[31456] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
[31492] | 3 | # CommonUtil.pm -- base class for file and directory plugins - aims to
|
---|
| 4 | # handle all encoding stuff, blocking stuff, to keep it in one place
|
---|
[31456] | 5 | # A component of the Greenstone digital library software
|
---|
| 6 | # from the New Zealand Digital Library Project at the
|
---|
| 7 | # University of Waikato, New Zealand.
|
---|
| 8 | #
|
---|
| 9 | # Copyright (C) 2017 New Zealand Digital Library Project
|
---|
| 10 | #
|
---|
| 11 | # This program is free software; you can redistribute it and/or modify
|
---|
| 12 | # it under the terms of the GNU General Public License as published by
|
---|
| 13 | # the Free Software Foundation; either version 2 of the License, or
|
---|
| 14 | # (at your option) any later version.
|
---|
| 15 | #
|
---|
| 16 | # This program is distributed in the hope that it will be useful,
|
---|
| 17 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 18 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 19 | # GNU General Public License for more details.
|
---|
| 20 | #
|
---|
| 21 | # You should have received a copy of the GNU General Public License
|
---|
| 22 | # along with this program; if not, write to the Free Software
|
---|
| 23 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 24 | #
|
---|
| 25 | ###########################################################################
|
---|
| 26 |
|
---|
[31492] | 27 | package CommonUtil;
|
---|
[31456] | 28 |
|
---|
| 29 | use strict;
|
---|
| 30 | no strict 'subs';
|
---|
| 31 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
| 32 |
|
---|
| 33 | use encodings;
|
---|
[31491] | 34 | use Unicode::Normalize 'normalize';
|
---|
[31456] | 35 |
|
---|
| 36 | use PrintInfo;
|
---|
[31478] | 37 | use Encode;
|
---|
[31487] | 38 | use Unicode::Normalize 'normalize';
|
---|
[31456] | 39 |
|
---|
| 40 | BEGIN {
|
---|
[31492] | 41 | @CommonUtil::ISA = ( 'PrintInfo' );
|
---|
[31456] | 42 | }
|
---|
| 43 |
|
---|
| 44 | our $encoding_list =
|
---|
| 45 | [ { 'name' => "ascii",
|
---|
[31492] | 46 | 'desc' => "{BaseImporter.encoding.ascii}" },
|
---|
[31456] | 47 | { 'name' => "utf8",
|
---|
[31492] | 48 | 'desc' => "{BaseImporter.encoding.utf8}" },
|
---|
[31456] | 49 | { 'name' => "unicode",
|
---|
[31492] | 50 | 'desc' => "{BaseImporter.encoding.unicode}" } ];
|
---|
[31456] | 51 |
|
---|
| 52 |
|
---|
| 53 | my $e = $encodings::encodings;
|
---|
| 54 | foreach my $enc (sort {$e->{$a}->{'name'} cmp $e->{$b}->{'name'}} keys (%$e))
|
---|
| 55 | {
|
---|
| 56 | my $hashEncode =
|
---|
| 57 | {'name' => $enc,
|
---|
| 58 | 'desc' => $e->{$enc}->{'name'}};
|
---|
| 59 |
|
---|
| 60 | push(@{$encoding_list},$hashEncode);
|
---|
| 61 | }
|
---|
| 62 |
|
---|
| 63 | our $encoding_plus_auto_list =
|
---|
| 64 | [ { 'name' => "auto",
|
---|
[31492] | 65 | 'desc' => "{BaseImporter.filename_encoding.auto}" },
|
---|
[31456] | 66 | { 'name' => "auto-language-analysis",
|
---|
[31492] | 67 | 'desc' => "{BaseImporter.filename_encoding.auto_language_analysis}" }, # textcat
|
---|
[31456] | 68 | { 'name' => "auto-filesystem-encoding",
|
---|
[31492] | 69 | 'desc' => "{BaseImporter.filename_encoding.auto_filesystem_encoding}" }, # locale
|
---|
[31456] | 70 | { 'name' => "auto-fl",
|
---|
[31492] | 71 | 'desc' => "{BaseImporter.filename_encoding.auto_fl}" }, # locale followed by textcat
|
---|
[31456] | 72 | { 'name' => "auto-lf",
|
---|
[31492] | 73 | 'desc' => "{BaseImporter.filename_encoding.auto_lf}" } ]; # texcat followed by locale
|
---|
[31456] | 74 |
|
---|
| 75 | push(@{$encoding_plus_auto_list},@{$encoding_list});
|
---|
| 76 |
|
---|
| 77 | my $arguments =
|
---|
[31478] | 78 | [ { 'name' => "block_exp",
|
---|
[31492] | 79 | 'desc' => "{BaseImporter.block_exp}",
|
---|
[31478] | 80 | 'type' => "regexp",
|
---|
| 81 | 'deft' => "",
|
---|
| 82 | 'reqd' => "no" },
|
---|
| 83 | { 'name' => "no_blocking",
|
---|
[31492] | 84 | 'desc' => "{BaseImporter.no_blocking}",
|
---|
[31478] | 85 | 'type' => "flag",
|
---|
| 86 | 'reqd' => "no"},
|
---|
| 87 | { 'name' => "filename_encoding",
|
---|
[31492] | 88 | 'desc' => "{BaseImporter.filename_encoding}",
|
---|
[31478] | 89 | 'type' => "enum",
|
---|
| 90 | 'deft' => "auto",
|
---|
| 91 | 'list' => $encoding_plus_auto_list,
|
---|
| 92 | 'reqd' => "no" }
|
---|
| 93 | ];
|
---|
[31456] | 94 |
|
---|
[31492] | 95 | my $options = { 'name' => "CommonUtil",
|
---|
| 96 | 'desc' => "{CommonUtil.desc}",
|
---|
[31456] | 97 | 'abstract' => "yes",
|
---|
| 98 | 'inherits' => "no",
|
---|
| 99 | 'args' => $arguments };
|
---|
| 100 |
|
---|
| 101 |
|
---|
| 102 | sub new {
|
---|
| 103 |
|
---|
| 104 | my ($class) = shift (@_);
|
---|
| 105 | my ($pluginlist,$inputargs,$hashArgOptLists,$auxiliary) = @_;
|
---|
| 106 | push(@$pluginlist, $class);
|
---|
| 107 |
|
---|
| 108 | push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});
|
---|
| 109 | push(@{$hashArgOptLists->{"OptList"}},$options);
|
---|
| 110 |
|
---|
| 111 | my $self = new PrintInfo($pluginlist, $inputargs, $hashArgOptLists,$auxiliary);
|
---|
| 112 |
|
---|
| 113 | return bless $self, $class;
|
---|
| 114 |
|
---|
| 115 | }
|
---|
| 116 |
|
---|
| 117 | sub init {
|
---|
| 118 | my $self = shift (@_);
|
---|
| 119 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
| 120 |
|
---|
| 121 | print STDERR "guess encoding = ".$self->guess_filesystem_encoding()."\n";
|
---|
| 122 | print STDERR "get encoding = ".$self->get_filesystem_encoding()."\n";
|
---|
| 123 |
|
---|
| 124 | # verbosity is passed through from the processor
|
---|
| 125 | $self->{'verbosity'} = $verbosity;
|
---|
| 126 |
|
---|
| 127 | # as are the outhandle and failhandle
|
---|
| 128 | $self->{'outhandle'} = $outhandle if defined $outhandle;
|
---|
| 129 | $self->{'failhandle'} = $failhandle;
|
---|
| 130 |
|
---|
| 131 | }
|
---|
| 132 |
|
---|
[31478] | 133 | # converts raw filesystem filename to perl unicode format
|
---|
| 134 | sub raw_filename_to_unicode {
|
---|
| 135 | my $self = shift (@_);
|
---|
| 136 | my ($file) = @_;
|
---|
[31456] | 137 |
|
---|
[31478] | 138 | my $unicode_file = "";
|
---|
| 139 | ### need it in perl unicode, not raw filesystem
|
---|
| 140 | my $filename_encoding = $self->guess_filesystem_encoding();
|
---|
| 141 |
|
---|
[31492] | 142 | # copied this from set_Source_metadata in BaseImporter
|
---|
[31478] | 143 | if ((defined $filename_encoding) && ($filename_encoding ne "ascii")) {
|
---|
| 144 | # Use filename_encoding to map raw filename to a Perl unicode-aware string
|
---|
| 145 | $unicode_file = decode($filename_encoding,$file);
|
---|
| 146 | }
|
---|
| 147 | else {
|
---|
| 148 | # otherwise generate %xx encoded version of filename for char > 127
|
---|
| 149 | $unicode_file = &unicode::raw_filename_to_url_encoded($file);
|
---|
| 150 | }
|
---|
| 151 | return $unicode_file;
|
---|
| 152 |
|
---|
| 153 | }
|
---|
[31456] | 154 | # just converts path as is to utf8.
|
---|
| 155 | sub filepath_to_utf8 {
|
---|
| 156 | my $self = shift (@_);
|
---|
| 157 | my ($file, $file_encoding) = @_;
|
---|
| 158 | my $filemeta = $file;
|
---|
| 159 |
|
---|
| 160 | my $filename_encoding = $self->{'filename_encoding'}; # filename encoding setting
|
---|
| 161 |
|
---|
| 162 | # Whenever filename-encoding is set to any of the auto settings, we
|
---|
| 163 | # check if the filename is already in UTF8. If it is, then we're done.
|
---|
| 164 | if($filename_encoding =~ m/auto/) {
|
---|
| 165 | if(&unicode::check_is_utf8($filemeta))
|
---|
| 166 | {
|
---|
| 167 | $filename_encoding = "utf8";
|
---|
| 168 | return $filemeta;
|
---|
| 169 | }
|
---|
| 170 | }
|
---|
| 171 |
|
---|
| 172 | # Auto setting, but filename is not utf8
|
---|
| 173 | if ($filename_encoding eq "auto")
|
---|
| 174 | {
|
---|
| 175 | # try textcat
|
---|
| 176 | $filename_encoding = $self->textcat_encoding($filemeta);
|
---|
| 177 |
|
---|
| 178 | # check the locale next
|
---|
| 179 | $filename_encoding = $self->locale_encoding() if $filename_encoding eq "undefined";
|
---|
| 180 |
|
---|
| 181 |
|
---|
| 182 | # now try the encoding of the document, if available
|
---|
| 183 | if ($filename_encoding eq "undefined" && defined $file_encoding) {
|
---|
| 184 | $filename_encoding = $file_encoding;
|
---|
| 185 | }
|
---|
| 186 |
|
---|
| 187 | }
|
---|
| 188 |
|
---|
| 189 | elsif ($filename_encoding eq "auto-language-analysis")
|
---|
| 190 | {
|
---|
| 191 | $filename_encoding = $self->textcat_encoding($filemeta);
|
---|
| 192 |
|
---|
| 193 | # now try the encoding of the document, if available
|
---|
| 194 | if ($filename_encoding eq "undefined" && defined $file_encoding) {
|
---|
| 195 | $filename_encoding = $file_encoding;
|
---|
| 196 | }
|
---|
| 197 | }
|
---|
| 198 |
|
---|
| 199 | elsif ($filename_encoding eq "auto-filesystem-encoding")
|
---|
| 200 | {
|
---|
| 201 | # try locale
|
---|
| 202 | $filename_encoding = $self->locale_encoding();
|
---|
| 203 | }
|
---|
| 204 |
|
---|
| 205 | elsif ($filename_encoding eq "auto-fl")
|
---|
| 206 | {
|
---|
| 207 | # filesystem-encoding (locale) then language-analysis (textcat)
|
---|
| 208 | $filename_encoding = $self->locale_encoding();
|
---|
| 209 |
|
---|
| 210 | # try textcat
|
---|
| 211 | $filename_encoding = $self->textcat_encoding($filemeta) if $filename_encoding eq "undefined";
|
---|
| 212 |
|
---|
| 213 | # else assume filename encoding is encoding of file content, if that's available
|
---|
| 214 | if ($filename_encoding eq "undefined" && defined $file_encoding) {
|
---|
| 215 | $filename_encoding = $file_encoding;
|
---|
| 216 | }
|
---|
| 217 | }
|
---|
| 218 |
|
---|
| 219 | elsif ($filename_encoding eq "auto-lf")
|
---|
| 220 | {
|
---|
| 221 | # language-analysis (textcat) then filesystem-encoding (locale)
|
---|
| 222 | $filename_encoding = $self->textcat_encoding($filemeta);
|
---|
| 223 |
|
---|
| 224 | # guess filename encoding from encoding of file content, if available
|
---|
| 225 | if ($filename_encoding eq "undefined" && defined $file_encoding) {
|
---|
| 226 | $filename_encoding = $file_encoding;
|
---|
| 227 | }
|
---|
| 228 |
|
---|
| 229 | # try locale
|
---|
| 230 | $filename_encoding = $self->locale_encoding() if $filename_encoding eq "undefined";
|
---|
| 231 | }
|
---|
| 232 |
|
---|
| 233 | # if still undefined, use utf8 as fallback
|
---|
| 234 | if ($filename_encoding eq "undefined") {
|
---|
| 235 | $filename_encoding = "utf8";
|
---|
| 236 | }
|
---|
| 237 |
|
---|
| 238 | #print STDERR "**** UTF8 encoding the filename $filemeta ";
|
---|
| 239 |
|
---|
| 240 | # if the filename encoding is set to utf8 but it isn't utf8 already--such as when
|
---|
| 241 | # 1. the utf8 fallback is used, or 2. if the system locale is used and happens to
|
---|
| 242 | # be always utf8 (in which case the filename's encoding is also set as utf8 even
|
---|
| 243 | # though the filename need not be if it originates from another system)--in such
|
---|
| 244 | # cases attempt to make the filename utf8 to match.
|
---|
| 245 | if($filename_encoding eq "utf8" && !&unicode::check_is_utf8($filemeta)) {
|
---|
| 246 | &unicode::ensure_utf8(\$filemeta);
|
---|
| 247 | }
|
---|
| 248 |
|
---|
| 249 | # convert non-unicode encodings to utf8
|
---|
| 250 | if ($filename_encoding !~ m/(?:ascii|utf8|unicode)/) {
|
---|
| 251 | $filemeta = &unicode::unicode2utf8(
|
---|
| 252 | &unicode::convert2unicode($filename_encoding, \$filemeta)
|
---|
| 253 | );
|
---|
| 254 | }
|
---|
| 255 |
|
---|
| 256 | #print STDERR " from encoding $filename_encoding -> $filemeta\n";
|
---|
| 257 | return $filemeta;
|
---|
| 258 | }
|
---|
| 259 |
|
---|
| 260 | # gets the filename with no path, converts to utf8, and then dm safes it.
|
---|
| 261 | # filename_encoding set by user
|
---|
| 262 | sub filename_to_utf8_metadata
|
---|
| 263 | {
|
---|
| 264 | my $self = shift (@_);
|
---|
| 265 | my ($file, $file_encoding) = @_;
|
---|
| 266 |
|
---|
| 267 | my $outhandle = $self->{'outhandle'};
|
---|
| 268 |
|
---|
[31492] | 269 | print $outhandle "****!!!!**** BaseImporter::filename_to_utf8_metadata now deprecated\n";
|
---|
[31456] | 270 | my ($cpackage,$cfilename,$cline,$csubr,$chas_args,$cwantarray) = caller(0);
|
---|
| 271 | print $outhandle "Calling method: $cfilename:$cline $cpackage->$csubr\n";
|
---|
| 272 |
|
---|
| 273 | my ($filemeta) = $file =~ /([^\\\/]+)$/; # getting the tail of the filepath (skips all string parts containing slashes upto the end)
|
---|
| 274 | $filemeta = $self->filepath_to_utf8($filemeta, $file_encoding);
|
---|
| 275 |
|
---|
| 276 | return $filemeta;
|
---|
| 277 | }
|
---|
| 278 |
|
---|
| 279 | sub locale_encoding {
|
---|
| 280 | my $self = shift(@_);
|
---|
| 281 |
|
---|
| 282 | if (!defined $self->{'filesystem_encoding'}) {
|
---|
| 283 | $self->{'filesystem_encoding'} = $self->get_filesystem_encoding();
|
---|
| 284 | }
|
---|
| 285 |
|
---|
| 286 | #print STDERR "*** filename encoding determined based on locale: " . $self->{'filesystem_encoding'} . "\n";
|
---|
| 287 | return $self->{'filesystem_encoding'}; # can be the string "undefined"
|
---|
| 288 | }
|
---|
| 289 |
|
---|
| 290 |
|
---|
| 291 | sub textcat_encoding {
|
---|
| 292 | my $self = shift(@_);
|
---|
| 293 | my ($filemeta) = @_;
|
---|
| 294 |
|
---|
| 295 | # analyse filenames without extensions and digits (and trimmed of
|
---|
| 296 | # surrounding whitespace), so that irrelevant chars don't confuse
|
---|
| 297 | # textcat
|
---|
| 298 | my $strictfilemeta = $filemeta;
|
---|
| 299 | $strictfilemeta =~ s/\.[^\.]+$//g;
|
---|
| 300 | $strictfilemeta =~ s/\d//g;
|
---|
| 301 | $strictfilemeta =~ s/^\s*//g;
|
---|
| 302 | $strictfilemeta =~ s/\s*$//g;
|
---|
| 303 |
|
---|
| 304 | my $filename_encoding = $self->encoding_from_language_analysis($strictfilemeta);
|
---|
| 305 | if(!defined $filename_encoding) {
|
---|
| 306 | $filename_encoding = "undefined";
|
---|
| 307 | }
|
---|
| 308 |
|
---|
| 309 | return $filename_encoding; # can be the string "undefined"
|
---|
| 310 | }
|
---|
| 311 |
|
---|
| 312 | # performs textcat
|
---|
| 313 | sub encoding_from_language_analysis {
|
---|
| 314 | my $self = shift(@_);
|
---|
| 315 | my ($text) = @_;
|
---|
| 316 |
|
---|
| 317 | my $outhandle = $self->{'outhandle'};
|
---|
| 318 | my $best_encoding = undef;
|
---|
| 319 |
|
---|
| 320 | # get the language/encoding of the textstring using textcat
|
---|
| 321 | require textcat; # Only load the textcat module if it is required
|
---|
| 322 | $self->{'textcat'} = new textcat() unless defined($self->{'textcat'});
|
---|
| 323 | my $results = $self->{'textcat'}->classify_cached_filename(\$text);
|
---|
| 324 |
|
---|
| 325 |
|
---|
| 326 | if (scalar @$results < 0) {
|
---|
| 327 | return undef;
|
---|
| 328 | }
|
---|
| 329 |
|
---|
| 330 | # We have some results, we choose the first
|
---|
| 331 | my ($language, $encoding) = $results->[0] =~ /^([^-]*)(?:-(.*))?$/;
|
---|
| 332 |
|
---|
| 333 | $best_encoding = $encoding;
|
---|
| 334 | if (!defined $best_encoding) {
|
---|
| 335 | return undef;
|
---|
| 336 | }
|
---|
| 337 |
|
---|
| 338 | if (defined $best_encoding && $best_encoding =~ m/^iso_8859/ && &unicode::check_is_utf8($text)) {
|
---|
| 339 | # the text is valid utf8, so assume that's the real encoding (since textcat is based on probabilities)
|
---|
| 340 | $best_encoding = 'utf8';
|
---|
| 341 | }
|
---|
| 342 |
|
---|
| 343 |
|
---|
| 344 | # check for equivalents where textcat doesn't have some encodings...
|
---|
| 345 | # eg MS versions of standard encodings
|
---|
| 346 | if (defined $best_encoding && $best_encoding =~ /^iso_8859_(\d+)/) {
|
---|
| 347 | my $iso = $1; # which variant of the iso standard?
|
---|
| 348 | # iso-8859 sets don't use chars 0x80-0x9f, windows codepages do
|
---|
| 349 | if ($text =~ /[\x80-\x9f]/) {
|
---|
| 350 | # Western Europe
|
---|
| 351 | if ($iso == 1 or $iso == 15) { $best_encoding = 'windows_1252' }
|
---|
| 352 | elsif ($iso == 2) {$best_encoding = 'windows_1250'} # Central Europe
|
---|
| 353 | elsif ($iso == 5) {$best_encoding = 'windows_1251'} # Cyrillic
|
---|
| 354 | elsif ($iso == 6) {$best_encoding = 'windows_1256'} # Arabic
|
---|
| 355 | elsif ($iso == 7) {$best_encoding = 'windows_1253'} # Greek
|
---|
| 356 | elsif ($iso == 8) {$best_encoding = 'windows_1255'} # Hebrew
|
---|
| 357 | elsif ($iso == 9) {$best_encoding = 'windows_1254'} # Turkish
|
---|
| 358 | }
|
---|
| 359 | }
|
---|
| 360 |
|
---|
| 361 | if (defined $best_encoding && $best_encoding !~ /^(ascii|utf8|unicode)$/ &&
|
---|
| 362 | !defined $encodings::encodings->{$best_encoding})
|
---|
| 363 | {
|
---|
| 364 | if ($self->{'verbosity'}) {
|
---|
[31492] | 365 | gsprintf($outhandle, "BaseImporter: {ReadTextFile.unsupported_encoding}\n", $text, $best_encoding, "undef");
|
---|
[31456] | 366 | }
|
---|
| 367 | $best_encoding = undef;
|
---|
| 368 | }
|
---|
| 369 |
|
---|
| 370 | return $best_encoding;
|
---|
| 371 | }
|
---|
| 372 |
|
---|
| 373 |
|
---|
| 374 |
|
---|
| 375 | sub deduce_filename_encoding
|
---|
| 376 | {
|
---|
| 377 | my $self = shift (@_);
|
---|
| 378 | my ($file,$metadata,$plugin_filename_encoding) = @_;
|
---|
| 379 |
|
---|
| 380 | my $gs_filename_encoding = $metadata->{"gs.filenameEncoding"};
|
---|
| 381 | my $deduced_filename_encoding = undef;
|
---|
| 382 |
|
---|
| 383 | # Start by looking for manually assigned metadata
|
---|
| 384 | if (defined $gs_filename_encoding) {
|
---|
| 385 | if (ref ($gs_filename_encoding) eq "ARRAY") {
|
---|
| 386 | my $outhandle = $self->{'outhandle'};
|
---|
| 387 |
|
---|
| 388 | $deduced_filename_encoding = $gs_filename_encoding->[0];
|
---|
| 389 |
|
---|
| 390 | my $num_vals = scalar(@$gs_filename_encoding);
|
---|
| 391 | if ($num_vals>1) {
|
---|
| 392 | print $outhandle "Warning: gs.filenameEncoding multiply defined for $file\n";
|
---|
| 393 | print $outhandle " Selecting first value: $deduced_filename_encoding\n";
|
---|
| 394 | }
|
---|
| 395 | }
|
---|
| 396 | else {
|
---|
| 397 | $deduced_filename_encoding = $gs_filename_encoding;
|
---|
| 398 | }
|
---|
| 399 | }
|
---|
| 400 |
|
---|
| 401 | if (!defined $deduced_filename_encoding || ($deduced_filename_encoding =~ m/^\s*$/)) {
|
---|
| 402 | # Look to see if plugin specifies this value
|
---|
| 403 |
|
---|
| 404 | if (defined $plugin_filename_encoding) {
|
---|
| 405 | # First look to see if we're using any of the "older" (i.e. deprecated auto-... plugin options)
|
---|
| 406 | if ($plugin_filename_encoding =~ m/^auto-.*$/) {
|
---|
| 407 | my $outhandle = $self->{'outhandle'};
|
---|
| 408 | print $outhandle "Warning: $plugin_filename_encoding is no longer supported\n";
|
---|
| 409 | print $outhandle " default to 'auto'\n";
|
---|
| 410 | $self->{'filename_encoding'} = $plugin_filename_encoding = "auto";
|
---|
| 411 | }
|
---|
| 412 |
|
---|
| 413 | if ($plugin_filename_encoding ne "auto") {
|
---|
| 414 | # We've been given a specific filenamne encoding
|
---|
| 415 | # => so use it!
|
---|
| 416 | $deduced_filename_encoding = $plugin_filename_encoding;
|
---|
| 417 | }
|
---|
| 418 | }
|
---|
| 419 | }
|
---|
| 420 |
|
---|
| 421 | if (!defined $deduced_filename_encoding || ($deduced_filename_encoding =~ m/^\s*$/)) {
|
---|
| 422 |
|
---|
| 423 | # Look to file system to provide a character encoding
|
---|
| 424 |
|
---|
| 425 | # If Windows NTFS, then -- assuming we work with long file names got through
|
---|
| 426 | # Win32::GetLongFilePath() -- then the underlying file system is UTF16
|
---|
| 427 |
|
---|
| 428 | if (($ENV{'GSDLOS'} =~ m/^windows$/i) && ($^O ne "cygwin")) {
|
---|
| 429 | # Can do better than working with the DOS character encoding returned by locale
|
---|
| 430 | $deduced_filename_encoding = "unicode";
|
---|
| 431 | }
|
---|
| 432 | else {
|
---|
| 433 | # Unix of some form or other
|
---|
| 434 |
|
---|
| 435 | # See if we can determine the file system encoding through locale
|
---|
| 436 | $deduced_filename_encoding = $self->locale_encoding();
|
---|
| 437 |
|
---|
| 438 | # if locale shows us filesystem is utf8, check to see filename is consistent
|
---|
| 439 | # => if not, then we have an "alien" filename on our hands
|
---|
| 440 |
|
---|
| 441 | if (defined $deduced_filename_encoding && $deduced_filename_encoding =~ m/^utf-?8$/i) {
|
---|
| 442 | if (!&unicode::check_is_utf8($file)) {
|
---|
| 443 | # "alien" filename, so revert
|
---|
| 444 | $deduced_filename_encoding = undef;
|
---|
| 445 | }
|
---|
| 446 | }
|
---|
| 447 | }
|
---|
| 448 | }
|
---|
| 449 |
|
---|
| 450 | # if (!defined $deduced_filename_encoding || ($deduced_filename_encoding =~ m/^\s*$/)) {
|
---|
| 451 | # # Last chance, apply textcat to deduce filename encoding
|
---|
| 452 | # $deduced_filename_encoding = $self->textcat_encoding($file);
|
---|
| 453 | # }
|
---|
| 454 |
|
---|
| 455 | if ($self->{'verbosity'}>3) {
|
---|
| 456 | my $outhandle = $self->{'outhandle'};
|
---|
| 457 |
|
---|
| 458 | if (defined $deduced_filename_encoding) {
|
---|
| 459 | print $outhandle " Deduced filename encoding as: $deduced_filename_encoding\n";
|
---|
| 460 | }
|
---|
| 461 | else {
|
---|
| 462 | print $outhandle " No filename encoding deduced\n";
|
---|
| 463 | }
|
---|
| 464 | }
|
---|
| 465 |
|
---|
| 466 | return $deduced_filename_encoding;
|
---|
| 467 | }
|
---|
| 468 |
|
---|
| 469 |
|
---|
| 470 | sub guess_filesystem_encoding
|
---|
| 471 | {
|
---|
| 472 | my $self = shift (@_);
|
---|
| 473 | # Look to file system to provide a character encoding
|
---|
| 474 | my $deduced_filename_encoding = "";
|
---|
| 475 | # If Windows NTFS, then -- assuming we work with long file names got through
|
---|
| 476 | # Win32::GetLongFilePath() -- then the underlying file system is UTF16
|
---|
| 477 |
|
---|
| 478 | if (($ENV{'GSDLOS'} =~ m/^windows$/i) && ($^O ne "cygwin")) {
|
---|
| 479 | # Can do better than working with the DOS character encoding returned by locale
|
---|
| 480 | $deduced_filename_encoding = "unicode";
|
---|
| 481 | }
|
---|
| 482 | else {
|
---|
| 483 | # Unix of some form or other
|
---|
| 484 |
|
---|
| 485 | # See if we can determine the file system encoding through locale
|
---|
| 486 | $deduced_filename_encoding = $self->locale_encoding(); #utf8??
|
---|
| 487 |
|
---|
| 488 | }
|
---|
| 489 | print STDERR "guessing filesystem encoding is $deduced_filename_encoding\n";
|
---|
| 490 | return $deduced_filename_encoding;
|
---|
| 491 | }
|
---|
| 492 |
|
---|
| 493 |
|
---|
| 494 | # uses locale
|
---|
| 495 | sub get_filesystem_encoding
|
---|
| 496 | {
|
---|
| 497 |
|
---|
| 498 | my $self = shift(@_);
|
---|
| 499 |
|
---|
| 500 | my $outhandle = $self->{'outhandle'};
|
---|
| 501 | my $filesystem_encoding = undef;
|
---|
| 502 |
|
---|
| 503 | eval {
|
---|
| 504 | # Works for Windows as well, returning the DOS code page in use
|
---|
| 505 | use POSIX qw(locale_h);
|
---|
| 506 |
|
---|
| 507 | # With only one parameter, setlocale retrieves the
|
---|
| 508 | # current value
|
---|
| 509 | my $current_locale = setlocale(LC_CTYPE);
|
---|
| 510 |
|
---|
| 511 | my $char_encoding = undef;
|
---|
| 512 | if ($current_locale =~ m/\./) {
|
---|
| 513 | ($char_encoding) = ($current_locale =~ m/^.*\.(.*?)$/);
|
---|
| 514 | $char_encoding = lc($char_encoding);
|
---|
| 515 | }
|
---|
| 516 | else {
|
---|
| 517 | if ($current_locale =~ m/^(posix|c)$/i) {
|
---|
| 518 | $char_encoding = "ascii";
|
---|
| 519 | }
|
---|
| 520 | }
|
---|
| 521 |
|
---|
| 522 | if (defined $char_encoding) {
|
---|
| 523 | if ($char_encoding =~ m/^(iso)(8859)-?(\d{1,2})$/) {
|
---|
| 524 | $char_encoding = "$1\_$2\_$3";
|
---|
| 525 | }
|
---|
| 526 |
|
---|
| 527 | $char_encoding =~ s/-/_/g;
|
---|
| 528 | $char_encoding =~ s/^utf_8$/utf8/;
|
---|
| 529 |
|
---|
| 530 | if ($char_encoding =~ m/^\d+$/) {
|
---|
| 531 | if (defined $encodings::encodings->{"windows_$char_encoding"}) {
|
---|
| 532 | $char_encoding = "windows_$char_encoding";
|
---|
| 533 | }
|
---|
| 534 | elsif (defined $encodings::encodings->{"dos_$char_encoding"}) {
|
---|
| 535 | $char_encoding = "dos_$char_encoding";
|
---|
| 536 | }
|
---|
| 537 | }
|
---|
| 538 |
|
---|
| 539 | if (($char_encoding =~ m/(?:ascii|utf8|unicode)/)
|
---|
| 540 | || (defined $encodings::encodings->{$char_encoding})) {
|
---|
| 541 | $filesystem_encoding = $char_encoding;
|
---|
| 542 | }
|
---|
| 543 | else {
|
---|
| 544 | print $outhandle "Warning: Unsupported character encoding '$char_encoding' from locale '$current_locale'\n";
|
---|
| 545 | }
|
---|
| 546 | }
|
---|
| 547 |
|
---|
| 548 |
|
---|
| 549 | };
|
---|
| 550 | if ($@) {
|
---|
| 551 | print $outhandle "$@\n";
|
---|
| 552 | print $outhandle "Warning: Unable to establish locale. Will assume filesystem is UTF-8\n";
|
---|
| 553 |
|
---|
| 554 | }
|
---|
| 555 |
|
---|
| 556 | return $filesystem_encoding;
|
---|
| 557 | }
|
---|
| 558 |
|
---|
| 559 |
|
---|
| 560 |
|
---|
| 561 | # write_file -- used by ConvertToPlug, for example in post processing
|
---|
| 562 | #
|
---|
| 563 | # where should this go, is here the best place??
|
---|
| 564 | sub utf8_write_file {
|
---|
| 565 | my $self = shift (@_);
|
---|
| 566 | my ($textref, $filename) = @_;
|
---|
| 567 |
|
---|
| 568 | if (!open (FILE, ">:utf8", $filename)) {
|
---|
| 569 | gsprintf(STDERR, "ConvertToPlug::write_file {ConvertToPlug.could_not_open_for_writing} ($!)\n", $filename);
|
---|
| 570 | die "\n";
|
---|
| 571 | }
|
---|
| 572 | print FILE $$textref;
|
---|
| 573 |
|
---|
| 574 | close FILE;
|
---|
| 575 | }
|
---|
| 576 |
|
---|
[31478] | 577 | sub block_raw_filename {
|
---|
| 578 |
|
---|
| 579 | my $self = shift (@_);
|
---|
| 580 | my ($block_hash,$filename_full_path) = @_;
|
---|
| 581 |
|
---|
| 582 | my $unicode_filename = $self->raw_filename_to_unicode($filename_full_path);
|
---|
| 583 | return $self->block_filename($block_hash, $unicode_filename);
|
---|
| 584 | }
|
---|
| 585 |
|
---|
| 586 | # block unicode string filename
|
---|
| 587 | sub block_filename
|
---|
| 588 | {
|
---|
| 589 | my $self = shift (@_);
|
---|
| 590 | my ($block_hash,$filename_full_path) = @_;
|
---|
| 591 | print STDERR "in block filename $filename_full_path\n";
|
---|
| 592 | print STDERR &unicode::debug_unicode_string($filename_full_path)."\n";
|
---|
| 593 |
|
---|
| 594 | if (($ENV{'GSDLOS'} =~ m/^windows$/) && ($^O ne "cygwin")) {
|
---|
| 595 | # block hash contains long names, lets make sure that we were passed a long name
|
---|
| 596 | $filename_full_path = &util::upgrade_if_dos_filename($filename_full_path);
|
---|
| 597 | # lower case the entire thing, eg for cover.jpg when its actually cover.JPG
|
---|
| 598 | my $lower_filename_full_path = lc($filename_full_path);
|
---|
| 599 | $block_hash->{'file_blocks'}->{$lower_filename_full_path} = 1;
|
---|
| 600 |
|
---|
| 601 | }
|
---|
| 602 | elsif ($ENV{'GSDLOS'} =~ m/^darwin$/) {
|
---|
| 603 | # we need to normalize the filenames
|
---|
| 604 | my $composed_filename_full_path = normalize('C', $filename_full_path);
|
---|
| 605 | print STDERR "darwin, composed filename =". &unicode::debug_unicode_string($composed_filename_full_path)."\n";
|
---|
| 606 | $block_hash->{'file_blocks'}->{$composed_filename_full_path} = 1;
|
---|
| 607 | }
|
---|
| 608 |
|
---|
| 609 | else {
|
---|
| 610 | $block_hash->{'file_blocks'}->{$filename_full_path} = 1;
|
---|
| 611 | }
|
---|
| 612 | }
|
---|
| 613 |
|
---|
| 614 |
|
---|
| 615 | # filename is raw filesystem name
|
---|
| 616 | sub raw_file_is_blocked {
|
---|
| 617 | my $self = shift (@_);
|
---|
| 618 | my ($block_hash, $filename_full_path) = @_;
|
---|
| 619 |
|
---|
| 620 | my $unicode_filename_full_path = $self->raw_filename_to_unicode($filename_full_path);
|
---|
| 621 | return $self->file_is_blocked($block_hash, $unicode_filename_full_path);
|
---|
| 622 | }
|
---|
| 623 |
|
---|
| 624 | # filename must be perl unicode string
|
---|
| 625 | sub file_is_blocked {
|
---|
| 626 | my $self = shift (@_);
|
---|
| 627 | my ($block_hash, $filename_full_path) = @_;
|
---|
| 628 |
|
---|
| 629 | #
|
---|
| 630 | print STDERR "in file is blocked $filename_full_path\n";
|
---|
| 631 | print STDERR &unicode::debug_unicode_string($filename_full_path)."\n";
|
---|
| 632 | if (($ENV{'GSDLOS'} =~ m/^windows$/) && ($^O ne "cygwin")) {
|
---|
| 633 | # convert to long filenames if needed
|
---|
| 634 | $filename_full_path = &util::upgrade_if_dos_filename($filename_full_path);
|
---|
| 635 | # all block paths are lowercased.
|
---|
| 636 | my $lower_filename = lc ($filename_full_path);
|
---|
| 637 | if (defined $block_hash->{'file_blocks'}->{$lower_filename}) {
|
---|
| 638 | $self->{'num_blocked'} ++;
|
---|
| 639 | return 1;
|
---|
| 640 | }
|
---|
| 641 | }
|
---|
[31491] | 642 | elsif ($ENV{'GSDLOS'} =~ m/^darwin$/) {
|
---|
| 643 |
|
---|
| 644 | # on mac, we want composed form in the block hash
|
---|
| 645 | my $composed_form = normalize('C', $filename_full_path);
|
---|
| 646 | print STDERR "gsdlos = darwin, composed = ". &unicode::debug_unicode_string($composed_form) ."\n";
|
---|
| 647 | if (defined $block_hash->{'file_blocks'}->{$composed_form}) {
|
---|
| 648 | $self->{'num_blocked'} ++;
|
---|
| 649 | print STDERR "BLOCKED 1\n";
|
---|
| 650 | return 1;
|
---|
| 651 | }
|
---|
| 652 | }
|
---|
| 653 |
|
---|
[31478] | 654 | else {
|
---|
| 655 | if (defined $block_hash->{'file_blocks'}->{$filename_full_path}) {
|
---|
| 656 | $self->{'num_blocked'} ++;
|
---|
| 657 | print STDERR "BLOCKED\n";
|
---|
| 658 | return 1;
|
---|
| 659 | }
|
---|
| 660 | }
|
---|
| 661 | # check Directory plugin's own block_exp
|
---|
| 662 | if ($self->{'block_exp'} ne "" && $filename_full_path =~ /$self->{'block_exp'}/) {
|
---|
| 663 | $self->{'num_blocked'} ++;
|
---|
| 664 | return 1; # blocked
|
---|
| 665 | }
|
---|
| 666 | print STDERR "NOT BLOCKED\n";
|
---|
| 667 | return 0;
|
---|
| 668 | }
|
---|
| 669 |
|
---|
| 670 |
|
---|
[31456] | 671 | 1;
|
---|
| 672 |
|
---|