | 646 | }
|
---|
[1242] | 647 |
|
---|
[15868] | 648 | # include any metadata passed in from previous plugins
|
---|
| 649 | # note that this metadata is associated with the top level section
|
---|
| 650 | my $section = $doc_obj->get_top_section();
|
---|
| 651 | # can we merge these two methods??
|
---|
| 652 | $self->add_associated_files($doc_obj, $filename_full_path);
|
---|
| 653 | $self->extra_metadata ($doc_obj, $section, $metadata);
|
---|
| 654 | $self->auto_extract_metadata($doc_obj);
|
---|
[1242] | 655 |
|
---|
[15868] | 656 | # if we haven't found any Title so far, assign one
|
---|
| 657 | # this was shifted to here from inside read()
|
---|
| 658 | $self->title_fallback($doc_obj,$section,$filename_no_path);
|
---|
| 659 |
|
---|
| 660 | $self->add_OID($doc_obj);
|
---|
| 661 |
|
---|
[21219] | 662 | $self->post_process_doc_obj($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli);
|
---|
[15868] | 663 | return (1,$doc_obj);
|
---|
| 664 | }
|
---|
[2785] | 665 |
|
---|
[21219] | 666 | sub post_process_doc_obj {
|
---|
| 667 | my $self = shift (@_);
|
---|
| 668 | my ($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
---|
| 669 |
|
---|
| 670 | return 1;
|
---|
| 671 | }
|
---|
| 672 |
|
---|
[15868] | 673 | sub add_dummy_text {
|
---|
| 674 | my $self = shift(@_);
|
---|
| 675 | my ($doc_obj, $section) = @_;
|
---|
[2785] | 676 |
|
---|
[15868] | 677 | # add NoText metadata so we can hide this dummy text in format statements
|
---|
| 678 | $doc_obj->add_metadata($section, "NoText", "1");
|
---|
[31420] | 679 |
|
---|
| 680 | # lookup_string with extra '1' arg returns perl internal unicode aware text, so we use add_utf8_text so no encoding is done on it.
|
---|
[31492] | 681 | $doc_obj->add_utf8_text($section, &gsprintf::lookup_string("{BaseImporter.dummy_text}",1));
|
---|
| 682 | #$doc_obj->add_text($section, &gsprintf::lookup_string("{BaseImporter.dummy_text}",1));
|
---|
[15868] | 683 |
|
---|
[31420] | 684 |
|
---|
[15868] | 685 | }
|
---|
[8510] | 686 |
|
---|
[15868] | 687 | # does nothing. Can be overridden by subclass
|
---|
| 688 | sub auto_extract_metadata {
|
---|
| 689 | my $self = shift(@_);
|
---|
| 690 | my ($doc_obj) = @_;
|
---|
| 691 | }
|
---|
[11122] | 692 |
|
---|
[15868] | 693 | # adds cover image, associate_file options stuff. Should be called by sub class
|
---|
| 694 | # read_into_doc_obj
|
---|
| 695 | sub add_associated_files {
|
---|
| 696 | my $self = shift(@_);
|
---|
| 697 | # whatis filename??
|
---|
| 698 | my ($doc_obj, $filename) = @_;
|
---|
| 699 |
|
---|
| 700 | # add in the cover image
|
---|
| 701 | if ($self->{'cover_image'}) {
|
---|
| 702 | $self->associate_cover_image($doc_obj, $filename);
|
---|
[8716] | 703 | }
|
---|
[22215] | 704 | # store the original (used for eg TextPlugin to store the original for OAI)
|
---|
| 705 | if ($self->{'store_original_file'}) {
|
---|
| 706 | $self->associate_source_file($doc_obj, $filename);
|
---|
| 707 | }
|
---|
[23457] | 708 |
|
---|
[15018] | 709 |
|
---|
[10280] | 710 | }
|
---|
[1242] | 711 |
|
---|
[16390] | 712 | # implement this if you are extracting metadata for other documents
|
---|
| 713 | sub metadata_read {
|
---|
| 714 | my $self = shift (@_);
|
---|
[19493] | 715 | my ($pluginfo, $base_dir, $file, $block_hash,
|
---|
| 716 | $extrametakeys, $extrametadata, $extrametafile,
|
---|
[23212] | 717 | $processor, $gli, $aux) = @_;
|
---|
[16390] | 718 |
|
---|
| 719 | # can we process this file??
|
---|
| 720 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
[24403] | 721 | return undef unless $self->can_process_this_file_for_metadata($filename_full_path);
|
---|
[16390] | 722 |
|
---|
| 723 | return 1; # we recognise the file, but don't actually do anything with it
|
---|
| 724 | }
|
---|
| 725 |
|
---|
| 726 |
|
---|
[31492] | 727 | # The BaseImporter read() function. This function calls read_into_doc_obj()
|
---|
[10280] | 728 | # to ensure all the right things to make general options work for a
|
---|
| 729 | # given plugin are done. It then calls the process() function which
|
---|
| 730 | # does all the work specific to a plugin (like the old read functions
|
---|
| 731 | # used to do). Most plugins should define their own process() function
|
---|
| 732 | # and let this read() function keep control.
|
---|
| 733 | #
|
---|
| 734 | # recursive plugins (e.g. RecPlug) and specialized plugins like those
|
---|
| 735 | # capable of processing many documents within a single file (e.g.
|
---|
| 736 | # GMLPlug) might want to implement their own version of read(), but
|
---|
| 737 | # more likely need to implement their own version of read_into_doc_obj()
|
---|
| 738 | #
|
---|
| 739 | # Return number of files processed, undef if can't recognise, -1 if can't
|
---|
| 740 | # process
|
---|
| 741 |
|
---|
| 742 | sub read {
|
---|
| 743 | my $self = shift (@_);
|
---|
[16390] | 744 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
[10280] | 745 |
|
---|
[16390] | 746 | # can we process this file??
|
---|
| 747 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
[20577] | 748 |
|
---|
[16390] | 749 | return undef unless $self->can_process_this_file($filename_full_path);
|
---|
| 750 |
|
---|
[31445] | 751 | #print STDERR "**** BEFORE READ INTO DOC OBJ: $file\n";
|
---|
[10280] | 752 | my ($process_status,$doc_obj) = $self->read_into_doc_obj(@_);
|
---|
[31445] | 753 | #print STDERR "**** AFTER READ INTO DOC OBJ: $file\n";
|
---|
| 754 |
|
---|
[10280] | 755 | if ((defined $process_status) && ($process_status == 1)) {
|
---|
[21219] | 756 |
|
---|
[10280] | 757 | # process the document
|
---|
| 758 | $processor->process($doc_obj);
|
---|
[15868] | 759 |
|
---|
[10280] | 760 | $self->{'num_processed'} ++;
|
---|
| 761 | undef $doc_obj;
|
---|
[9398] | 762 | }
|
---|
[15868] | 763 | # delete any temp files that we may have created
|
---|
| 764 | $self->clean_up_after_doc_obj_processing();
|
---|
[9398] | 765 |
|
---|
[18469] | 766 |
|
---|
[10280] | 767 | # if process_status == 1, then the file has been processed.
|
---|
| 768 | return $process_status;
|
---|
| 769 |
|
---|
[4] | 770 | }
|
---|
| 771 |
|
---|
[1244] | 772 | # returns undef if file is rejected by the plugin
|
---|
[1242] | 773 | sub process {
|
---|
| 774 | my $self = shift (@_);
|
---|
[11089] | 775 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
---|
[1242] | 776 |
|
---|
[31492] | 777 | gsprintf(STDERR, "BaseImporter::process {common.must_be_implemented}\n");
|
---|
[1244] | 778 |
|
---|
[23457] | 779 | my ($cpackage,$cfilename,$cline,$csubr,$chas_args,$cwantarray) = caller(1);
|
---|
[23419] | 780 | print STDERR "Calling method: $cfilename:$cline $cpackage->$csubr\n";
|
---|
| 781 |
|
---|
[23457] | 782 | die "\n";
|
---|
[23419] | 783 |
|
---|
[1244] | 784 | return undef; # never gets here
|
---|
[1242] | 785 | }
|
---|
| 786 |
|
---|
[15868] | 787 | # overwrite this method to delete any temp files that we have created
|
---|
| 788 | sub clean_up_after_doc_obj_processing {
|
---|
| 789 | my $self = shift(@_);
|
---|
[4] | 790 |
|
---|
[10280] | 791 | }
|
---|
[16390] | 792 |
|
---|
[1219] | 793 |
|
---|
[10280] | 794 |
|
---|
[7504] | 795 | sub filename_based_title
|
---|
| 796 | {
|
---|
| 797 | my $self = shift (@_);
|
---|
| 798 | my ($file) = @_;
|
---|
| 799 |
|
---|
| 800 | my $file_derived_title = $file;
|
---|
| 801 | $file_derived_title =~ s/_/ /g;
|
---|
[23335] | 802 | $file_derived_title =~ s/\.[^.]+$//;
|
---|
[7504] | 803 |
|
---|
| 804 | return $file_derived_title;
|
---|
| 805 | }
|
---|
| 806 |
|
---|
[9398] | 807 |
|
---|
[7504] | 808 | sub title_fallback
|
---|
| 809 | {
|
---|
| 810 | my $self = shift (@_);
|
---|
| 811 | my ($doc_obj,$section,$file) = @_;
|
---|
| 812 |
|
---|
[22705] | 813 | if (!defined $doc_obj->get_metadata_element ($section, "Title")
|
---|
[23335] | 814 | || $doc_obj->get_metadata_element($section, "Title") eq "") {
|
---|
[7504] | 815 |
|
---|
[23335] | 816 | my $source_file = $doc_obj->get_metadata_element($section, "Source");
|
---|
| 817 | my $file_derived_title;
|
---|
| 818 | if (defined $source_file) {
|
---|
| 819 | $file_derived_title = $self->filename_based_title($source_file);
|
---|
| 820 | }
|
---|
| 821 | else {
|
---|
| 822 | # pp = pretty print
|
---|
| 823 | my $pp_file = (defined $source_file) ? $source_file : $file;
|
---|
| 824 |
|
---|
| 825 | my $raw_title = $self->filename_based_title($file);
|
---|
| 826 | my $file_derived_title = &unicode::raw_filename_to_url_encoded($raw_title);
|
---|
| 827 | }
|
---|
| 828 |
|
---|
| 829 |
|
---|
[15868] | 830 | if (!defined $doc_obj->get_metadata_element ($section, "Title")) {
|
---|
| 831 | $doc_obj->add_utf8_metadata ($section, "Title", $file_derived_title);
|
---|
[9413] | 832 | }
|
---|
[15868] | 833 | else {
|
---|
[16995] | 834 | $doc_obj->set_utf8_metadata_element ($section, "Title", $file_derived_title);
|
---|
[15868] | 835 | }
|
---|
[9413] | 836 | }
|
---|
[23457] | 837 |
|
---|
[1844] | 838 | }
|
---|
[23457] | 839 |
|
---|
[1219] | 840 | # add any extra metadata that's been passed around from one
|
---|
| 841 | # plugin to another.
|
---|
| 842 | # extra_metadata uses add_utf8_metadata so it expects metadata values
|
---|
| 843 | # to already be in utf8
|
---|
| 844 | sub extra_metadata {
|
---|
| 845 | my $self = shift (@_);
|
---|
| 846 | my ($doc_obj, $cursection, $metadata) = @_;
|
---|
| 847 |
|
---|
[11122] | 848 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 849 |
|
---|
[27949] | 850 | # Sort the extra metadata for diffcol so these meta appear in a consistent order
|
---|
| 851 | # in doc.xml. Necessary for the ex.PDF.* and ex.File.* meta that's extracted in
|
---|
| 852 | # the PDFBox collection, as the order of these varies between CentOS and Ubuntu.
|
---|
[27973] | 853 | foreach my $field (sort keys(%$metadata)) {
|
---|
| 854 | # foreach my $field (keys(%$metadata)) {
|
---|
[839] | 855 | # $metadata->{$field} may be an array reference
|
---|
[8510] | 856 | if ($field eq "gsdlassocfile_tobe") {
|
---|
| 857 | # 'gsdlassocfile_tobe' is artificially introduced metadata
|
---|
| 858 | # that is used to signal that certain additional files should
|
---|
| 859 | # be tied to this document. Useful in situations where a
|
---|
| 860 | # metadata pass in the plugin pipeline works out some files
|
---|
| 861 | # need to be associated with a document, but the document hasn't
|
---|
| 862 | # been formed yet.
|
---|
| 863 | my $equiv_form = "";
|
---|
| 864 | foreach my $gaf (@{$metadata->{$field}}) {
|
---|
| 865 | my ($full_filename,$mimetype) = ($gaf =~ m/^(.*):(.*):$/);
|
---|
[18171] | 866 | my ($tail_filename) = ($full_filename =~ /^.*[\/\\](.+?)$/);
|
---|
| 867 |
|
---|
| 868 | # we need to make sure the filename is valid utf-8 - we do
|
---|
[18320] | 869 | # this by url or base64 encoding it
|
---|
[18171] | 870 | # $tail_filename is the name that we store the file as
|
---|
[18320] | 871 | $tail_filename = &util::rename_file($tail_filename, $self->{'file_rename_method'});
|
---|
[8510] | 872 | $doc_obj->associate_file($full_filename,$tail_filename,$mimetype);
|
---|
[20778] | 873 | $doc_obj->associate_source_file($full_filename);
|
---|
[18320] | 874 | # If the filename is url_encoded, we need to encode the % signs
|
---|
| 875 | # in the filename, so that it works in a url
|
---|
[18404] | 876 | my $url_tail_filename = &unicode::filename_to_url($tail_filename);
|
---|
[11122] | 877 | # work out extended tail extension (i.e. matching tail re)
|
---|
| 878 |
|
---|
| 879 | my ($file_prefix,$file_extended_ext)
|
---|
[16390] | 880 | = &util::get_prefix_and_tail_by_regex($tail_filename,$associate_tail_re);
|
---|
[11122] | 881 | my ($pre_doc_ext) = ($file_extended_ext =~ m/^(.*)\..*$/);
|
---|
[8510] | 882 | my ($doc_ext) = ($tail_filename =~ m/^.*\.(.*)$/);
|
---|
[24219] | 883 |
|
---|
[25345] | 884 | # the greenstone 2 stuff
|
---|
| 885 | my $start_doclink = "<a href=\"_httpprefix_/collect/[collection]/index/assoc/{Or}{[parent(Top):assocfilepath],[assocfilepath]}/$url_tail_filename\">";
|
---|
| 886 | #my $start_doclink = "<a href=\'_httpprefix_/collect/[collection]/index/assoc/[assocfilepath]/$url_tail_filename\'>";
|
---|
[24219] | 887 | my $start_doclink_gs3 = "<a href=\'_httpprefix_/collect/[collection]/index/assoc/[assocfilepath]/$url_tail_filename\'>";
|
---|
| 888 |
|
---|
[8510] | 889 | my $srcicon = "_icon".$doc_ext."_";
|
---|
| 890 | my $end_doclink = "</a>";
|
---|
[24219] | 891 |
|
---|
[11122] | 892 | my $assoc_form = "$start_doclink\{If\}{$srcicon,$srcicon,$doc_ext\}$end_doclink";
|
---|
| 893 |
|
---|
[24219] | 894 |
|
---|
[18171] | 895 | if (defined $pre_doc_ext && $pre_doc_ext ne "") {
|
---|
[11122] | 896 | # for metadata such as [mp3._edited] [mp3._full] ...
|
---|
| 897 | $doc_obj->add_utf8_metadata ($cursection, "$doc_ext.$pre_doc_ext", $assoc_form);
|
---|
| 898 | }
|
---|
| 899 |
|
---|
| 900 | # for multiple metadata such as [mp3.assoclink]
|
---|
| 901 | $doc_obj->add_utf8_metadata ($cursection, "$doc_ext.assoclink", $assoc_form);
|
---|
[24219] | 902 |
|
---|
[11122] | 903 | $equiv_form .= " $assoc_form";
|
---|
[25345] | 904 |
|
---|
| 905 | # following are used for greenstone 3,
|
---|
| 906 | $doc_obj->add_utf8_metadata ($cursection, "equivDocLink", $start_doclink_gs3);
|
---|
| 907 | $doc_obj->add_utf8_metadata ($cursection, "equivDocIcon", $srcicon);
|
---|
| 908 | $doc_obj->add_utf8_metadata ($cursection, "/equivDocLink", $end_doclink);
|
---|
| 909 |
|
---|
[8510] | 910 | }
|
---|
| 911 | $doc_obj->add_utf8_metadata ($cursection, "equivlink", $equiv_form);
|
---|
| 912 | }
|
---|
[23279] | 913 | elsif ($field eq "gsdlzipfilename") {
|
---|
| 914 | # special case for when files have come out of a zip. source_path
|
---|
| 915 | # (used for archives dbs and keeping track for incremental import)
|
---|
| 916 | # must be set to the zip file name
|
---|
| 917 | my $zip_filename = $metadata->{$field};
|
---|
| 918 | # overwrite the source_path
|
---|
[23363] | 919 | $doc_obj->set_source_path($zip_filename);
|
---|
[23279] | 920 | # and set the metadata
|
---|
| 921 | $zip_filename = &util::filename_within_collection($zip_filename);
|
---|
| 922 | $zip_filename = $doc_obj->encode_filename($zip_filename, $self->{'file_rename_method'});
|
---|
| 923 | $doc_obj->add_utf8_metadata ($cursection, $field, $zip_filename);
|
---|
| 924 | }
|
---|
[8510] | 925 | elsif (ref ($metadata->{$field}) eq "ARRAY") {
|
---|
[32501] | 926 | if ($field =~ /(.+?)\/\/\/Section\/([\d.]*)/m){
|
---|
| 927 | my $field_new_name = $1;
|
---|
| 928 | my $specified_section = $2;
|
---|
| 929 | map {
|
---|
| 930 | $doc_obj->add_utf8_metadata ($specified_section, $field_new_name, $_);
|
---|
| 931 | } @{$metadata->{$field}};
|
---|
| 932 | } else {
|
---|
| 933 | map {
|
---|
| 934 | $doc_obj->add_utf8_metadata ($cursection, $field, $_);
|
---|
| 935 | } @{$metadata->{$field}};
|
---|
| 936 | }
|
---|
| 937 |
|
---|
[839] | 938 | } else {
|
---|
[32501] | 939 | if ($field =~ /(.+?)\/\/\/Section\/([\d.]*)/m){
|
---|
| 940 | my $field_new_name = $1;
|
---|
| 941 | my $specified_section = $2;
|
---|
| 942 | $doc_obj->add_utf8_metadata ($specified_section, $field_new_name, $metadata->{$field});
|
---|
| 943 | } else {
|
---|
| 944 | $doc_obj->add_utf8_metadata ($cursection, $field, $metadata->{$field});
|
---|
| 945 | }
|
---|
[839] | 946 | }
|
---|
| 947 | }
|
---|
| 948 | }
|
---|
| 949 |
|
---|
[1396] | 950 |
|
---|
[2785] | 951 | sub compile_stats {
|
---|
| 952 | my $self = shift(@_);
|
---|
| 953 | my ($stats) = @_;
|
---|
| 954 |
|
---|
| 955 | $stats->{'num_processed'} += $self->{'num_processed'};
|
---|
| 956 | $stats->{'num_not_processed'} += $self->{'num_not_processed'};
|
---|
[2796] | 957 | $stats->{'num_archives'} += $self->{'num_archives'};
|
---|
[2785] | 958 |
|
---|
| 959 | }
|
---|
[22215] | 960 | sub associate_source_file {
|
---|
| 961 | my $self = shift(@_);
|
---|
| 962 |
|
---|
| 963 | my ($doc_obj, $filename) = @_;
|
---|
| 964 | my $cursection = $doc_obj->get_top_section();
|
---|
| 965 | my $assocfilename = $doc_obj->get_assocfile_from_sourcefile();
|
---|
| 966 |
|
---|
| 967 | $doc_obj->associate_file($filename, $assocfilename, undef, $cursection);
|
---|
[24225] | 968 | # srclink_file is now deprecated because of the "_" in the metadataname. Use srclinkFile
|
---|
[22663] | 969 | $doc_obj->add_utf8_metadata ($cursection, "srclink_file", $doc_obj->get_sourcefile());
|
---|
[24225] | 970 | $doc_obj->add_utf8_metadata ($cursection, "srclinkFile", $doc_obj->get_sourcefile());
|
---|
[22215] | 971 | }
|
---|
[2785] | 972 |
|
---|
[2816] | 973 | sub associate_cover_image {
|
---|
[22215] | 974 | my $self = shift(@_);
|
---|
[2816] | 975 | my ($doc_obj, $filename) = @_;
|
---|
| 976 |
|
---|
[23363] | 977 | my $upgraded_filename = &util::upgrade_if_dos_filename($filename);
|
---|
| 978 |
|
---|
[10833] | 979 | $filename =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
[23363] | 980 | $upgraded_filename =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
| 981 |
|
---|
| 982 | if (exists $self->{'covers_missing_cache'}->{$upgraded_filename}) {
|
---|
| 983 | # don't stat() for existence e.g. for multiple document input files
|
---|
[10833] | 984 | # (eg SplitPlug)
|
---|
| 985 | return;
|
---|
| 986 | }
|
---|
| 987 |
|
---|
[9413] | 988 | my $top_section=$doc_obj->get_top_section();
|
---|
| 989 |
|
---|
[27306] | 990 | if (&FileUtils::fileExists($upgraded_filename)) {
|
---|
[20778] | 991 | $doc_obj->associate_source_file($filename);
|
---|
[13968] | 992 | $doc_obj->associate_file($filename, "cover.jpg", "image/jpeg");
|
---|
[9413] | 993 | $doc_obj->add_utf8_metadata($top_section, "hascover", 1);
|
---|
[3086] | 994 | } else {
|
---|
[10833] | 995 | my $upper_filename = $filename;
|
---|
[23363] | 996 | my $upgraded_upper_filename = $upgraded_filename;
|
---|
| 997 |
|
---|
[10833] | 998 | $upper_filename =~ s/jpg$/JPG/;
|
---|
[23363] | 999 | $upgraded_upper_filename =~ s/jpg$/JPG/;
|
---|
| 1000 |
|
---|
[27306] | 1001 | if (&FileUtils::fileExists($upgraded_upper_filename)) {
|
---|
[20778] | 1002 | $doc_obj->associate_source_file($upper_filename);
|
---|
[10833] | 1003 | $doc_obj->associate_file($upper_filename, "cover.jpg",
|
---|
| 1004 | "image/jpeg");
|
---|
[9413] | 1005 | $doc_obj->add_utf8_metadata($top_section, "hascover", 1);
|
---|
[10833] | 1006 | } else {
|
---|
| 1007 | # file doesn't exist, so record the fact that it's missing so
|
---|
| 1008 | # we don't stat() again (stat is slow)
|
---|
[23363] | 1009 | $self->{'covers_missing_cache'}->{$upgraded_filename} = 1;
|
---|
[3086] | 1010 | }
|
---|
[2816] | 1011 | }
|
---|
[10833] | 1012 |
|
---|
[2816] | 1013 | }
|
---|
| 1014 |
|
---|
[11332] | 1015 |
|
---|
| 1016 | # Overridden by exploding plugins (eg. ISISPlug)
|
---|
| 1017 | sub clean_up_after_exploding
|
---|
| 1018 | {
|
---|
| 1019 | my $self = shift(@_);
|
---|
| 1020 | }
|
---|
| 1021 |
|
---|
| 1022 |
|
---|
[16390] | 1023 |
|
---|
[4] | 1024 | 1;
|
---|