[537] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
[15868] | 3 | # BasePlugin.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 |
|
---|
[15868] | 26 | package BasePlugin;
|
---|
[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';
|
---|
[4] | 41 |
|
---|
[15868] | 42 | use PrintInfo;
|
---|
[10218] | 43 |
|
---|
[15868] | 44 | BEGIN {
|
---|
| 45 | @BasePlugin::ISA = ( 'PrintInfo' );
|
---|
| 46 | }
|
---|
[5681] | 47 |
|
---|
[18320] | 48 | # the different methods that can be applied when renaming
|
---|
| 49 | # imported documents and their associated files
|
---|
| 50 | our $file_rename_method_list =
|
---|
| 51 | [ { 'name' => "url",
|
---|
| 52 | 'desc' => "{BasePlugin.rename_method.url}" },
|
---|
| 53 | { 'name' => "base64",
|
---|
[18398] | 54 | 'desc' => "{BasePlugin.rename_method.base64}" },
|
---|
[18404] | 55 | { 'name' => "none",
|
---|
| 56 | 'desc' => "{BasePlugin.rename_method.none}",
|
---|
[18398] | 57 | 'hiddengli' => "yes" } ];
|
---|
[18320] | 58 |
|
---|
[15868] | 59 | our $encoding_list =
|
---|
[10218] | 60 | [ { 'name' => "ascii",
|
---|
[16014] | 61 | 'desc' => "{BasePlugin.encoding.ascii}" },
|
---|
[4744] | 62 | { 'name' => "utf8",
|
---|
[16014] | 63 | 'desc' => "{BasePlugin.encoding.utf8}" },
|
---|
[4744] | 64 | { 'name' => "unicode",
|
---|
[16014] | 65 | 'desc' => "{BasePlugin.encoding.unicode}" } ];
|
---|
[3540] | 66 |
|
---|
[23457] | 67 |
|
---|
[10620] | 68 | my $e = $encodings::encodings;
|
---|
| 69 | foreach my $enc (sort {$e->{$a}->{'name'} cmp $e->{$b}->{'name'}} keys (%$e))
|
---|
| 70 | {
|
---|
| 71 | my $hashEncode =
|
---|
| 72 | {'name' => $enc,
|
---|
| 73 | 'desc' => $e->{$enc}->{'name'}};
|
---|
| 74 |
|
---|
[15868] | 75 | push(@{$encoding_list},$hashEncode);
|
---|
[10620] | 76 | }
|
---|
| 77 |
|
---|
[15868] | 78 | our $encoding_plus_auto_list =
|
---|
| 79 | [ { 'name' => "auto",
|
---|
[16557] | 80 | 'desc' => "{BasePlugin.filename_encoding.auto}" },
|
---|
[23457] | 81 | { 'name' => "auto-language-analysis",
|
---|
[16557] | 82 | 'desc' => "{BasePlugin.filename_encoding.auto_language_analysis}" }, # textcat
|
---|
| 83 | { 'name' => "auto-filesystem-encoding",
|
---|
| 84 | 'desc' => "{BasePlugin.filename_encoding.auto_filesystem_encoding}" }, # locale
|
---|
| 85 | { 'name' => "auto-fl",
|
---|
| 86 | 'desc' => "{BasePlugin.filename_encoding.auto_fl}" }, # locale followed by textcat
|
---|
| 87 | { 'name' => "auto-lf",
|
---|
| 88 | 'desc' => "{BasePlugin.filename_encoding.auto_lf}" } ]; # texcat followed by locale
|
---|
[10620] | 89 |
|
---|
[15868] | 90 | push(@{$encoding_plus_auto_list},@{$encoding_list});
|
---|
| 91 |
|
---|
[16698] | 92 | our $oidtype_list =
|
---|
[17026] | 93 | [ { 'name' => "auto",
|
---|
| 94 | 'desc' => "{BasePlugin.OIDtype.auto}" },
|
---|
| 95 | { 'name' => "hash",
|
---|
[16698] | 96 | 'desc' => "{import.OIDtype.hash}" },
|
---|
[24290] | 97 | { 'name' => "hash_on_ga_xml",
|
---|
| 98 | 'desc' => "{import.OIDtype.hash_on_ga_xml}" },
|
---|
[26536] | 99 | { 'name' => "hash_on_full_filename",
|
---|
| 100 | 'desc' => "{import.OIDtype.hash_on_full_filename}" },
|
---|
[16698] | 101 | { 'name' => "assigned",
|
---|
| 102 | 'desc' => "{import.OIDtype.assigned}" },
|
---|
| 103 | { 'name' => "incremental",
|
---|
| 104 | 'desc' => "{import.OIDtype.incremental}" },
|
---|
[26221] | 105 | { 'name' => "filename",
|
---|
| 106 | 'desc' => "{import.OIDtype.filename}" },
|
---|
[16698] | 107 | { 'name' => "dirname",
|
---|
[26536] | 108 | 'desc' => "{import.OIDtype.dirname}" },
|
---|
| 109 | { 'name' => "full_filename",
|
---|
| 110 | 'desc' => "{import.OIDtype.full_filename}" } ];
|
---|
[16698] | 111 |
|
---|
[4873] | 112 | my $arguments =
|
---|
[3540] | 113 | [ { 'name' => "process_exp",
|
---|
[15868] | 114 | 'desc' => "{BasePlugin.process_exp}",
|
---|
[6408] | 115 | 'type' => "regexp",
|
---|
[3540] | 116 | 'deft' => "",
|
---|
| 117 | 'reqd' => "no" },
|
---|
[16390] | 118 | { 'name' => "no_blocking",
|
---|
| 119 | 'desc' => "{BasePlugin.no_blocking}",
|
---|
| 120 | 'type' => "flag",
|
---|
| 121 | 'reqd' => "no"},
|
---|
[3540] | 122 | { 'name' => "block_exp",
|
---|
[15868] | 123 | 'desc' => "{BasePlugin.block_exp}",
|
---|
[6408] | 124 | 'type' => "regexp",
|
---|
[3540] | 125 | 'deft' => "",
|
---|
| 126 | 'reqd' => "no" },
|
---|
[22215] | 127 | { 'name' => "store_original_file",
|
---|
| 128 | 'desc' => "{BasePlugin.store_original_file}",
|
---|
| 129 | 'type' => "flag",
|
---|
| 130 | 'reqd' => "no" },
|
---|
[8892] | 131 | { 'name' => "associate_ext",
|
---|
[15868] | 132 | 'desc' => "{BasePlugin.associate_ext}",
|
---|
[8892] | 133 | 'type' => "string",
|
---|
| 134 | 'reqd' => "no" },
|
---|
[11122] | 135 | { 'name' => "associate_tail_re",
|
---|
[15868] | 136 | 'desc' => "{BasePlugin.associate_tail_re}",
|
---|
[11122] | 137 | 'type' => "string",
|
---|
| 138 | 'reqd' => "no" },
|
---|
[16698] | 139 | { 'name' => "OIDtype",
|
---|
| 140 | 'desc' => "{import.OIDtype}",
|
---|
| 141 | 'type' => "enum",
|
---|
| 142 | 'list' => $oidtype_list,
|
---|
[16847] | 143 | # leave default empty so we can tell if its been set or not - if not set will use option from import.pl
|
---|
[17026] | 144 | 'deft' => "auto",
|
---|
[18591] | 145 | 'reqd' => "no" },
|
---|
[16698] | 146 | { 'name' => "OIDmetadata",
|
---|
| 147 | 'desc' => "{import.OIDmetadata}",
|
---|
| 148 | 'type' => "metadata",
|
---|
| 149 | 'deft' => "dc.Identifier",
|
---|
[18591] | 150 | 'reqd' => "no" },
|
---|
[16698] | 151 | # { 'name' => "use_as_doc_identifier",
|
---|
| 152 | # 'desc' => "{BasePlugin.use_as_doc_identifier}",
|
---|
| 153 | # 'type' => "string",
|
---|
| 154 | # 'reqd' => "no" ,
|
---|
| 155 | # 'deft' => "" } ,
|
---|
[18320] | 156 | { 'name' => "no_cover_image",
|
---|
[15868] | 157 | 'desc' => "{BasePlugin.no_cover_image}",
|
---|
[3540] | 158 | 'type' => "flag",
|
---|
| 159 | 'reqd' => "no" },
|
---|
[15868] | 160 | { 'name' => "filename_encoding",
|
---|
| 161 | 'desc' => "{BasePlugin.filename_encoding}",
|
---|
| 162 | 'type' => "enum",
|
---|
| 163 | 'deft' => "auto",
|
---|
| 164 | 'list' => $encoding_plus_auto_list,
|
---|
[16390] | 165 | 'reqd' => "no" },
|
---|
| 166 | { 'name' => "smart_block",
|
---|
[16520] | 167 | 'desc' => "{common.deprecated}. {BasePlugin.smart_block}",
|
---|
[16390] | 168 | 'type' => "flag",
|
---|
| 169 | 'reqd' => "no",
|
---|
[18320] | 170 | 'hiddengli' => "yes" }, # deprecated, but leave in for old collections
|
---|
| 171 | { 'name' => "file_rename_method",
|
---|
| 172 | 'desc' => "{BasePlugin.file_rename_method}",
|
---|
| 173 | 'type' => "enum",
|
---|
| 174 | 'deft' => &get_default_file_rename_method(), # by default rename imported files and assoc files using this encoding
|
---|
| 175 | 'list' => $file_rename_method_list,
|
---|
| 176 | 'reqd' => "no"
|
---|
[23457] | 177 | }
|
---|
[15868] | 178 |
|
---|
| 179 | ];
|
---|
[3540] | 180 |
|
---|
[9398] | 181 |
|
---|
[15868] | 182 | my $options = { 'name' => "BasePlugin",
|
---|
| 183 | 'desc' => "{BasePlugin.desc}",
|
---|
[6408] | 184 | 'abstract' => "yes",
|
---|
| 185 | 'inherits' => "no",
|
---|
[4750] | 186 | 'args' => $arguments };
|
---|
[3540] | 187 |
|
---|
[4] | 188 | sub new {
|
---|
[10218] | 189 |
|
---|
[15868] | 190 | my ($class) = shift (@_);
|
---|
[16698] | 191 | my ($pluginlist,$inputargs,$hashArgOptLists,$auxiliary) = @_;
|
---|
[10218] | 192 | push(@$pluginlist, $class);
|
---|
[9398] | 193 |
|
---|
[15868] | 194 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
| 195 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
[10218] | 196 |
|
---|
[16698] | 197 | my $self = new PrintInfo($pluginlist, $inputargs, $hashArgOptLists,$auxiliary);
|
---|
[16390] | 198 |
|
---|
| 199 | if ($self->{'info_only'}) {
|
---|
| 200 | # don't worry about any options etc
|
---|
| 201 | return bless $self, $class;
|
---|
| 202 | }
|
---|
[10579] | 203 |
|
---|
[16390] | 204 | if ($self->{'smart_block'}) {
|
---|
| 205 | print STDERR "WARNING: -smart_block option has been deprecated and is no longer useful\n";
|
---|
| 206 | }
|
---|
| 207 | $self->{'smart_block'} = undef;
|
---|
| 208 |
|
---|
[15868] | 209 | my $plugin_name = (defined $pluginlist->[0]) ? $pluginlist->[0] : $class;
|
---|
| 210 | $self->{'plugin_type'} = $plugin_name;
|
---|
[10579] | 211 |
|
---|
[24404] | 212 | # remove ex. from OIDmetadata iff it's the only namespace prefix
|
---|
| 213 | $self->{'OIDmetadata'} =~ s/^ex\.([^.]+)$/$1/ if defined $self->{'OIDmetadata'};
|
---|
[2785] | 214 | $self->{'num_processed'} = 0;
|
---|
| 215 | $self->{'num_not_processed'} = 0;
|
---|
| 216 | $self->{'num_blocked'} = 0;
|
---|
| 217 | $self->{'num_archives'} = 0;
|
---|
[8678] | 218 | $self->{'cover_image'} = 1; # cover image is on by default
|
---|
[10218] | 219 | $self->{'cover_image'} = 0 if ($self->{'no_cover_image'});
|
---|
[19222] | 220 | $self->{'can_process_directories'} = 0;
|
---|
[10579] | 221 | #$self->{'option_list'} = $hashArgOptLists->{"OptList"};
|
---|
[3540] | 222 |
|
---|
[8892] | 223 | my $associate_ext = $self->{'associate_ext'};
|
---|
| 224 | if ((defined $associate_ext) && ($associate_ext ne "")) {
|
---|
[9351] | 225 |
|
---|
[11122] | 226 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 227 | if ((defined $associate_tail_re) && ($associate_tail_re ne "")) {
|
---|
| 228 | my $outhandle = $self->{'outhandle'};
|
---|
| 229 | print $outhandle "Warning: can only specify 'associate_ext' or 'associate_tail_re'\n";
|
---|
| 230 | print $outhandle " defaulting to 'associate_tail_re'\n";
|
---|
[8892] | 231 | }
|
---|
[11122] | 232 | else {
|
---|
| 233 | my @exts = split(/,/,$associate_ext);
|
---|
[8892] | 234 |
|
---|
[11122] | 235 | my @exts_bracketed = map { $_ = "(?:\\.$_)" } @exts;
|
---|
| 236 | my $associate_tail_re = join("|",@exts_bracketed);
|
---|
| 237 | $self->{'associate_tail_re'} = $associate_tail_re;
|
---|
| 238 | }
|
---|
| 239 |
|
---|
| 240 | delete $self->{'associate_ext'};
|
---|
[8892] | 241 | }
|
---|
| 242 |
|
---|
[15868] | 243 | return bless $self, $class;
|
---|
[11089] | 244 |
|
---|
[4] | 245 | }
|
---|
| 246 |
|
---|
[16821] | 247 | sub merge_inheritance
|
---|
| 248 | {
|
---|
| 249 | my $self = {};
|
---|
| 250 | my @child_selfs = @_;
|
---|
| 251 |
|
---|
| 252 | foreach my $child_self (@child_selfs) {
|
---|
| 253 | foreach my $key (keys %$child_self) {
|
---|
| 254 | if (defined $self->{$key}) {
|
---|
| 255 | if ($self->{$key} ne $child_self->{$key}) {
|
---|
| 256 | # print STDERR "Warning: Conflicting value in multiple inheritance for '$key'\n";
|
---|
| 257 | # print STDERR "Existing stored value = $self->{$key}\n";
|
---|
| 258 | # print STDERR "New (child) value = $child_self->{$key}\n";
|
---|
| 259 | # print STDERR "Keeping existing value\n";
|
---|
| 260 | # Existing value seems to be option specified in collect.cfg
|
---|
| 261 |
|
---|
| 262 | ### $self->{$key} = $child_self->{$key};
|
---|
| 263 |
|
---|
| 264 | }
|
---|
| 265 | else {
|
---|
| 266 | ## print STDERR "****Info: Value $self->{$key} for $key already defined through multiple inheritance as the same value\n";
|
---|
| 267 | }
|
---|
| 268 |
|
---|
| 269 | }
|
---|
| 270 | else {
|
---|
| 271 | $self->{$key} = $child_self->{$key};
|
---|
| 272 | }
|
---|
| 273 | }
|
---|
| 274 | }
|
---|
| 275 |
|
---|
| 276 | return $self;
|
---|
| 277 | }
|
---|
| 278 |
|
---|
[15868] | 279 | # initialize BasePlugin options
|
---|
| 280 | # if init() is overridden in a sub-class, remember to call BasePlugin::init()
|
---|
[1242] | 281 | sub init {
|
---|
| 282 | my $self = shift (@_);
|
---|
[2785] | 283 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
[1242] | 284 |
|
---|
| 285 | # verbosity is passed through from the processor
|
---|
| 286 | $self->{'verbosity'} = $verbosity;
|
---|
| 287 |
|
---|
[2785] | 288 | # as are the outhandle and failhandle
|
---|
[1424] | 289 | $self->{'outhandle'} = $outhandle if defined $outhandle;
|
---|
[2785] | 290 | $self->{'failhandle'} = $failhandle;
|
---|
[16390] | 291 | # $self->SUPER::init(@_);
|
---|
| 292 |
|
---|
[1242] | 293 | # set process_exp and block_exp to defaults unless they were
|
---|
| 294 | # explicitly set
|
---|
[1244] | 295 |
|
---|
| 296 | if ((!$self->is_recursive()) and
|
---|
[1242] | 297 | (!defined $self->{'process_exp'}) || ($self->{'process_exp'} eq "")) {
|
---|
[1244] | 298 |
|
---|
[1242] | 299 | $self->{'process_exp'} = $self->get_default_process_exp ();
|
---|
| 300 | if ($self->{'process_exp'} eq "") {
|
---|
[1244] | 301 | warn ref($self) . " Warning: Non-recursive plugin has no process_exp\n";
|
---|
[1242] | 302 | }
|
---|
| 303 | }
|
---|
| 304 |
|
---|
| 305 | if ((!defined $self->{'block_exp'}) || ($self->{'block_exp'} eq "")) {
|
---|
| 306 | $self->{'block_exp'} = $self->get_default_block_exp ();
|
---|
| 307 | }
|
---|
[11089] | 308 |
|
---|
[1242] | 309 | }
|
---|
| 310 |
|
---|
[839] | 311 | sub begin {
|
---|
| 312 | my $self = shift (@_);
|
---|
| 313 | my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
|
---|
[16821] | 314 |
|
---|
[17026] | 315 | if ($self->{'OIDtype'} eq "auto") {
|
---|
| 316 | # hasn't been set in the plugin, use the processor values
|
---|
| 317 | $self->{'OIDtype'} = $processor->{'OIDtype'};
|
---|
| 318 | $self->{'OIDmetadata'} = $processor->{'OIDmetadata'};
|
---|
| 319 | }
|
---|
| 320 | if ($self->{'OIDtype'} eq "hash") {
|
---|
| 321 | # should we hash on the file or on the doc xml??
|
---|
| 322 | $self->{'OIDtype'} = $self->get_oid_hash_type();
|
---|
| 323 | if ($self->{'OIDtype'} !~ /^(hash_on_file|hash_on_ga_xml)$/) {
|
---|
| 324 | $self->{'OIDtype'} = "hash_on_file";
|
---|
| 325 | }
|
---|
| 326 | }
|
---|
[839] | 327 | }
|
---|
| 328 |
|
---|
[21308] | 329 | # This is called once if removeold is set with import.pl. Most plugins will do
|
---|
| 330 | # nothing but if a plugin does any stuff outside of creating doc obj, then
|
---|
| 331 | # it may need to clear something.
|
---|
| 332 | sub remove_all {
|
---|
[21286] | 333 | my $self = shift (@_);
|
---|
| 334 | my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
|
---|
| 335 | }
|
---|
| 336 |
|
---|
[21308] | 337 | # This is called per document for docs that have been deleted from the
|
---|
| 338 | # collection. Most plugins will do nothing
|
---|
| 339 | # but if a plugin does any stuff outside of creating doc obj, then it may need
|
---|
| 340 | # to clear something.
|
---|
| 341 | sub remove_one {
|
---|
| 342 | my $self = shift (@_);
|
---|
| 343 |
|
---|
[21315] | 344 | my ($file, $oids, $archivedir) = @_;
|
---|
| 345 | return 0 if $self->can_process_this_file($file);
|
---|
| 346 | return undef;
|
---|
[21308] | 347 | }
|
---|
| 348 |
|
---|
[839] | 349 | sub end {
|
---|
[10155] | 350 | # potentially called at the end of each plugin pass
|
---|
| 351 | # import.pl only has one plugin pass, but buildcol.pl has multiple ones
|
---|
| 352 |
|
---|
[15868] | 353 | my ($self) = shift (@_);
|
---|
[839] | 354 | }
|
---|
| 355 |
|
---|
[10155] | 356 | sub deinit {
|
---|
| 357 | # called only once, after all plugin passes have been done
|
---|
| 358 |
|
---|
| 359 | my ($self) = @_;
|
---|
| 360 | }
|
---|
| 361 |
|
---|
[17026] | 362 | # default hashing type is to hash on the original file (or converted file)
|
---|
| 363 | # override this to return hash_on_ga_xml for filetypes where hashing on the
|
---|
| 364 | # file is no good eg video
|
---|
| 365 | sub get_oid_hash_type {
|
---|
| 366 |
|
---|
| 367 | my $self = shift (@_);
|
---|
| 368 |
|
---|
| 369 | return "hash_on_file";
|
---|
| 370 | }
|
---|
| 371 |
|
---|
[15868] | 372 |
|
---|
[1242] | 373 | # this function should be overridden to return 1
|
---|
| 374 | # in recursive plugins
|
---|
[4] | 375 | sub is_recursive {
|
---|
| 376 | my $self = shift (@_);
|
---|
| 377 |
|
---|
[1242] | 378 | return 0;
|
---|
[4] | 379 | }
|
---|
| 380 |
|
---|
[1242] | 381 | sub get_default_block_exp {
|
---|
| 382 | my $self = shift (@_);
|
---|
| 383 |
|
---|
| 384 | return "";
|
---|
| 385 | }
|
---|
| 386 |
|
---|
| 387 | sub get_default_process_exp {
|
---|
| 388 | my $self = shift (@_);
|
---|
| 389 |
|
---|
| 390 | return "";
|
---|
| 391 | }
|
---|
| 392 |
|
---|
[23419] | 393 |
|
---|
[18320] | 394 | # rename imported files and assoc files using URL encoding by default
|
---|
| 395 | # as this will work for most plugins and give more legible filenames
|
---|
| 396 | sub get_default_file_rename_method() {
|
---|
| 397 | my $self = shift (@_);
|
---|
| 398 | return "url";
|
---|
| 399 | }
|
---|
| 400 |
|
---|
[18398] | 401 | # returns this plugin's active (possibly user-selected) file_rename_method
|
---|
| 402 | sub get_file_rename_method() {
|
---|
| 403 | my $self = shift (@_);
|
---|
| 404 | my $rename_method = $self->{'file_rename_method'};
|
---|
| 405 | if($rename_method) {
|
---|
| 406 | return $rename_method;
|
---|
| 407 | } else {
|
---|
| 408 | return $self->get_default_file_rename_method();
|
---|
| 409 | }
|
---|
| 410 | }
|
---|
| 411 |
|
---|
[16390] | 412 | # default implementation is to do nothing
|
---|
| 413 | sub store_block_files {
|
---|
| 414 |
|
---|
[9067] | 415 | my $self =shift (@_);
|
---|
[16390] | 416 | my ($filename_full_path, $block_hash) = @_;
|
---|
| 417 |
|
---|
[9067] | 418 | }
|
---|
| 419 |
|
---|
[16390] | 420 | # put files to block into hash
|
---|
| 421 | sub use_block_expressions {
|
---|
| 422 |
|
---|
| 423 | my $self =shift (@_);
|
---|
| 424 | my ($filename_full_path, $block_hash) = @_;
|
---|
| 425 |
|
---|
[23363] | 426 | $filename_full_path = &util::upgrade_if_dos_filename($filename_full_path);
|
---|
| 427 |
|
---|
[16390] | 428 | if ($self->{'block_exp'} ne "" && $filename_full_path =~ /$self->{'block_exp'}/) {
|
---|
[23561] | 429 | &util::block_filename($block_hash,$filename_full_path);
|
---|
[16390] | 430 | }
|
---|
| 431 |
|
---|
| 432 | }
|
---|
| 433 |
|
---|
[9067] | 434 | #default implementation is to block a file with same name as this, but extension jpg or JPG, if cover_images is on.
|
---|
| 435 | sub block_cover_image
|
---|
| 436 | {
|
---|
[10833] | 437 | my $self =shift;
|
---|
[16390] | 438 | my ($filename, $block_hash) = @_;
|
---|
[10833] | 439 |
|
---|
[23363] | 440 | $filename = &util::upgrade_if_dos_filename($filename);
|
---|
| 441 |
|
---|
[9067] | 442 | if ($self->{'cover_image'}) {
|
---|
| 443 | my $coverfile = $filename;
|
---|
| 444 | $coverfile =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
[23759] | 445 |
|
---|
[25743] | 446 | #if there is no file extension, coverfile will be the same as filename
|
---|
| 447 | return if $coverfile eq $filename;
|
---|
| 448 |
|
---|
[23363] | 449 | if (!&util::fd_exists($coverfile)) {
|
---|
[9067] | 450 | $coverfile =~ s/jpg$/JPG/;
|
---|
| 451 | }
|
---|
[23363] | 452 | if (&util::fd_exists($coverfile)) {
|
---|
[23561] | 453 | &util::block_filename($block_hash,$coverfile);
|
---|
[11089] | 454 | }
|
---|
[9067] | 455 | }
|
---|
| 456 |
|
---|
| 457 | return;
|
---|
| 458 | }
|
---|
[11122] | 459 |
|
---|
| 460 |
|
---|
[16390] | 461 | # discover all the files that should be blocked by this plugin
|
---|
| 462 | # check the args ...
|
---|
| 463 | sub file_block_read {
|
---|
[11122] | 464 |
|
---|
[8510] | 465 | my $self = shift (@_);
|
---|
[16390] | 466 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $gli) = @_;
|
---|
[8892] | 467 | # Keep track of filenames with same root but different extensions
|
---|
[11122] | 468 | # Used to support -associate_ext and the more generalised
|
---|
| 469 | # -associate_tail_re
|
---|
[16390] | 470 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
[8892] | 471 |
|
---|
[18441] | 472 | if (!-d $filename_full_path) {
|
---|
| 473 | $block_hash->{'all_files'}->{$file} = 1;
|
---|
| 474 | }
|
---|
| 475 |
|
---|
[11122] | 476 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 477 | if ((defined $associate_tail_re) && ($associate_tail_re ne "")) {
|
---|
| 478 | my ($file_prefix,$file_ext)
|
---|
[16390] | 479 | = &util::get_prefix_and_tail_by_regex($filename_full_path,$associate_tail_re);
|
---|
[8892] | 480 | if ((defined $file_prefix) && (defined $file_ext)) {
|
---|
[16390] | 481 | my $shared_fileroot = $block_hash->{'shared_fileroot'};
|
---|
[8892] | 482 | if (!defined $shared_fileroot->{$file_prefix}) {
|
---|
[11122] | 483 | my $file_prefix_rec = { 'tie_to' => undef,
|
---|
| 484 | 'exts' => {} };
|
---|
[8892] | 485 | $shared_fileroot->{$file_prefix} = $file_prefix_rec;
|
---|
| 486 | }
|
---|
| 487 |
|
---|
| 488 | my $file_prefix_rec = $shared_fileroot->{$file_prefix};
|
---|
| 489 |
|
---|
[16390] | 490 | if ($self->can_process_this_file($filename_full_path)) {
|
---|
[8892] | 491 | # This is the document the others should be tied to
|
---|
| 492 | $file_prefix_rec->{'tie_to'} = $file_ext;
|
---|
| 493 | }
|
---|
| 494 | else {
|
---|
[11122] | 495 | if ($file_ext =~ m/$associate_tail_re$/) {
|
---|
[16390] | 496 | # this file should be associated to the main one
|
---|
[9351] | 497 | $file_prefix_rec->{'exts'}->{$file_ext} = 1;
|
---|
| 498 | }
|
---|
[8892] | 499 | }
|
---|
[11122] | 500 |
|
---|
[8892] | 501 | }
|
---|
| 502 | }
|
---|
[11122] | 503 |
|
---|
[16390] | 504 | # check block expressions
|
---|
| 505 | $self->use_block_expressions($filename_full_path, $block_hash) unless $self->{'no_blocking'};
|
---|
| 506 |
|
---|
[9067] | 507 | # now check whether we are actually processing this
|
---|
[16390] | 508 | if (!-f $filename_full_path || !$self->can_process_this_file($filename_full_path)) {
|
---|
[9067] | 509 | return undef; # can't recognise
|
---|
| 510 | }
|
---|
[23457] | 511 |
|
---|
[16852] | 512 | # if we have a block_exp, then this overrides the normal 'smart' blocking
|
---|
| 513 | $self->store_block_files($filename_full_path, $block_hash) unless ($self->{'no_blocking'} || $self->{'block_exp'} ne "");
|
---|
[8892] | 514 |
|
---|
[11089] | 515 | # block the cover image if there is one
|
---|
| 516 | if ($self->{'cover_image'}) {
|
---|
[16852] | 517 | $self->block_cover_image($filename_full_path, $block_hash);
|
---|
[11089] | 518 | }
|
---|
[23457] | 519 |
|
---|
[9067] | 520 | return 1;
|
---|
[8510] | 521 | }
|
---|
| 522 |
|
---|
[16390] | 523 | # plugins that rely on more than process_exp (eg XML plugins) can override this method
|
---|
| 524 | sub can_process_this_file {
|
---|
| 525 | my $self = shift(@_);
|
---|
| 526 | my ($filename) = @_;
|
---|
[8892] | 527 |
|
---|
[19222] | 528 | if (-d $filename && !$self->{'can_process_directories'}) {
|
---|
| 529 | return 0;
|
---|
| 530 | }
|
---|
[22427] | 531 |
|
---|
[16390] | 532 | if ($self->{'process_exp'} ne "" && $filename =~ /$self->{'process_exp'}/) {
|
---|
| 533 | return 1;
|
---|
[8892] | 534 | }
|
---|
| 535 | return 0;
|
---|
[10280] | 536 |
|
---|
| 537 | }
|
---|
| 538 |
|
---|
[24403] | 539 | # Even if a plugin can extract metadata in its metadata_read pass,
|
---|
| 540 | # make the default return 'undef' so processing of the file continues
|
---|
| 541 | # down the pipeline, so other plugins can also have the opportunity to
|
---|
| 542 | # locate metadata and set it up in the extrametakeys variables that
|
---|
| 543 | # are passed around.
|
---|
| 544 |
|
---|
| 545 | sub can_process_this_file_for_metadata {
|
---|
| 546 | my $self = shift(@_);
|
---|
| 547 |
|
---|
| 548 | return undef;
|
---|
| 549 | }
|
---|
| 550 |
|
---|
| 551 |
|
---|
[16390] | 552 | # just converts path as is to utf8.
|
---|
| 553 | sub filepath_to_utf8 {
|
---|
[10280] | 554 | my $self = shift (@_);
|
---|
[15868] | 555 | my ($file, $file_encoding) = @_;
|
---|
[16390] | 556 | my $filemeta = $file;
|
---|
[10280] | 557 |
|
---|
[16557] | 558 | my $filename_encoding = $self->{'filename_encoding'}; # filename encoding setting
|
---|
| 559 |
|
---|
| 560 | # Whenever filename-encoding is set to any of the auto settings, we
|
---|
| 561 | # check if the filename is already in UTF8. If it is, then we're done.
|
---|
| 562 | if($filename_encoding =~ m/auto/) {
|
---|
[23457] | 563 | if(&unicode::check_is_utf8($filemeta))
|
---|
| 564 | {
|
---|
| 565 | $filename_encoding = "utf8";
|
---|
| 566 | return $filemeta;
|
---|
| 567 | }
|
---|
[16557] | 568 | }
|
---|
| 569 |
|
---|
| 570 | # Auto setting, but filename is not utf8
|
---|
| 571 | if ($filename_encoding eq "auto")
|
---|
| 572 | {
|
---|
[23457] | 573 | # try textcat
|
---|
| 574 | $filename_encoding = $self->textcat_encoding($filemeta);
|
---|
[16557] | 575 |
|
---|
[23457] | 576 | # check the locale next
|
---|
| 577 | $filename_encoding = $self->locale_encoding() if $filename_encoding eq "undefined";
|
---|
[16557] | 578 |
|
---|
[23457] | 579 |
|
---|
| 580 | # now try the encoding of the document, if available
|
---|
| 581 | if ($filename_encoding eq "undefined" && defined $file_encoding) {
|
---|
| 582 | $filename_encoding = $file_encoding;
|
---|
| 583 | }
|
---|
[16557] | 584 |
|
---|
| 585 | }
|
---|
| 586 |
|
---|
| 587 | elsif ($filename_encoding eq "auto-language-analysis")
|
---|
| 588 | {
|
---|
| 589 | $filename_encoding = $self->textcat_encoding($filemeta);
|
---|
| 590 |
|
---|
| 591 | # now try the encoding of the document, if available
|
---|
| 592 | if ($filename_encoding eq "undefined" && defined $file_encoding) {
|
---|
| 593 | $filename_encoding = $file_encoding;
|
---|
| 594 | }
|
---|
| 595 | }
|
---|
| 596 |
|
---|
| 597 | elsif ($filename_encoding eq "auto-filesystem-encoding")
|
---|
[23352] | 598 | {
|
---|
[16557] | 599 | # try locale
|
---|
| 600 | $filename_encoding = $self->locale_encoding();
|
---|
| 601 | }
|
---|
| 602 |
|
---|
| 603 | elsif ($filename_encoding eq "auto-fl")
|
---|
| 604 | {
|
---|
| 605 | # filesystem-encoding (locale) then language-analysis (textcat)
|
---|
| 606 | $filename_encoding = $self->locale_encoding();
|
---|
| 607 |
|
---|
| 608 | # try textcat
|
---|
| 609 | $filename_encoding = $self->textcat_encoding($filemeta) if $filename_encoding eq "undefined";
|
---|
[23457] | 610 |
|
---|
[16557] | 611 | # else assume filename encoding is encoding of file content, if that's available
|
---|
| 612 | if ($filename_encoding eq "undefined" && defined $file_encoding) {
|
---|
| 613 | $filename_encoding = $file_encoding;
|
---|
[15868] | 614 | }
|
---|
[16557] | 615 | }
|
---|
[23457] | 616 |
|
---|
[16557] | 617 | elsif ($filename_encoding eq "auto-lf")
|
---|
| 618 | {
|
---|
| 619 | # language-analysis (textcat) then filesystem-encoding (locale)
|
---|
| 620 | $filename_encoding = $self->textcat_encoding($filemeta);
|
---|
| 621 |
|
---|
| 622 | # guess filename encoding from encoding of file content, if available
|
---|
| 623 | if ($filename_encoding eq "undefined" && defined $file_encoding) {
|
---|
| 624 | $filename_encoding = $file_encoding;
|
---|
| 625 | }
|
---|
| 626 |
|
---|
| 627 | # try locale
|
---|
| 628 | $filename_encoding = $self->locale_encoding() if $filename_encoding eq "undefined";
|
---|
[15868] | 629 | }
|
---|
[23457] | 630 |
|
---|
[16557] | 631 | # if still undefined, use utf8 as fallback
|
---|
| 632 | if ($filename_encoding eq "undefined") {
|
---|
| 633 | $filename_encoding = "utf8";
|
---|
| 634 | }
|
---|
| 635 |
|
---|
[18171] | 636 | #print STDERR "**** UTF8 encoding the filename $filemeta ";
|
---|
[16767] | 637 |
|
---|
[16557] | 638 | # if the filename encoding is set to utf8 but it isn't utf8 already--such as when
|
---|
| 639 | # 1. the utf8 fallback is used, or 2. if the system locale is used and happens to
|
---|
| 640 | # be always utf8 (in which case the filename's encoding is also set as utf8 even
|
---|
| 641 | # though the filename need not be if it originates from another system)--in such
|
---|
| 642 | # cases attempt to make the filename utf8 to match.
|
---|
| 643 | if($filename_encoding eq "utf8" && !&unicode::check_is_utf8($filemeta)) {
|
---|
| 644 | &unicode::ensure_utf8(\$filemeta);
|
---|
| 645 | }
|
---|
| 646 |
|
---|
| 647 | # convert non-unicode encodings to utf8
|
---|
| 648 | if ($filename_encoding !~ m/(?:ascii|utf8|unicode)/) {
|
---|
| 649 | $filemeta = &unicode::unicode2utf8(
|
---|
[23457] | 650 | &unicode::convert2unicode($filename_encoding, \$filemeta)
|
---|
| 651 | );
|
---|
[15868] | 652 | }
|
---|
[16390] | 653 |
|
---|
[18171] | 654 | #print STDERR " from encoding $filename_encoding -> $filemeta\n";
|
---|
[16390] | 655 | return $filemeta;
|
---|
| 656 | }
|
---|
| 657 |
|
---|
| 658 | # gets the filename with no path, converts to utf8, and then dm safes it.
|
---|
[18320] | 659 | # filename_encoding set by user
|
---|
[16390] | 660 | sub filename_to_utf8_metadata
|
---|
| 661 | {
|
---|
| 662 | my $self = shift (@_);
|
---|
| 663 | my ($file, $file_encoding) = @_;
|
---|
| 664 |
|
---|
| 665 | my $outhandle = $self->{'outhandle'};
|
---|
| 666 |
|
---|
[23457] | 667 | print $outhandle "****!!!!**** BasePlugin::filename_to_utf8_metadata now deprecated\n";
|
---|
| 668 | my ($cpackage,$cfilename,$cline,$csubr,$chas_args,$cwantarray) = caller(0);
|
---|
| 669 | print $outhandle "Calling method: $cfilename:$cline $cpackage->$csubr\n";
|
---|
[23347] | 670 |
|
---|
| 671 |
|
---|
[16390] | 672 | my ($filemeta) = $file =~ /([^\\\/]+)$/; # getting the tail of the filepath (skips all string parts containing slashes upto the end)
|
---|
| 673 | $filemeta = $self->filepath_to_utf8($filemeta, $file_encoding);
|
---|
[15868] | 674 |
|
---|
[22652] | 675 | return $filemeta;
|
---|
[10280] | 676 | }
|
---|
| 677 |
|
---|
[16557] | 678 | sub locale_encoding {
|
---|
| 679 | my $self = shift(@_);
|
---|
| 680 |
|
---|
| 681 | if (!defined $self->{'filesystem_encoding'}) {
|
---|
| 682 | $self->{'filesystem_encoding'} = $self->get_filesystem_encoding();
|
---|
| 683 | }
|
---|
[10280] | 684 |
|
---|
[18171] | 685 | #print STDERR "*** filename encoding determined based on locale: " . $self->{'filesystem_encoding'} . "\n";
|
---|
[16557] | 686 | return $self->{'filesystem_encoding'}; # can be the string "undefined"
|
---|
| 687 | }
|
---|
| 688 |
|
---|
| 689 | sub textcat_encoding {
|
---|
| 690 | my $self = shift(@_);
|
---|
| 691 | my ($filemeta) = @_;
|
---|
| 692 |
|
---|
[22705] | 693 | # analyse filenames without extensions and digits (and trimmed of
|
---|
| 694 | # surrounding whitespace), so that irrelevant chars don't confuse
|
---|
| 695 | # textcat
|
---|
[16557] | 696 | my $strictfilemeta = $filemeta;
|
---|
| 697 | $strictfilemeta =~ s/\.[^\.]+$//g;
|
---|
| 698 | $strictfilemeta =~ s/\d//g;
|
---|
| 699 | $strictfilemeta =~ s/^\s*//g;
|
---|
| 700 | $strictfilemeta =~ s/\s*$//g;
|
---|
| 701 |
|
---|
| 702 | my $filename_encoding = $self->encoding_from_language_analysis($strictfilemeta);
|
---|
| 703 | if(!defined $filename_encoding) {
|
---|
| 704 | $filename_encoding = "undefined";
|
---|
| 705 | }
|
---|
| 706 |
|
---|
| 707 | return $filename_encoding; # can be the string "undefined"
|
---|
| 708 | }
|
---|
| 709 |
|
---|
| 710 | # performs textcat
|
---|
| 711 | sub encoding_from_language_analysis {
|
---|
| 712 | my $self = shift(@_);
|
---|
| 713 | my ($text) = @_;
|
---|
| 714 |
|
---|
| 715 | my $outhandle = $self->{'outhandle'};
|
---|
| 716 | my $best_encoding = undef;
|
---|
[23457] | 717 |
|
---|
[17212] | 718 | # get the language/encoding of the textstring using textcat
|
---|
[22632] | 719 | require textcat; # Only load the textcat module if it is required
|
---|
[16557] | 720 | $self->{'textcat'} = new textcat() unless defined($self->{'textcat'});
|
---|
[16767] | 721 | my $results = $self->{'textcat'}->classify_cached_filename(\$text);
|
---|
[16557] | 722 |
|
---|
| 723 |
|
---|
| 724 | if (scalar @$results < 0) {
|
---|
| 725 | return undef;
|
---|
| 726 | }
|
---|
| 727 |
|
---|
| 728 | # We have some results, we choose the first
|
---|
| 729 | my ($language, $encoding) = $results->[0] =~ /^([^-]*)(?:-(.*))?$/;
|
---|
| 730 |
|
---|
| 731 | $best_encoding = $encoding;
|
---|
| 732 | if (!defined $best_encoding) {
|
---|
| 733 | return undef;
|
---|
| 734 | }
|
---|
[23457] | 735 |
|
---|
[16557] | 736 | if (defined $best_encoding && $best_encoding =~ m/^iso_8859/ && &unicode::check_is_utf8($text)) {
|
---|
| 737 | # the text is valid utf8, so assume that's the real encoding (since textcat is based on probabilities)
|
---|
| 738 | $best_encoding = 'utf8';
|
---|
| 739 | }
|
---|
| 740 |
|
---|
| 741 |
|
---|
| 742 | # check for equivalents where textcat doesn't have some encodings...
|
---|
| 743 | # eg MS versions of standard encodings
|
---|
| 744 | if (defined $best_encoding && $best_encoding =~ /^iso_8859_(\d+)/) {
|
---|
| 745 | my $iso = $1; # which variant of the iso standard?
|
---|
| 746 | # iso-8859 sets don't use chars 0x80-0x9f, windows codepages do
|
---|
| 747 | if ($text =~ /[\x80-\x9f]/) {
|
---|
| 748 | # Western Europe
|
---|
| 749 | if ($iso == 1 or $iso == 15) { $best_encoding = 'windows_1252' }
|
---|
| 750 | elsif ($iso == 2) {$best_encoding = 'windows_1250'} # Central Europe
|
---|
| 751 | elsif ($iso == 5) {$best_encoding = 'windows_1251'} # Cyrillic
|
---|
| 752 | elsif ($iso == 6) {$best_encoding = 'windows_1256'} # Arabic
|
---|
| 753 | elsif ($iso == 7) {$best_encoding = 'windows_1253'} # Greek
|
---|
| 754 | elsif ($iso == 8) {$best_encoding = 'windows_1255'} # Hebrew
|
---|
| 755 | elsif ($iso == 9) {$best_encoding = 'windows_1254'} # Turkish
|
---|
| 756 | }
|
---|
| 757 | }
|
---|
| 758 |
|
---|
| 759 | if (defined $best_encoding && $best_encoding !~ /^(ascii|utf8|unicode)$/ &&
|
---|
| 760 | !defined $encodings::encodings->{$best_encoding})
|
---|
| 761 | {
|
---|
| 762 | if ($self->{'verbosity'}) {
|
---|
| 763 | gsprintf($outhandle, "BasePlugin: {ReadTextFile.unsupported_encoding}\n", $text, $best_encoding, "undef");
|
---|
| 764 | }
|
---|
| 765 | $best_encoding = undef;
|
---|
| 766 | }
|
---|
| 767 |
|
---|
| 768 | return $best_encoding;
|
---|
| 769 | }
|
---|
| 770 |
|
---|
| 771 | # uses locale
|
---|
[23352] | 772 | sub get_filesystem_encoding
|
---|
| 773 | {
|
---|
[10280] | 774 |
|
---|
[15868] | 775 | my $self = shift(@_);
|
---|
| 776 |
|
---|
[14961] | 777 | my $outhandle = $self->{'outhandle'};
|
---|
| 778 | my $filesystem_encoding = undef;
|
---|
| 779 |
|
---|
| 780 | eval {
|
---|
[23352] | 781 | # Works for Windows as well, returning the DOS code page in use
|
---|
[14961] | 782 | use POSIX qw(locale_h);
|
---|
[15868] | 783 |
|
---|
| 784 | # With only one parameter, setlocale retrieves the
|
---|
| 785 | # current value
|
---|
[14961] | 786 | my $current_locale = setlocale(LC_CTYPE);
|
---|
[15868] | 787 |
|
---|
[23364] | 788 | my $char_encoding = undef;
|
---|
| 789 | if ($current_locale =~ m/\./) {
|
---|
| 790 | ($char_encoding) = ($current_locale =~ m/^.*\.(.*?)$/);
|
---|
| 791 | $char_encoding = lc($char_encoding);
|
---|
| 792 | }
|
---|
| 793 | else {
|
---|
| 794 | if ($current_locale =~ m/^(posix|c)$/i) {
|
---|
| 795 | $char_encoding = "ascii";
|
---|
| 796 | }
|
---|
| 797 | }
|
---|
| 798 |
|
---|
| 799 | if (defined $char_encoding) {
|
---|
[24564] | 800 | if ($char_encoding =~ m/^(iso)(8859)-?(\d{1,2})$/) {
|
---|
[15446] | 801 | $char_encoding = "$1\_$2\_$3";
|
---|
| 802 | }
|
---|
| 803 |
|
---|
[14961] | 804 | $char_encoding =~ s/-/_/g;
|
---|
| 805 | $char_encoding =~ s/^utf_8$/utf8/;
|
---|
[15868] | 806 |
|
---|
[14961] | 807 | if ($char_encoding =~ m/^\d+$/) {
|
---|
[15607] | 808 | if (defined $encodings::encodings->{"windows_$char_encoding"}) {
|
---|
[14961] | 809 | $char_encoding = "windows_$char_encoding";
|
---|
| 810 | }
|
---|
[15607] | 811 | elsif (defined $encodings::encodings->{"dos_$char_encoding"}) {
|
---|
[14961] | 812 | $char_encoding = "dos_$char_encoding";
|
---|
| 813 | }
|
---|
| 814 | }
|
---|
[15868] | 815 |
|
---|
[14961] | 816 | if (($char_encoding =~ m/(?:ascii|utf8|unicode)/)
|
---|
[15446] | 817 | || (defined $encodings::encodings->{$char_encoding})) {
|
---|
[14961] | 818 | $filesystem_encoding = $char_encoding;
|
---|
| 819 | }
|
---|
| 820 | else {
|
---|
| 821 | print $outhandle "Warning: Unsupported character encoding '$char_encoding' from locale '$current_locale'\n";
|
---|
| 822 | }
|
---|
| 823 | }
|
---|
[15868] | 824 |
|
---|
[14961] | 825 |
|
---|
| 826 | };
|
---|
| 827 | if ($@) {
|
---|
| 828 | print $outhandle "$@\n";
|
---|
[18398] | 829 | print $outhandle "Warning: Unable to establish locale. Will assume filesystem is UTF-8\n";
|
---|
[14961] | 830 |
|
---|
| 831 | }
|
---|
[23352] | 832 |
|
---|
[15868] | 833 | return $filesystem_encoding;
|
---|
| 834 | }
|
---|
[14961] | 835 |
|
---|
[23335] | 836 |
|
---|
| 837 |
|
---|
| 838 | sub deduce_filename_encoding
|
---|
| 839 | {
|
---|
| 840 | my $self = shift (@_);
|
---|
[23347] | 841 | my ($file,$metadata,$plugin_filename_encoding) = @_;
|
---|
[23335] | 842 |
|
---|
[23465] | 843 | my $gs_filename_encoding = $metadata->{"gs.filenameEncoding"};
|
---|
[23335] | 844 | my $deduced_filename_encoding = undef;
|
---|
| 845 |
|
---|
| 846 | # Start by looking for manually assigned metadata
|
---|
| 847 | if (defined $gs_filename_encoding) {
|
---|
[23457] | 848 | if (ref ($gs_filename_encoding) eq "ARRAY") {
|
---|
| 849 | my $outhandle = $self->{'outhandle'};
|
---|
| 850 |
|
---|
| 851 | $deduced_filename_encoding = $gs_filename_encoding->[0];
|
---|
| 852 |
|
---|
| 853 | my $num_vals = scalar(@$gs_filename_encoding);
|
---|
| 854 | if ($num_vals>1) {
|
---|
[23465] | 855 | print $outhandle "Warning: gs.filenameEncoding multiply defined for $file\n";
|
---|
[23457] | 856 | print $outhandle " Selecting first value: $deduced_filename_encoding\n";
|
---|
| 857 | }
|
---|
| 858 | }
|
---|
| 859 | else {
|
---|
| 860 | $deduced_filename_encoding = $gs_filename_encoding;
|
---|
| 861 | }
|
---|
[23335] | 862 | }
|
---|
[23457] | 863 |
|
---|
[23335] | 864 | if (!defined $deduced_filename_encoding || ($deduced_filename_encoding =~ m/^\s*$/)) {
|
---|
[23457] | 865 | # Look to see if plugin specifies this value
|
---|
[23347] | 866 |
|
---|
[23457] | 867 | if (defined $plugin_filename_encoding) {
|
---|
| 868 | # First look to see if we're using any of the "older" (i.e. deprecated auto-... plugin options)
|
---|
| 869 | if ($plugin_filename_encoding =~ m/^auto-.*$/) {
|
---|
| 870 | my $outhandle = $self->{'outhandle'};
|
---|
| 871 | print $outhandle "Warning: $plugin_filename_encoding is no longer supported\n";
|
---|
| 872 | print $outhandle " default to 'auto'\n";
|
---|
| 873 | $self->{'filename_encoding'} = $plugin_filename_encoding = "auto";
|
---|
| 874 | }
|
---|
| 875 |
|
---|
| 876 | if ($plugin_filename_encoding ne "auto") {
|
---|
| 877 | # We've been given a specific filenamne encoding
|
---|
| 878 | # => so use it!
|
---|
| 879 | $deduced_filename_encoding = $plugin_filename_encoding;
|
---|
| 880 | }
|
---|
| 881 | }
|
---|
[23335] | 882 | }
|
---|
| 883 |
|
---|
| 884 | if (!defined $deduced_filename_encoding || ($deduced_filename_encoding =~ m/^\s*$/)) {
|
---|
[23347] | 885 |
|
---|
[23352] | 886 | # Look to file system to provide a character encoding
|
---|
[23347] | 887 |
|
---|
[23352] | 888 | # If Windows NTFS, then -- assuming we work with long file names got through
|
---|
| 889 | # Win32::GetLongFilePath() -- then the underlying file system is UTF16
|
---|
| 890 |
|
---|
| 891 | if ($ENV{'GSDLOS'} =~ m/^windows$/i) {
|
---|
| 892 | # Can do better than working with the DOS character encoding returned by locale
|
---|
| 893 | $deduced_filename_encoding = "unicode";
|
---|
| 894 | }
|
---|
| 895 | else {
|
---|
| 896 | # Unix of some form or other
|
---|
| 897 |
|
---|
| 898 | # See if we can determine the file system encoding through locale
|
---|
| 899 | $deduced_filename_encoding = $self->locale_encoding();
|
---|
[23457] | 900 |
|
---|
[23352] | 901 | # if locale shows us filesystem is utf8, check to see filename is consistent
|
---|
| 902 | # => if not, then we have an "alien" filename on our hands
|
---|
[23457] | 903 |
|
---|
[23458] | 904 | if (defined $deduced_filename_encoding && $deduced_filename_encoding =~ m/^utf-?8$/i) {
|
---|
[23352] | 905 | if (!&unicode::check_is_utf8($file)) {
|
---|
| 906 | # "alien" filename, so revert
|
---|
| 907 | $deduced_filename_encoding = undef;
|
---|
[23347] | 908 | }
|
---|
[23352] | 909 | }
|
---|
| 910 | }
|
---|
[23335] | 911 | }
|
---|
| 912 |
|
---|
[23347] | 913 | # if (!defined $deduced_filename_encoding || ($deduced_filename_encoding =~ m/^\s*$/)) {
|
---|
| 914 | # # Last chance, apply textcat to deduce filename encoding
|
---|
| 915 | # $deduced_filename_encoding = $self->textcat_encoding($file);
|
---|
| 916 | # }
|
---|
| 917 |
|
---|
[23457] | 918 | if ($self->{'verbosity'}>3) {
|
---|
| 919 | my $outhandle = $self->{'outhandle'};
|
---|
[23347] | 920 |
|
---|
[23457] | 921 | if (defined $deduced_filename_encoding) {
|
---|
| 922 | print $outhandle " Deduced filename encoding as: $deduced_filename_encoding\n";
|
---|
[23347] | 923 | }
|
---|
[23457] | 924 | else {
|
---|
| 925 | print $outhandle " No filename encoding deduced\n";
|
---|
| 926 | }
|
---|
| 927 | }
|
---|
| 928 |
|
---|
[23335] | 929 | return $deduced_filename_encoding;
|
---|
| 930 | }
|
---|
| 931 |
|
---|
| 932 |
|
---|
| 933 |
|
---|
| 934 |
|
---|
| 935 | # Notionally written to be called once for each document, it is however safe to
|
---|
| 936 | # call multiple times (as in the case of ImagePlugin) which calls this later on
|
---|
| 937 | # after the original image has potentially been converted to a *new* source image
|
---|
| 938 | # format (e.g. TIFF to PNG)
|
---|
| 939 |
|
---|
[15868] | 940 | sub set_Source_metadata {
|
---|
| 941 | my $self = shift (@_);
|
---|
[23461] | 942 | my ($doc_obj, $raw_filename, $filename_encoding, $section) = @_;
|
---|
| 943 |
|
---|
[23335] | 944 | # 1. Sets the filename (Source) for display encoded as Unicode if possible,
|
---|
| 945 | # and (as a fallback) using %xx if not for non-ascii chars
|
---|
| 946 | # 2. Sets the url ref (SourceFile) to the URL encoded version
|
---|
| 947 | # of filename for generated files
|
---|
[23352] | 948 |
|
---|
| 949 | my ($unused_full_rf, $raw_file) = &util::get_full_filenames("", $raw_filename);
|
---|
[23335] | 950 |
|
---|
[23461] | 951 | my $this_section = (defined $section)? $section : $doc_obj->get_top_section();
|
---|
[23352] | 952 |
|
---|
| 953 | my $octet_file = $raw_file;
|
---|
| 954 |
|
---|
| 955 | # UTF-8 version of filename
|
---|
[23387] | 956 | # if ((defined $ENV{"DEBUG_UNICODE"}) && ($ENV{"DEBUG_UNICODE"})) {
|
---|
| 957 | # print STDERR "**** Setting Source Metadata given: $octet_file\n";
|
---|
| 958 | # }
|
---|
[14961] | 959 |
|
---|
[23352] | 960 | # Deal with (on Windows) raw filenames that are in their
|
---|
| 961 | # abbreviated DOS form
|
---|
| 962 |
|
---|
| 963 | if ($ENV{'GSDLOS'} =~ m/^windows$/i) {
|
---|
| 964 | if ((defined $filename_encoding) && ($filename_encoding eq "unicode")) {
|
---|
| 965 | if (-e $raw_filename) {
|
---|
| 966 | my $unicode_filename = Win32::GetLongPathName($raw_filename);
|
---|
| 967 |
|
---|
| 968 | my $unused_full_uf;
|
---|
| 969 | ($unused_full_uf, $octet_file) = &util::get_full_filenames("", $unicode_filename);
|
---|
| 970 | }
|
---|
[23347] | 971 | }
|
---|
[23352] | 972 | }
|
---|
[23335] | 973 |
|
---|
| 974 | my $url_encoded_filename;
|
---|
[23364] | 975 | if ((defined $filename_encoding) && ($filename_encoding ne "ascii")) {
|
---|
[23352] | 976 | # => Generate a pretty print version of filename that is mapped to Unicode
|
---|
| 977 |
|
---|
| 978 | # Use filename_encoding to map raw filename to a Perl unicode-aware string
|
---|
| 979 | $url_encoded_filename = decode($filename_encoding,$octet_file);
|
---|
[23335] | 980 | }
|
---|
| 981 | else {
|
---|
[23352] | 982 | # otherwise generate %xx encoded version of filename for char > 127
|
---|
| 983 | $url_encoded_filename = &unicode::raw_filename_to_url_encoded($octet_file);
|
---|
[23335] | 984 | }
|
---|
[18320] | 985 |
|
---|
[23387] | 986 | # if ((defined $ENV{"DEBUG_UNICODE"}) && ($ENV{"DEBUG_UNICODE"})) {
|
---|
| 987 | # print STDERR "****** saving Source as: $url_encoded_filename\n";
|
---|
| 988 | # }
|
---|
[23759] | 989 |
|
---|
| 990 | # In the case of converted files and (generalized) exploded documents, there
|
---|
| 991 | # will already be a source filename => store as OrigSource before overriding
|
---|
| 992 | my $orig_source = $doc_obj->get_metadata_element ($this_section, "Source");
|
---|
| 993 | if ((defined $orig_source) && ($orig_source !~ m/^\s*$/)) {
|
---|
| 994 | $doc_obj->set_utf8_metadata_element($this_section, "OrigSource", $orig_source);
|
---|
| 995 | }
|
---|
| 996 |
|
---|
[16919] | 997 | # Source is the UTF8 display name - not necessarily the name of the file on the system
|
---|
[23829] | 998 | if ($ENV{'GSDLOS'} =~ m/^darwin$/i) {
|
---|
| 999 | # on Darwin want all display strings to be in composed form, then can search on that
|
---|
| 1000 | $url_encoded_filename = normalize('C', $url_encoded_filename); # Normalisation Form 'C' (composition)
|
---|
| 1001 | }
|
---|
[23461] | 1002 | $doc_obj->set_utf8_metadata_element($this_section, "Source", $url_encoded_filename);
|
---|
[23759] | 1003 |
|
---|
[23352] | 1004 |
|
---|
[23335] | 1005 | my $renamed_raw_file = &util::rename_file($raw_file, $self->{'file_rename_method'});
|
---|
[18320] | 1006 | # If using URL encoding, then SourceFile is the url-reference to url-encoded
|
---|
[23335] | 1007 | # renamed_raw_url: it's a url that refers to the actual file on the system
|
---|
| 1008 | my $renamed_raw_url = &unicode::filename_to_url($renamed_raw_file);
|
---|
[18320] | 1009 |
|
---|
[23461] | 1010 | $doc_obj->set_utf8_metadata_element($this_section, "SourceFile",
|
---|
[23335] | 1011 | $renamed_raw_url);
|
---|
| 1012 |
|
---|
[23387] | 1013 | # if ((defined $ENV{"DEBUG_UNICODE"}) && ($ENV{"DEBUG_UNICODE"})) {
|
---|
| 1014 | # print STDERR "****** saving SourceFile as: $renamed_raw_url\n";
|
---|
| 1015 | # }
|
---|
[14961] | 1016 | }
|
---|
[23457] | 1017 |
|
---|
[17026] | 1018 | # this should be called by all plugins to set the oid of the doc obj, rather
|
---|
| 1019 | # than calling doc_obj->set_OID directly
|
---|
[15868] | 1020 | sub add_OID {
|
---|
[15018] | 1021 | my $self = shift (@_);
|
---|
| 1022 | my ($doc_obj) = @_;
|
---|
| 1023 |
|
---|
[26221] | 1024 | # don't add one if there is one already set
|
---|
| 1025 | return unless $doc_obj->get_OID() =~ /^NULL$/;
|
---|
[17026] | 1026 | $doc_obj->set_OIDtype($self->{'OIDtype'}, $self->{'OIDmetadata'});
|
---|
[15018] | 1027 |
|
---|
[17026] | 1028 | # see if there is a plugin specific set_OID function
|
---|
| 1029 | if (defined ($self->can('set_OID'))) {
|
---|
| 1030 | $self->set_OID(@_); # pass through doc_obj and any extra arguments
|
---|
[15018] | 1031 | }
|
---|
[17026] | 1032 | else {
|
---|
[15018] | 1033 | # use the default set_OID() in doc.pm
|
---|
| 1034 | $doc_obj->set_OID();
|
---|
| 1035 | }
|
---|
[17026] | 1036 |
|
---|
[15018] | 1037 | }
|
---|
[23457] | 1038 |
|
---|
[15868] | 1039 | # The BasePlugin read_into_doc_obj() function. This function does all the
|
---|
| 1040 | # 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] | 1041 | # a file and sets up a slew of metadata all saved in doc_obj, which
|
---|
| 1042 | # it then returns as part of a tuple (process_status,doc_obj)
|
---|
| 1043 | #
|
---|
| 1044 | # Much of this functionality used to reside in read, but it was broken
|
---|
| 1045 | # down into a supporting routine to make the code more flexible.
|
---|
| 1046 | #
|
---|
| 1047 | # recursive plugins (e.g. RecPlug) and specialized plugins like those
|
---|
| 1048 | # capable of processing many documents within a single file (e.g.
|
---|
| 1049 | # GMLPlug) will normally want to implement their own version of
|
---|
| 1050 | # read_into_doc_obj()
|
---|
| 1051 | #
|
---|
| 1052 | # Note that $base_dir might be "" and that $file might
|
---|
| 1053 | # include directories
|
---|
[15868] | 1054 |
|
---|
| 1055 | # currently blocking has been done before it gets here - does this affect secondary plugin stuff??
|
---|
[10280] | 1056 | sub read_into_doc_obj {
|
---|
| 1057 | my $self = shift (@_);
|
---|
[16390] | 1058 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
[10280] | 1059 |
|
---|
[15868] | 1060 | my $outhandle = $self->{'outhandle'};
|
---|
[10280] | 1061 |
|
---|
[15868] | 1062 | # should we move this to read? What about secondary plugins?
|
---|
[23759] | 1063 | my $pp_file = &util::prettyprint_file($base_dir,$file,$gli);
|
---|
[15868] | 1064 | print STDERR "<Processing n='$file' p='$self->{'plugin_type'}'>\n" if ($gli);
|
---|
[23363] | 1065 | print $outhandle "$self->{'plugin_type'} processing $pp_file\n"
|
---|
[23457] | 1066 | if $self->{'verbosity'} > 1;
|
---|
[10280] | 1067 |
|
---|
[16390] | 1068 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
[23457] | 1069 |
|
---|
[1242] | 1070 | # create a new document
|
---|
[18320] | 1071 | my $doc_obj = new doc ($filename_full_path, "indexed_doc", $self->{'file_rename_method'});
|
---|
[14961] | 1072 | my $top_section = $doc_obj->get_top_section();
|
---|
| 1073 |
|
---|
| 1074 | $doc_obj->add_utf8_metadata($top_section, "Plugin", "$self->{'plugin_type'}");
|
---|
[15868] | 1075 | $doc_obj->add_utf8_metadata($top_section, "FileSize", (-s $filename_full_path));
|
---|
[23457] | 1076 |
|
---|
[18469] | 1077 |
|
---|
[23352] | 1078 | my $plugin_filename_encoding = $self->{'filename_encoding'};
|
---|
[23347] | 1079 | my $filename_encoding = $self->deduce_filename_encoding($file,$metadata,$plugin_filename_encoding);
|
---|
[23472] | 1080 | $self->set_Source_metadata($doc_obj,$filename_full_path,$filename_encoding,$top_section);
|
---|
[8166] | 1081 |
|
---|
[15868] | 1082 | # plugin specific stuff - what args do we need here??
|
---|
| 1083 | unless (defined ($self->process($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli))) {
|
---|
| 1084 | print STDERR "<ProcessingError n='$file'>\n" if ($gli);
|
---|
| 1085 | return -1;
|
---|
[2816] | 1086 | }
|
---|
[1242] | 1087 |
|
---|
[15868] | 1088 | # include any metadata passed in from previous plugins
|
---|
| 1089 | # note that this metadata is associated with the top level section
|
---|
| 1090 | my $section = $doc_obj->get_top_section();
|
---|
| 1091 | # can we merge these two methods??
|
---|
| 1092 | $self->add_associated_files($doc_obj, $filename_full_path);
|
---|
| 1093 | $self->extra_metadata ($doc_obj, $section, $metadata);
|
---|
| 1094 | $self->auto_extract_metadata($doc_obj);
|
---|
[1242] | 1095 |
|
---|
[15868] | 1096 | # if we haven't found any Title so far, assign one
|
---|
| 1097 | # this was shifted to here from inside read()
|
---|
| 1098 | $self->title_fallback($doc_obj,$section,$filename_no_path);
|
---|
| 1099 |
|
---|
| 1100 | $self->add_OID($doc_obj);
|
---|
| 1101 |
|
---|
[21219] | 1102 | $self->post_process_doc_obj($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli);
|
---|
[15868] | 1103 | return (1,$doc_obj);
|
---|
| 1104 | }
|
---|
[2785] | 1105 |
|
---|
[21219] | 1106 | sub post_process_doc_obj {
|
---|
| 1107 | my $self = shift (@_);
|
---|
| 1108 | my ($pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
---|
| 1109 |
|
---|
| 1110 | return 1;
|
---|
| 1111 | }
|
---|
| 1112 |
|
---|
[15868] | 1113 | sub add_dummy_text {
|
---|
| 1114 | my $self = shift(@_);
|
---|
| 1115 | my ($doc_obj, $section) = @_;
|
---|
[2785] | 1116 |
|
---|
[15868] | 1117 | # add NoText metadata so we can hide this dummy text in format statements
|
---|
| 1118 | $doc_obj->add_metadata($section, "NoText", "1");
|
---|
| 1119 | $doc_obj->add_text($section, &gsprintf::lookup_string("{BasePlugin.dummy_text}",1));
|
---|
| 1120 |
|
---|
| 1121 | }
|
---|
[8510] | 1122 |
|
---|
[15868] | 1123 | # does nothing. Can be overridden by subclass
|
---|
| 1124 | sub auto_extract_metadata {
|
---|
| 1125 | my $self = shift(@_);
|
---|
| 1126 | my ($doc_obj) = @_;
|
---|
| 1127 | }
|
---|
[11122] | 1128 |
|
---|
[15868] | 1129 | # adds cover image, associate_file options stuff. Should be called by sub class
|
---|
| 1130 | # read_into_doc_obj
|
---|
| 1131 | sub add_associated_files {
|
---|
| 1132 | my $self = shift(@_);
|
---|
| 1133 | # whatis filename??
|
---|
| 1134 | my ($doc_obj, $filename) = @_;
|
---|
| 1135 |
|
---|
| 1136 | # add in the cover image
|
---|
| 1137 | if ($self->{'cover_image'}) {
|
---|
| 1138 | $self->associate_cover_image($doc_obj, $filename);
|
---|
[8716] | 1139 | }
|
---|
[22215] | 1140 | # store the original (used for eg TextPlugin to store the original for OAI)
|
---|
| 1141 | if ($self->{'store_original_file'}) {
|
---|
| 1142 | $self->associate_source_file($doc_obj, $filename);
|
---|
| 1143 | }
|
---|
[23457] | 1144 |
|
---|
[15018] | 1145 |
|
---|
[10280] | 1146 | }
|
---|
[1242] | 1147 |
|
---|
[16390] | 1148 | # implement this if you are extracting metadata for other documents
|
---|
| 1149 | sub metadata_read {
|
---|
| 1150 | my $self = shift (@_);
|
---|
[19493] | 1151 | my ($pluginfo, $base_dir, $file, $block_hash,
|
---|
| 1152 | $extrametakeys, $extrametadata, $extrametafile,
|
---|
[23212] | 1153 | $processor, $gli, $aux) = @_;
|
---|
[16390] | 1154 |
|
---|
| 1155 | # can we process this file??
|
---|
| 1156 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
[24403] | 1157 | return undef unless $self->can_process_this_file_for_metadata($filename_full_path);
|
---|
[16390] | 1158 |
|
---|
| 1159 | return 1; # we recognise the file, but don't actually do anything with it
|
---|
| 1160 | }
|
---|
| 1161 |
|
---|
| 1162 |
|
---|
[15868] | 1163 | # The BasePlugin read() function. This function calls read_into_doc_obj()
|
---|
[10280] | 1164 | # to ensure all the right things to make general options work for a
|
---|
| 1165 | # given plugin are done. It then calls the process() function which
|
---|
| 1166 | # does all the work specific to a plugin (like the old read functions
|
---|
| 1167 | # used to do). Most plugins should define their own process() function
|
---|
| 1168 | # and let this read() function keep control.
|
---|
| 1169 | #
|
---|
| 1170 | # recursive plugins (e.g. RecPlug) and specialized plugins like those
|
---|
| 1171 | # capable of processing many documents within a single file (e.g.
|
---|
| 1172 | # GMLPlug) might want to implement their own version of read(), but
|
---|
| 1173 | # more likely need to implement their own version of read_into_doc_obj()
|
---|
| 1174 | #
|
---|
| 1175 | # Return number of files processed, undef if can't recognise, -1 if can't
|
---|
| 1176 | # process
|
---|
| 1177 |
|
---|
| 1178 | sub read {
|
---|
| 1179 | my $self = shift (@_);
|
---|
[16390] | 1180 | my ($pluginfo, $base_dir, $file, $block_hash, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
[10280] | 1181 |
|
---|
[16390] | 1182 | # can we process this file??
|
---|
| 1183 | my ($filename_full_path, $filename_no_path) = &util::get_full_filenames($base_dir, $file);
|
---|
[20577] | 1184 |
|
---|
[16390] | 1185 | return undef unless $self->can_process_this_file($filename_full_path);
|
---|
| 1186 |
|
---|
[10280] | 1187 | my ($process_status,$doc_obj) = $self->read_into_doc_obj(@_);
|
---|
[21219] | 1188 |
|
---|
[10280] | 1189 | if ((defined $process_status) && ($process_status == 1)) {
|
---|
[21219] | 1190 |
|
---|
[10280] | 1191 | # process the document
|
---|
| 1192 | $processor->process($doc_obj);
|
---|
[15868] | 1193 |
|
---|
[10280] | 1194 | $self->{'num_processed'} ++;
|
---|
| 1195 | undef $doc_obj;
|
---|
[9398] | 1196 | }
|
---|
[15868] | 1197 | # delete any temp files that we may have created
|
---|
| 1198 | $self->clean_up_after_doc_obj_processing();
|
---|
[9398] | 1199 |
|
---|
[18469] | 1200 |
|
---|
[10280] | 1201 | # if process_status == 1, then the file has been processed.
|
---|
| 1202 | return $process_status;
|
---|
| 1203 |
|
---|
[4] | 1204 | }
|
---|
| 1205 |
|
---|
[1244] | 1206 | # returns undef if file is rejected by the plugin
|
---|
[1242] | 1207 | sub process {
|
---|
| 1208 | my $self = shift (@_);
|
---|
[11089] | 1209 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
---|
[1242] | 1210 |
|
---|
[23419] | 1211 | gsprintf(STDERR, "BasePlugin::process {common.must_be_implemented}\n");
|
---|
[1244] | 1212 |
|
---|
[23457] | 1213 | my ($cpackage,$cfilename,$cline,$csubr,$chas_args,$cwantarray) = caller(1);
|
---|
[23419] | 1214 | print STDERR "Calling method: $cfilename:$cline $cpackage->$csubr\n";
|
---|
| 1215 |
|
---|
[23457] | 1216 | die "\n";
|
---|
[23419] | 1217 |
|
---|
[1244] | 1218 | return undef; # never gets here
|
---|
[1242] | 1219 | }
|
---|
| 1220 |
|
---|
[15868] | 1221 | # overwrite this method to delete any temp files that we have created
|
---|
| 1222 | sub clean_up_after_doc_obj_processing {
|
---|
| 1223 | my $self = shift(@_);
|
---|
[4] | 1224 |
|
---|
[10280] | 1225 | }
|
---|
[16390] | 1226 |
|
---|
[10280] | 1227 | # write_file -- used by ConvertToPlug, for example in post processing
|
---|
| 1228 | #
|
---|
[15868] | 1229 | # where should this go, is here the best place??
|
---|
[10280] | 1230 | sub utf8_write_file {
|
---|
| 1231 | my $self = shift (@_);
|
---|
| 1232 | my ($textref, $filename) = @_;
|
---|
| 1233 |
|
---|
[22953] | 1234 | if (!open (FILE, ">:utf8", $filename)) {
|
---|
[10280] | 1235 | gsprintf(STDERR, "ConvertToPlug::write_file {ConvertToPlug.could_not_open_for_writing} ($!)\n", $filename);
|
---|
[23457] | 1236 | die "\n";
|
---|
| 1237 | }
|
---|
[10280] | 1238 | print FILE $$textref;
|
---|
| 1239 |
|
---|
[1219] | 1240 | close FILE;
|
---|
| 1241 | }
|
---|
| 1242 |
|
---|
[10280] | 1243 |
|
---|
[7504] | 1244 | sub filename_based_title
|
---|
| 1245 | {
|
---|
| 1246 | my $self = shift (@_);
|
---|
| 1247 | my ($file) = @_;
|
---|
| 1248 |
|
---|
| 1249 | my $file_derived_title = $file;
|
---|
| 1250 | $file_derived_title =~ s/_/ /g;
|
---|
[23335] | 1251 | $file_derived_title =~ s/\.[^.]+$//;
|
---|
[7504] | 1252 |
|
---|
| 1253 | return $file_derived_title;
|
---|
| 1254 | }
|
---|
| 1255 |
|
---|
[9398] | 1256 |
|
---|
[7504] | 1257 | sub title_fallback
|
---|
| 1258 | {
|
---|
| 1259 | my $self = shift (@_);
|
---|
| 1260 | my ($doc_obj,$section,$file) = @_;
|
---|
| 1261 |
|
---|
[22705] | 1262 | if (!defined $doc_obj->get_metadata_element ($section, "Title")
|
---|
[23335] | 1263 | || $doc_obj->get_metadata_element($section, "Title") eq "") {
|
---|
[7504] | 1264 |
|
---|
[23335] | 1265 | my $source_file = $doc_obj->get_metadata_element($section, "Source");
|
---|
| 1266 | my $file_derived_title;
|
---|
| 1267 | if (defined $source_file) {
|
---|
| 1268 | $file_derived_title = $self->filename_based_title($source_file);
|
---|
| 1269 | }
|
---|
| 1270 | else {
|
---|
| 1271 | # pp = pretty print
|
---|
| 1272 | my $pp_file = (defined $source_file) ? $source_file : $file;
|
---|
| 1273 |
|
---|
| 1274 | my $raw_title = $self->filename_based_title($file);
|
---|
| 1275 | my $file_derived_title = &unicode::raw_filename_to_url_encoded($raw_title);
|
---|
| 1276 | }
|
---|
| 1277 |
|
---|
| 1278 |
|
---|
[15868] | 1279 | if (!defined $doc_obj->get_metadata_element ($section, "Title")) {
|
---|
| 1280 | $doc_obj->add_utf8_metadata ($section, "Title", $file_derived_title);
|
---|
[9413] | 1281 | }
|
---|
[15868] | 1282 | else {
|
---|
[16995] | 1283 | $doc_obj->set_utf8_metadata_element ($section, "Title", $file_derived_title);
|
---|
[15868] | 1284 | }
|
---|
[9413] | 1285 | }
|
---|
[23457] | 1286 |
|
---|
[1844] | 1287 | }
|
---|
[23457] | 1288 |
|
---|
[1219] | 1289 | # add any extra metadata that's been passed around from one
|
---|
| 1290 | # plugin to another.
|
---|
| 1291 | # extra_metadata uses add_utf8_metadata so it expects metadata values
|
---|
| 1292 | # to already be in utf8
|
---|
| 1293 | sub extra_metadata {
|
---|
| 1294 | my $self = shift (@_);
|
---|
| 1295 | my ($doc_obj, $cursection, $metadata) = @_;
|
---|
| 1296 |
|
---|
[11122] | 1297 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 1298 |
|
---|
[1219] | 1299 | foreach my $field (keys(%$metadata)) {
|
---|
[839] | 1300 | # $metadata->{$field} may be an array reference
|
---|
[8510] | 1301 | if ($field eq "gsdlassocfile_tobe") {
|
---|
| 1302 | # 'gsdlassocfile_tobe' is artificially introduced metadata
|
---|
| 1303 | # that is used to signal that certain additional files should
|
---|
| 1304 | # be tied to this document. Useful in situations where a
|
---|
| 1305 | # metadata pass in the plugin pipeline works out some files
|
---|
| 1306 | # need to be associated with a document, but the document hasn't
|
---|
| 1307 | # been formed yet.
|
---|
| 1308 | my $equiv_form = "";
|
---|
| 1309 | foreach my $gaf (@{$metadata->{$field}}) {
|
---|
| 1310 | my ($full_filename,$mimetype) = ($gaf =~ m/^(.*):(.*):$/);
|
---|
[18171] | 1311 | my ($tail_filename) = ($full_filename =~ /^.*[\/\\](.+?)$/);
|
---|
| 1312 |
|
---|
| 1313 | # we need to make sure the filename is valid utf-8 - we do
|
---|
[18320] | 1314 | # this by url or base64 encoding it
|
---|
[18171] | 1315 | # $tail_filename is the name that we store the file as
|
---|
[18320] | 1316 | $tail_filename = &util::rename_file($tail_filename, $self->{'file_rename_method'});
|
---|
[8510] | 1317 | $doc_obj->associate_file($full_filename,$tail_filename,$mimetype);
|
---|
[20778] | 1318 | $doc_obj->associate_source_file($full_filename);
|
---|
[18320] | 1319 | # If the filename is url_encoded, we need to encode the % signs
|
---|
| 1320 | # in the filename, so that it works in a url
|
---|
[18404] | 1321 | my $url_tail_filename = &unicode::filename_to_url($tail_filename);
|
---|
[11122] | 1322 | # work out extended tail extension (i.e. matching tail re)
|
---|
| 1323 |
|
---|
| 1324 | my ($file_prefix,$file_extended_ext)
|
---|
[16390] | 1325 | = &util::get_prefix_and_tail_by_regex($tail_filename,$associate_tail_re);
|
---|
[11122] | 1326 | my ($pre_doc_ext) = ($file_extended_ext =~ m/^(.*)\..*$/);
|
---|
[8510] | 1327 | my ($doc_ext) = ($tail_filename =~ m/^.*\.(.*)$/);
|
---|
[24219] | 1328 |
|
---|
[25345] | 1329 | # the greenstone 2 stuff
|
---|
| 1330 | my $start_doclink = "<a href=\"_httpprefix_/collect/[collection]/index/assoc/{Or}{[parent(Top):assocfilepath],[assocfilepath]}/$url_tail_filename\">";
|
---|
| 1331 | #my $start_doclink = "<a href=\'_httpprefix_/collect/[collection]/index/assoc/[assocfilepath]/$url_tail_filename\'>";
|
---|
[24219] | 1332 | my $start_doclink_gs3 = "<a href=\'_httpprefix_/collect/[collection]/index/assoc/[assocfilepath]/$url_tail_filename\'>";
|
---|
| 1333 |
|
---|
[8510] | 1334 | my $srcicon = "_icon".$doc_ext."_";
|
---|
| 1335 | my $end_doclink = "</a>";
|
---|
[24219] | 1336 |
|
---|
[11122] | 1337 | my $assoc_form = "$start_doclink\{If\}{$srcicon,$srcicon,$doc_ext\}$end_doclink";
|
---|
| 1338 |
|
---|
[24219] | 1339 |
|
---|
[18171] | 1340 | if (defined $pre_doc_ext && $pre_doc_ext ne "") {
|
---|
[11122] | 1341 | # for metadata such as [mp3._edited] [mp3._full] ...
|
---|
| 1342 | $doc_obj->add_utf8_metadata ($cursection, "$doc_ext.$pre_doc_ext", $assoc_form);
|
---|
| 1343 | }
|
---|
| 1344 |
|
---|
| 1345 | # for multiple metadata such as [mp3.assoclink]
|
---|
| 1346 | $doc_obj->add_utf8_metadata ($cursection, "$doc_ext.assoclink", $assoc_form);
|
---|
[24219] | 1347 |
|
---|
[11122] | 1348 | $equiv_form .= " $assoc_form";
|
---|
[25345] | 1349 |
|
---|
| 1350 | # following are used for greenstone 3,
|
---|
| 1351 | $doc_obj->add_utf8_metadata ($cursection, "equivDocLink", $start_doclink_gs3);
|
---|
| 1352 | $doc_obj->add_utf8_metadata ($cursection, "equivDocIcon", $srcicon);
|
---|
| 1353 | $doc_obj->add_utf8_metadata ($cursection, "/equivDocLink", $end_doclink);
|
---|
| 1354 |
|
---|
[8510] | 1355 | }
|
---|
| 1356 | $doc_obj->add_utf8_metadata ($cursection, "equivlink", $equiv_form);
|
---|
| 1357 | }
|
---|
[23279] | 1358 | elsif ($field eq "gsdlzipfilename") {
|
---|
| 1359 | # special case for when files have come out of a zip. source_path
|
---|
| 1360 | # (used for archives dbs and keeping track for incremental import)
|
---|
| 1361 | # must be set to the zip file name
|
---|
| 1362 | my $zip_filename = $metadata->{$field};
|
---|
| 1363 | # overwrite the source_path
|
---|
[23363] | 1364 | $doc_obj->set_source_path($zip_filename);
|
---|
[23279] | 1365 | # and set the metadata
|
---|
| 1366 | $zip_filename = &util::filename_within_collection($zip_filename);
|
---|
| 1367 | $zip_filename = $doc_obj->encode_filename($zip_filename, $self->{'file_rename_method'});
|
---|
| 1368 | $doc_obj->add_utf8_metadata ($cursection, $field, $zip_filename);
|
---|
| 1369 | }
|
---|
[8510] | 1370 | elsif (ref ($metadata->{$field}) eq "ARRAY") {
|
---|
[839] | 1371 | map {
|
---|
[1219] | 1372 | $doc_obj->add_utf8_metadata ($cursection, $field, $_);
|
---|
[839] | 1373 | } @{$metadata->{$field}};
|
---|
| 1374 | } else {
|
---|
[1219] | 1375 | $doc_obj->add_utf8_metadata ($cursection, $field, $metadata->{$field});
|
---|
[839] | 1376 | }
|
---|
| 1377 | }
|
---|
| 1378 | }
|
---|
| 1379 |
|
---|
[1396] | 1380 |
|
---|
[2785] | 1381 | sub compile_stats {
|
---|
| 1382 | my $self = shift(@_);
|
---|
| 1383 | my ($stats) = @_;
|
---|
| 1384 |
|
---|
| 1385 | $stats->{'num_processed'} += $self->{'num_processed'};
|
---|
| 1386 | $stats->{'num_not_processed'} += $self->{'num_not_processed'};
|
---|
[2796] | 1387 | $stats->{'num_archives'} += $self->{'num_archives'};
|
---|
[2785] | 1388 |
|
---|
| 1389 | }
|
---|
[22215] | 1390 | sub associate_source_file {
|
---|
| 1391 | my $self = shift(@_);
|
---|
| 1392 |
|
---|
| 1393 | my ($doc_obj, $filename) = @_;
|
---|
| 1394 | my $cursection = $doc_obj->get_top_section();
|
---|
| 1395 | my $assocfilename = $doc_obj->get_assocfile_from_sourcefile();
|
---|
| 1396 |
|
---|
| 1397 | $doc_obj->associate_file($filename, $assocfilename, undef, $cursection);
|
---|
[24225] | 1398 | # srclink_file is now deprecated because of the "_" in the metadataname. Use srclinkFile
|
---|
[22663] | 1399 | $doc_obj->add_utf8_metadata ($cursection, "srclink_file", $doc_obj->get_sourcefile());
|
---|
[24225] | 1400 | $doc_obj->add_utf8_metadata ($cursection, "srclinkFile", $doc_obj->get_sourcefile());
|
---|
[22215] | 1401 | }
|
---|
[2785] | 1402 |
|
---|
[2816] | 1403 | sub associate_cover_image {
|
---|
[22215] | 1404 | my $self = shift(@_);
|
---|
[2816] | 1405 | my ($doc_obj, $filename) = @_;
|
---|
| 1406 |
|
---|
[23363] | 1407 | my $upgraded_filename = &util::upgrade_if_dos_filename($filename);
|
---|
| 1408 |
|
---|
[10833] | 1409 | $filename =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
[23363] | 1410 | $upgraded_filename =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
| 1411 |
|
---|
| 1412 | if (exists $self->{'covers_missing_cache'}->{$upgraded_filename}) {
|
---|
| 1413 | # don't stat() for existence e.g. for multiple document input files
|
---|
[10833] | 1414 | # (eg SplitPlug)
|
---|
| 1415 | return;
|
---|
| 1416 | }
|
---|
| 1417 |
|
---|
[9413] | 1418 | my $top_section=$doc_obj->get_top_section();
|
---|
| 1419 |
|
---|
[23363] | 1420 | if (&util::fd_exists($upgraded_filename)) {
|
---|
[20778] | 1421 | $doc_obj->associate_source_file($filename);
|
---|
[13968] | 1422 | $doc_obj->associate_file($filename, "cover.jpg", "image/jpeg");
|
---|
[9413] | 1423 | $doc_obj->add_utf8_metadata($top_section, "hascover", 1);
|
---|
[3086] | 1424 | } else {
|
---|
[10833] | 1425 | my $upper_filename = $filename;
|
---|
[23363] | 1426 | my $upgraded_upper_filename = $upgraded_filename;
|
---|
| 1427 |
|
---|
[10833] | 1428 | $upper_filename =~ s/jpg$/JPG/;
|
---|
[23363] | 1429 | $upgraded_upper_filename =~ s/jpg$/JPG/;
|
---|
| 1430 |
|
---|
| 1431 | if (&util::fd_exists($upgraded_upper_filename)) {
|
---|
[20778] | 1432 | $doc_obj->associate_source_file($upper_filename);
|
---|
[10833] | 1433 | $doc_obj->associate_file($upper_filename, "cover.jpg",
|
---|
| 1434 | "image/jpeg");
|
---|
[9413] | 1435 | $doc_obj->add_utf8_metadata($top_section, "hascover", 1);
|
---|
[10833] | 1436 | } else {
|
---|
| 1437 | # file doesn't exist, so record the fact that it's missing so
|
---|
| 1438 | # we don't stat() again (stat is slow)
|
---|
[23363] | 1439 | $self->{'covers_missing_cache'}->{$upgraded_filename} = 1;
|
---|
[3086] | 1440 | }
|
---|
[2816] | 1441 | }
|
---|
[10833] | 1442 |
|
---|
[2816] | 1443 | }
|
---|
| 1444 |
|
---|
[11332] | 1445 |
|
---|
| 1446 | # Overridden by exploding plugins (eg. ISISPlug)
|
---|
| 1447 | sub clean_up_after_exploding
|
---|
| 1448 | {
|
---|
| 1449 | my $self = shift(@_);
|
---|
| 1450 | }
|
---|
| 1451 |
|
---|
| 1452 |
|
---|
[16390] | 1453 |
|
---|
[4] | 1454 | 1;
|
---|