[537] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
| 3 | # BasPlug.pm -- base class for all the import plugins
|
---|
| 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 |
|
---|
| 26 | package BasPlug;
|
---|
[2219] | 27 |
|
---|
[9398] | 28 | BEGIN {
|
---|
| 29 | die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
|
---|
| 30 | }
|
---|
| 31 |
|
---|
[3834] | 32 | eval {require bytes};
|
---|
[3767] | 33 |
|
---|
[2219] | 34 | # suppress the annoying "subroutine redefined" warning that various
|
---|
| 35 | # plugins cause under perl 5.6
|
---|
| 36 | $SIG{__WARN__} = sub {warn($_[0]) unless ($_[0] =~ /Subroutine\s+\S+\sredefined/)};
|
---|
| 37 |
|
---|
[10254] | 38 | use strict;
|
---|
| 39 | no strict 'subs';
|
---|
| 40 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
[9413] | 41 |
|
---|
[8892] | 42 | use File::Basename;
|
---|
| 43 |
|
---|
[1954] | 44 | use Kea;
|
---|
[1219] | 45 | use multiread;
|
---|
[1870] | 46 | use encodings;
|
---|
[11389] | 47 | use unicode;
|
---|
[1219] | 48 | use cnseg;
|
---|
[1242] | 49 | use acronym;
|
---|
[1317] | 50 | use textcat;
|
---|
[1242] | 51 | use doc;
|
---|
[7645] | 52 | eval "require diagnostics"; # some perl distros (eg mac) don't have this
|
---|
[1411] | 53 | use DateExtract;
|
---|
[2751] | 54 | use ghtml;
|
---|
[9413] | 55 | use gsprintf 'gsprintf';
|
---|
[4778] | 56 | use printusage;
|
---|
[10218] | 57 | use parse2;
|
---|
[4] | 58 |
|
---|
[10218] | 59 |
|
---|
[9398] | 60 | use GISBasPlug;
|
---|
[5681] | 61 |
|
---|
[9413] | 62 | @BasPlug::ISA = ( GISBasPlug );
|
---|
[9398] | 63 |
|
---|
[4873] | 64 | my $unicode_list =
|
---|
[10218] | 65 | [ { 'name' => "ascii",
|
---|
[4873] | 66 | 'desc' => "{BasPlug.input_encoding.ascii}" },
|
---|
[4744] | 67 | { 'name' => "utf8",
|
---|
[4873] | 68 | 'desc' => "{BasPlug.input_encoding.utf8}" },
|
---|
[4744] | 69 | { 'name' => "unicode",
|
---|
[4873] | 70 | 'desc' => "{BasPlug.input_encoding.unicode}" } ];
|
---|
[3540] | 71 |
|
---|
[10218] | 72 | my $auto_unicode_list =
|
---|
| 73 | [ { 'name' => "auto",
|
---|
| 74 | 'desc' => "{BasPlug.input_encoding.auto}" } ];
|
---|
| 75 |
|
---|
[10620] | 76 | my $e = $encodings::encodings;
|
---|
| 77 | foreach my $enc (sort {$e->{$a}->{'name'} cmp $e->{$b}->{'name'}} keys (%$e))
|
---|
| 78 | {
|
---|
| 79 | my $hashEncode =
|
---|
| 80 | {'name' => $enc,
|
---|
| 81 | 'desc' => $e->{$enc}->{'name'}};
|
---|
| 82 |
|
---|
| 83 | push(@{$unicode_list},$hashEncode);
|
---|
| 84 | }
|
---|
| 85 |
|
---|
| 86 | push(@{$auto_unicode_list},@{$unicode_list});
|
---|
| 87 |
|
---|
[4873] | 88 | my $arguments =
|
---|
[3540] | 89 | [ { 'name' => "process_exp",
|
---|
[4873] | 90 | 'desc' => "{BasPlug.process_exp}",
|
---|
[6408] | 91 | 'type' => "regexp",
|
---|
[3540] | 92 | 'deft' => "",
|
---|
| 93 | 'reqd' => "no" },
|
---|
| 94 | { 'name' => "block_exp",
|
---|
[4873] | 95 | 'desc' => "{BasPlug.block_exp}",
|
---|
[6408] | 96 | 'type' => "regexp",
|
---|
[3540] | 97 | 'deft' => "",
|
---|
| 98 | 'reqd' => "no" },
|
---|
[9067] | 99 | { 'name' => "smart_block",
|
---|
| 100 | 'desc' => "{BasPlug.smart_block}",
|
---|
| 101 | 'type' => "flag",
|
---|
[11368] | 102 | 'reqd' => "no" },
|
---|
[8892] | 103 | { 'name' => "associate_ext",
|
---|
| 104 | 'desc' => "{BasPlug.associate_ext}",
|
---|
| 105 | 'type' => "string",
|
---|
| 106 | 'reqd' => "no" },
|
---|
[11122] | 107 | { 'name' => "associate_tail_re",
|
---|
| 108 | 'desc' => "{BasPlug.associate_tail_re}",
|
---|
| 109 | 'type' => "string",
|
---|
| 110 | 'reqd' => "no" },
|
---|
[3540] | 111 | { 'name' => "input_encoding",
|
---|
[4873] | 112 | 'desc' => "{BasPlug.input_encoding}",
|
---|
[3540] | 113 | 'type' => "enum",
|
---|
[10218] | 114 | 'list' => $auto_unicode_list,
|
---|
[3540] | 115 | 'reqd' => "no" ,
|
---|
| 116 | 'deft' => "auto" } ,
|
---|
| 117 | { 'name' => "default_encoding",
|
---|
[4873] | 118 | 'desc' => "{BasPlug.default_encoding}",
|
---|
[4744] | 119 | 'type' => "enum",
|
---|
[6332] | 120 | 'list' => $unicode_list,
|
---|
[4744] | 121 | 'reqd' => "no",
|
---|
| 122 | 'deft' => "utf8" },
|
---|
[3540] | 123 | { 'name' => "extract_language",
|
---|
[4873] | 124 | 'desc' => "{BasPlug.extract_language}",
|
---|
[3540] | 125 | 'type' => "flag",
|
---|
| 126 | 'reqd' => "no" },
|
---|
| 127 | { 'name' => "default_language",
|
---|
[4873] | 128 | 'desc' => "{BasPlug.default_language}",
|
---|
[10329] | 129 | 'type' => "string",
|
---|
[3540] | 130 | 'deft' => "en",
|
---|
| 131 | 'reqd' => "no" },
|
---|
| 132 | { 'name' => "extract_acronyms",
|
---|
[4873] | 133 | 'desc' => "{BasPlug.extract_acronyms}",
|
---|
[3540] | 134 | 'type' => "flag",
|
---|
| 135 | 'reqd' => "no" },
|
---|
| 136 | { 'name' => "markup_acronyms",
|
---|
[4873] | 137 | 'desc' => "{BasPlug.markup_acronyms}",
|
---|
[3540] | 138 | 'type' => "flag",
|
---|
[9398] | 139 | 'reqd' => "no" },
|
---|
[8789] | 140 | { 'name' => "extract_keyphrases",
|
---|
| 141 | 'desc' => "{BasPlug.extract_keyphrases}",
|
---|
| 142 | 'type' => "flag",
|
---|
[8814] | 143 | 'reqd' => "no" },
|
---|
[11069] | 144 | { 'name' => "extract_keyphrases_kea4",
|
---|
| 145 | 'desc' => "{BasPlug.extract_keyphrases_kea4}",
|
---|
| 146 | 'type' => "flag",
|
---|
| 147 | 'reqd' => "no" },
|
---|
[8789] | 148 | { 'name' => "extract_keyphrase_options",
|
---|
| 149 | 'desc' => "{BasPlug.extract_keyphrase_options}",
|
---|
| 150 | 'type' => "string",
|
---|
| 151 | 'deft' => "",
|
---|
[8814] | 152 | 'reqd' => "no" },
|
---|
[3540] | 153 | { 'name' => "first",
|
---|
[4873] | 154 | 'desc' => "{BasPlug.first}",
|
---|
[3540] | 155 | 'type' => "string",
|
---|
| 156 | 'reqd' => "no" },
|
---|
| 157 | { 'name' => "extract_email",
|
---|
[4873] | 158 | 'desc' => "{BasPlug.extract_email}",
|
---|
[3540] | 159 | 'type' => "flag",
|
---|
| 160 | 'reqd' => "no" },
|
---|
| 161 | { 'name' => "extract_historical_years",
|
---|
[4873] | 162 | 'desc' => "{BasPlug.extract_historical_years}",
|
---|
[3540] | 163 | 'type' => "flag",
|
---|
| 164 | 'reqd' => "no" },
|
---|
| 165 | { 'name' => "maximum_year",
|
---|
[4873] | 166 | 'desc' => "{BasPlug.maximum_year}",
|
---|
[3540] | 167 | 'type' => "int",
|
---|
[4744] | 168 | 'deft' => (localtime)[5]+1900,
|
---|
[10218] | 169 | 'char_length' => "4",
|
---|
| 170 | #'range' => "2,100",
|
---|
[3540] | 171 | 'reqd' => "no"},
|
---|
| 172 | { 'name' => "maximum_century",
|
---|
[4873] | 173 | 'desc' => "{BasPlug.maximum_century}",
|
---|
[7105] | 174 | 'type' => "string",
|
---|
[10218] | 175 | 'deft' => "-1",
|
---|
[3540] | 176 | 'reqd' => "no" },
|
---|
| 177 | { 'name' => "no_bibliography",
|
---|
[4873] | 178 | 'desc' => "{BasPlug.no_bibliography}",
|
---|
[3540] | 179 | 'type' => "flag",
|
---|
| 180 | 'reqd' => "no"},
|
---|
[8678] | 181 | { 'name' => "no_cover_image",
|
---|
| 182 | 'desc' => "{BasPlug.no_cover_image}",
|
---|
[3540] | 183 | 'type' => "flag",
|
---|
[10218] | 184 | 'reqd' => "no" },
|
---|
| 185 | { 'name' => "separate_cjk",
|
---|
| 186 | 'desc' => "{BasPlug.separate_cjk}",
|
---|
| 187 | 'type' => "flag",
|
---|
| 188 | 'reqd' => "no",
|
---|
| 189 | 'hiddengli' => "yes" },
|
---|
| 190 | { 'name' => "new_extract_email",
|
---|
| 191 | 'desc' => "",
|
---|
| 192 | 'type' => "flag",
|
---|
| 193 | 'reqd' => "no",
|
---|
| 194 | 'hiddengli' => "yes" } ];
|
---|
[3540] | 195 |
|
---|
[9398] | 196 | my $gis_arguments =
|
---|
| 197 | [ { 'name' => "extract_placenames",
|
---|
| 198 | 'desc' => "{GISBasPlug.extract_placenames}",
|
---|
| 199 | 'type' => "flag",
|
---|
| 200 | 'reqd' => "no" },
|
---|
| 201 | { 'name' => "gazetteer",
|
---|
| 202 | 'desc' => "{GISBasPlug.gazetteer}",
|
---|
| 203 | 'type' => "string",
|
---|
| 204 | 'reqd' => "no" },
|
---|
| 205 | { 'name' => "place_list",
|
---|
| 206 | 'desc' => "{GISBasPlug.place_list}",
|
---|
| 207 | 'type' => "flag",
|
---|
| 208 | 'reqd' => "no" } ];
|
---|
| 209 |
|
---|
| 210 |
|
---|
[3540] | 211 | my $options = { 'name' => "BasPlug",
|
---|
[5681] | 212 | 'desc' => "{BasPlug.desc}",
|
---|
[6408] | 213 | 'abstract' => "yes",
|
---|
| 214 | 'inherits' => "no",
|
---|
[4750] | 215 | 'args' => $arguments };
|
---|
[3540] | 216 |
|
---|
[4778] | 217 |
|
---|
[10478] | 218 | sub set_keepold {
|
---|
| 219 | my $self = shift(@_);
|
---|
| 220 | my ($keepold) = @_;
|
---|
| 221 |
|
---|
| 222 | $self->{'keepold'} = $keepold;
|
---|
| 223 | }
|
---|
| 224 |
|
---|
[4873] | 225 | sub get_arguments
|
---|
| 226 | {
|
---|
[8716] | 227 | my $self = shift(@_);
|
---|
| 228 | my $optionlistref = $self->{'option_list'};
|
---|
| 229 | my @optionlist = @$optionlistref;
|
---|
| 230 | my $pluginoptions = pop(@$optionlistref);
|
---|
| 231 | my $pluginarguments = $pluginoptions->{'args'};
|
---|
[4873] | 232 | return $pluginarguments;
|
---|
| 233 | }
|
---|
| 234 |
|
---|
| 235 |
|
---|
[4778] | 236 | sub print_xml_usage
|
---|
| 237 | {
|
---|
[8716] | 238 | my $self = shift(@_);
|
---|
[11681] | 239 | my $header = shift(@_);
|
---|
| 240 |
|
---|
[6945] | 241 | # XML output is always in UTF-8
|
---|
[9413] | 242 | gsprintf::output_strings_in_UTF8;
|
---|
[6945] | 243 |
|
---|
[11681] | 244 | if ($header) {
|
---|
| 245 | &PrintUsage::print_xml_header("plugin");
|
---|
| 246 | }
|
---|
[6925] | 247 | $self->print_xml();
|
---|
[3540] | 248 | }
|
---|
| 249 |
|
---|
[4778] | 250 |
|
---|
| 251 | sub print_xml
|
---|
| 252 | {
|
---|
[8716] | 253 | my $self = shift(@_);
|
---|
[4778] | 254 |
|
---|
[8716] | 255 | my $optionlistref = $self->{'option_list'};
|
---|
| 256 | my @optionlist = @$optionlistref;
|
---|
[10229] | 257 | my $pluginoptions = shift(@$optionlistref);
|
---|
[4778] | 258 | return if (!defined($pluginoptions));
|
---|
| 259 |
|
---|
[9413] | 260 | gsprintf(STDERR, "<PlugInfo>\n");
|
---|
| 261 | gsprintf(STDERR, " <Name>$pluginoptions->{'name'}</Name>\n");
|
---|
| 262 | my $desc = gsprintf::lookup_string($pluginoptions->{'desc'});
|
---|
[7023] | 263 | $desc =~ s/</&lt;/g; # doubly escaped
|
---|
| 264 | $desc =~ s/>/&gt;/g;
|
---|
| 265 |
|
---|
[9413] | 266 | gsprintf(STDERR, " <Desc>$desc</Desc>\n");
|
---|
| 267 | gsprintf(STDERR, " <Abstract>$pluginoptions->{'abstract'}</Abstract>\n");
|
---|
| 268 | gsprintf(STDERR, " <Inherits>$pluginoptions->{'inherits'}</Inherits>\n");
|
---|
| 269 | gsprintf(STDERR, " <Explodes>" . ($pluginoptions->{'explodes'} || "no") . "</Explodes>\n");
|
---|
| 270 | gsprintf(STDERR, " <Arguments>\n");
|
---|
[4778] | 271 | if (defined($pluginoptions->{'args'})) {
|
---|
[6925] | 272 | &PrintUsage::print_options_xml($pluginoptions->{'args'});
|
---|
[3540] | 273 | }
|
---|
[11669] | 274 | gsprintf(STDERR, " </Arguments>\n");
|
---|
[4778] | 275 | # Recurse up the plugin hierarchy
|
---|
[6925] | 276 | $self->print_xml();
|
---|
[9413] | 277 | gsprintf(STDERR, "</PlugInfo>\n");
|
---|
[3540] | 278 | }
|
---|
| 279 |
|
---|
[4744] | 280 |
|
---|
[4778] | 281 | sub print_txt_usage
|
---|
[4744] | 282 | {
|
---|
[8716] | 283 | my $self = shift(@_);
|
---|
[4750] | 284 | # Print the usage message for a plugin (recursively)
|
---|
[8716] | 285 | my $descoffset = $self->determine_description_offset(0);
|
---|
[6925] | 286 | $self->print_plugin_usage($descoffset, 1);
|
---|
[4750] | 287 | }
|
---|
[4744] | 288 |
|
---|
| 289 |
|
---|
[4750] | 290 | sub determine_description_offset
|
---|
| 291 | {
|
---|
[8716] | 292 | my $self = shift(@_);
|
---|
| 293 | my $maxoffset = shift(@_);
|
---|
[4750] | 294 |
|
---|
[8716] | 295 | my $optionlistref = $self->{'option_list'};
|
---|
| 296 | my @optionlist = @$optionlistref;
|
---|
[10229] | 297 | my $pluginoptions = shift(@$optionlistref);
|
---|
[4750] | 298 | return $maxoffset if (!defined($pluginoptions));
|
---|
| 299 |
|
---|
| 300 | # Find the length of the longest option string of this plugin
|
---|
[8716] | 301 | my $pluginargs = $pluginoptions->{'args'};
|
---|
[4744] | 302 | if (defined($pluginargs)) {
|
---|
[8716] | 303 | my $longest = &PrintUsage::find_longest_option_string($pluginargs);
|
---|
[4778] | 304 | if ($longest > $maxoffset) {
|
---|
| 305 | $maxoffset = $longest;
|
---|
[4744] | 306 | }
|
---|
[4750] | 307 | }
|
---|
[4744] | 308 |
|
---|
[4750] | 309 | # Recurse up the plugin hierarchy
|
---|
| 310 | $maxoffset = $self->determine_description_offset($maxoffset);
|
---|
| 311 | $self->{'option_list'} = \@optionlist;
|
---|
| 312 | return $maxoffset;
|
---|
| 313 | }
|
---|
| 314 |
|
---|
| 315 |
|
---|
| 316 | sub print_plugin_usage
|
---|
| 317 | {
|
---|
[8716] | 318 | my $self = shift(@_);
|
---|
| 319 | my $descoffset = shift(@_);
|
---|
| 320 | my $isleafclass = shift(@_);
|
---|
[4750] | 321 |
|
---|
[8716] | 322 | my $optionlistref = $self->{'option_list'};
|
---|
| 323 | my @optionlist = @$optionlistref;
|
---|
[10229] | 324 | my $pluginoptions = shift(@$optionlistref);
|
---|
[4750] | 325 | return if (!defined($pluginoptions));
|
---|
| 326 |
|
---|
[8716] | 327 | my $pluginname = $pluginoptions->{'name'};
|
---|
| 328 | my $pluginargs = $pluginoptions->{'args'};
|
---|
| 329 | my $plugindesc = $pluginoptions->{'desc'};
|
---|
[4750] | 330 |
|
---|
| 331 | # Produce the usage information using the data structure above
|
---|
| 332 | if ($isleafclass) {
|
---|
[6932] | 333 | if (defined($plugindesc)) {
|
---|
[9413] | 334 | gsprintf(STDERR, "$plugindesc\n\n");
|
---|
[6932] | 335 | }
|
---|
[9413] | 336 | gsprintf(STDERR, " {common.usage}: plugin $pluginname [{common.options}]\n\n");
|
---|
[4750] | 337 | }
|
---|
| 338 |
|
---|
| 339 | # Display the plugin options, if there are some
|
---|
| 340 | if (defined($pluginargs)) {
|
---|
[4744] | 341 | # Calculate the column offset of the option descriptions
|
---|
[8716] | 342 | my $optiondescoffset = $descoffset + 2; # 2 spaces between options & descriptions
|
---|
[4744] | 343 |
|
---|
[4750] | 344 | if ($isleafclass) {
|
---|
[9413] | 345 | gsprintf(STDERR, " {common.specific_options}:\n");
|
---|
[4750] | 346 | }
|
---|
| 347 | else {
|
---|
[9413] | 348 | gsprintf(STDERR, " {common.general_options}:\n", $pluginname);
|
---|
[4750] | 349 | }
|
---|
| 350 |
|
---|
[4744] | 351 | # Display the plugin options
|
---|
[6925] | 352 | &PrintUsage::print_options_txt($pluginargs, $optiondescoffset);
|
---|
[4744] | 353 | }
|
---|
| 354 |
|
---|
[4750] | 355 | # Recurse up the plugin hierarchy
|
---|
[6925] | 356 | $self->print_plugin_usage($descoffset, 0);
|
---|
[4750] | 357 | $self->{'option_list'} = \@optionlist;
|
---|
[4744] | 358 | }
|
---|
| 359 |
|
---|
| 360 |
|
---|
[4] | 361 | sub new {
|
---|
[10218] | 362 | # Set Encodings to the list!!
|
---|
| 363 |
|
---|
| 364 |
|
---|
| 365 | # Start the BasPlug Constructor
|
---|
[1219] | 366 | my $class = shift (@_);
|
---|
[10218] | 367 | my ($pluginlist,$args,$hashArgOptLists) = @_;
|
---|
| 368 | push(@$pluginlist, $class);
|
---|
| 369 | my $plugin_name = (defined $pluginlist->[0]) ? $pluginlist->[0] : $class;
|
---|
[9398] | 370 |
|
---|
[10218] | 371 | if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
|
---|
| 372 | if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
|
---|
| 373 |
|
---|
[9398] | 374 | if (GISBasPlug::has_mapdata()) {
|
---|
| 375 | push(@$arguments,@$gis_arguments);
|
---|
| 376 | }
|
---|
[10218] | 377 |
|
---|
| 378 | my $self = {};
|
---|
[10579] | 379 | $self->{'outhandle'} = STDERR;
|
---|
| 380 | $self->{'option_list'} = $hashArgOptLists->{"OptList"};
|
---|
| 381 | $self->{"info_only"} = 0;
|
---|
| 382 |
|
---|
| 383 | # Check if gsdlinfo is in the argument list or not - if it is, don't parse
|
---|
| 384 | # the args, just return the object.
|
---|
| 385 | foreach my $strArg (@{$args})
|
---|
| 386 | {
|
---|
| 387 | if($strArg eq "-gsdlinfo")
|
---|
| 388 | {
|
---|
| 389 | $self->{"info_only"} = 1;
|
---|
| 390 | return bless $self, $class;
|
---|
| 391 | }
|
---|
| 392 | }
|
---|
| 393 |
|
---|
[10218] | 394 | if(!parse2::parse($args,$hashArgOptLists->{"ArgList"},$self))
|
---|
| 395 | {
|
---|
| 396 | my $classTempClass = bless $self, $class;
|
---|
[10620] | 397 | print STDERR "<BadPlugin p=$plugin_name>\n";
|
---|
[10218] | 398 | &gsprintf(STDERR, "\n{BasPlug.bad_general_option}\n", $plugin_name);
|
---|
| 399 | $classTempClass->print_txt_usage(""); # Use default resource bundle
|
---|
| 400 | die "\n";
|
---|
| 401 | }
|
---|
[9398] | 402 |
|
---|
[10280] | 403 |
|
---|
[10579] | 404 | delete $self->{"info_only"};
|
---|
[10218] | 405 | # else parsing was successful.
|
---|
| 406 |
|
---|
| 407 | $self->{'plugin_type'} = $plugin_name;
|
---|
[10579] | 408 | #$self->{'outhandle'} = STDERR;
|
---|
[2785] | 409 | $self->{'num_processed'} = 0;
|
---|
| 410 | $self->{'num_not_processed'} = 0;
|
---|
| 411 | $self->{'num_blocked'} = 0;
|
---|
| 412 | $self->{'num_archives'} = 0;
|
---|
[8678] | 413 | $self->{'cover_image'} = 1; # cover image is on by default
|
---|
[10218] | 414 | $self->{'cover_image'} = 0 if ($self->{'no_cover_image'});
|
---|
[10579] | 415 | #$self->{'option_list'} = $hashArgOptLists->{"OptList"};
|
---|
[3540] | 416 |
|
---|
[8892] | 417 | my $associate_ext = $self->{'associate_ext'};
|
---|
| 418 | if ((defined $associate_ext) && ($associate_ext ne "")) {
|
---|
[9351] | 419 |
|
---|
[11122] | 420 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 421 | if ((defined $associate_tail_re) && ($associate_tail_re ne "")) {
|
---|
| 422 | my $outhandle = $self->{'outhandle'};
|
---|
| 423 | print $outhandle "Warning: can only specify 'associate_ext' or 'associate_tail_re'\n";
|
---|
| 424 | print $outhandle " defaulting to 'associate_tail_re'\n";
|
---|
[8892] | 425 | }
|
---|
[11122] | 426 | else {
|
---|
| 427 | my @exts = split(/,/,$associate_ext);
|
---|
[8892] | 428 |
|
---|
[11122] | 429 | my @exts_bracketed = map { $_ = "(?:\\.$_)" } @exts;
|
---|
| 430 | my $associate_tail_re = join("|",@exts_bracketed);
|
---|
| 431 | $self->{'associate_tail_re'} = $associate_tail_re;
|
---|
| 432 | }
|
---|
| 433 |
|
---|
| 434 | delete $self->{'associate_ext'};
|
---|
[8892] | 435 | }
|
---|
| 436 |
|
---|
| 437 | $self->{'shared_fileroot'} = {};
|
---|
[8510] | 438 | $self->{'file_blocks'} = {};
|
---|
[1219] | 439 |
|
---|
[9398] | 440 | if ($self->{'extract_placenames'}) {
|
---|
| 441 |
|
---|
| 442 | my $outhandle = $self->{'outhandle'};
|
---|
[10218] | 443 |
|
---|
[9398] | 444 | my $places_ref
|
---|
| 445 | = GISBasPlug::loadGISDatabase($outhandle,$self->{'gazetteer'});
|
---|
[10218] | 446 |
|
---|
[9398] | 447 | if (!defined $places_ref) {
|
---|
| 448 | print $outhandle "Warning: Error loading mapdata gazetteer \"$self->{'gazetteer'}\"\n";
|
---|
| 449 | print $outhandle " No placename extraction will take place.\n";
|
---|
| 450 | $self->{'extract_placenames'} = undef;
|
---|
| 451 | }
|
---|
| 452 | else {
|
---|
| 453 | $self->{'places'} = $places_ref;
|
---|
| 454 | }
|
---|
[10280] | 455 | }
|
---|
[11089] | 456 |
|
---|
[1219] | 457 | return bless $self, $class;
|
---|
[10218] | 458 |
|
---|
[4] | 459 | }
|
---|
| 460 |
|
---|
[1242] | 461 | # initialize BasPlug options
|
---|
| 462 | # if init() is overridden in a sub-class, remember to call BasPlug::init()
|
---|
| 463 | sub init {
|
---|
| 464 | my $self = shift (@_);
|
---|
[2785] | 465 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
[1242] | 466 |
|
---|
| 467 | # verbosity is passed through from the processor
|
---|
| 468 | $self->{'verbosity'} = $verbosity;
|
---|
| 469 |
|
---|
[2785] | 470 | # as are the outhandle and failhandle
|
---|
[1424] | 471 | $self->{'outhandle'} = $outhandle if defined $outhandle;
|
---|
[2785] | 472 | $self->{'failhandle'} = $failhandle;
|
---|
[1424] | 473 |
|
---|
[1242] | 474 | # set process_exp and block_exp to defaults unless they were
|
---|
| 475 | # explicitly set
|
---|
[1244] | 476 |
|
---|
| 477 | if ((!$self->is_recursive()) and
|
---|
[1242] | 478 | (!defined $self->{'process_exp'}) || ($self->{'process_exp'} eq "")) {
|
---|
[1244] | 479 |
|
---|
[1242] | 480 | $self->{'process_exp'} = $self->get_default_process_exp ();
|
---|
| 481 | if ($self->{'process_exp'} eq "") {
|
---|
[1244] | 482 | warn ref($self) . " Warning: Non-recursive plugin has no process_exp\n";
|
---|
[1242] | 483 | }
|
---|
| 484 | }
|
---|
| 485 |
|
---|
| 486 | if ((!defined $self->{'block_exp'}) || ($self->{'block_exp'} eq "")) {
|
---|
| 487 | $self->{'block_exp'} = $self->get_default_block_exp ();
|
---|
| 488 | }
|
---|
[11089] | 489 |
|
---|
[1242] | 490 | }
|
---|
| 491 |
|
---|
[839] | 492 | sub begin {
|
---|
| 493 | my $self = shift (@_);
|
---|
| 494 | my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
|
---|
[1396] | 495 | $self->initialise_extractors();
|
---|
[839] | 496 | }
|
---|
| 497 |
|
---|
| 498 | sub end {
|
---|
[10155] | 499 | # potentially called at the end of each plugin pass
|
---|
| 500 | # import.pl only has one plugin pass, but buildcol.pl has multiple ones
|
---|
| 501 |
|
---|
[839] | 502 | my ($self) = @_;
|
---|
[1396] | 503 | $self->finalise_extractors();
|
---|
[839] | 504 | }
|
---|
| 505 |
|
---|
[10155] | 506 | sub deinit {
|
---|
| 507 | # called only once, after all plugin passes have been done
|
---|
| 508 |
|
---|
| 509 | my ($self) = @_;
|
---|
| 510 | }
|
---|
| 511 |
|
---|
[1242] | 512 | # this function should be overridden to return 1
|
---|
| 513 | # in recursive plugins
|
---|
[4] | 514 | sub is_recursive {
|
---|
| 515 | my $self = shift (@_);
|
---|
| 516 |
|
---|
[1242] | 517 | return 0;
|
---|
[4] | 518 | }
|
---|
| 519 |
|
---|
[1242] | 520 | sub get_default_block_exp {
|
---|
| 521 | my $self = shift (@_);
|
---|
| 522 |
|
---|
| 523 | return "";
|
---|
| 524 | }
|
---|
| 525 |
|
---|
| 526 | sub get_default_process_exp {
|
---|
| 527 | my $self = shift (@_);
|
---|
| 528 |
|
---|
| 529 | return "";
|
---|
| 530 | }
|
---|
| 531 |
|
---|
[9067] | 532 | # default implementation is to do nothing.
|
---|
| 533 | sub store_block_files
|
---|
| 534 | {
|
---|
| 535 | my $self =shift (@_);
|
---|
| 536 | my ($filename) = @_;
|
---|
| 537 | return;
|
---|
| 538 | }
|
---|
| 539 |
|
---|
| 540 | #default implementation is to block a file with same name as this, but extension jpg or JPG, if cover_images is on.
|
---|
| 541 | sub block_cover_image
|
---|
| 542 | {
|
---|
[10833] | 543 | my $self =shift;
|
---|
| 544 | my $filename = shift;
|
---|
| 545 |
|
---|
[9067] | 546 | if ($self->{'cover_image'}) {
|
---|
| 547 | my $coverfile = $filename;
|
---|
| 548 | $coverfile =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
| 549 | if (!-e $coverfile) {
|
---|
| 550 | $coverfile =~ s/jpg$/JPG/;
|
---|
| 551 | }
|
---|
| 552 | if (-e $coverfile) {
|
---|
| 553 | $self->{'file_blocks'}->{$coverfile} = 1;
|
---|
[11089] | 554 | }
|
---|
[9067] | 555 | }
|
---|
| 556 |
|
---|
| 557 | return;
|
---|
| 558 | }
|
---|
[11122] | 559 |
|
---|
| 560 | sub root_ext_split
|
---|
| 561 | {
|
---|
| 562 | my $self = shift (@_);
|
---|
| 563 | my ($filename,$tail_re) = @_;
|
---|
[9067] | 564 |
|
---|
[11122] | 565 | my ($file_prefix,$file_ext) = ($filename =~ m/^(.*?)($tail_re)$/);
|
---|
| 566 |
|
---|
| 567 | if ((!defined $file_prefix) || (!defined $file_ext)) {
|
---|
| 568 | ($file_prefix,$file_ext) = ($filename =~ m/^(.*)(\..*?)$/);
|
---|
| 569 | }
|
---|
| 570 |
|
---|
| 571 | return ($file_prefix,$file_ext);
|
---|
| 572 | }
|
---|
| 573 |
|
---|
[8510] | 574 | sub metadata_read {
|
---|
| 575 | my $self = shift (@_);
|
---|
| 576 | my ($pluginfo, $base_dir, $file, $metadata, $extrametakeys, $extrametadata, $processor, $maxdocs, $gli) = @_;
|
---|
[8892] | 577 | # Keep track of filenames with same root but different extensions
|
---|
[11122] | 578 | # Used to support -associate_ext and the more generalised
|
---|
| 579 | # -associate_tail_re
|
---|
[8892] | 580 |
|
---|
[11122] | 581 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 582 | if ((defined $associate_tail_re) && ($associate_tail_re ne "")) {
|
---|
[8892] | 583 |
|
---|
[11122] | 584 | my ($file_prefix,$file_ext)
|
---|
| 585 | = $self->root_ext_split($file,$associate_tail_re);
|
---|
| 586 |
|
---|
[8892] | 587 | if ((defined $file_prefix) && (defined $file_ext)) {
|
---|
[11122] | 588 |
|
---|
[8892] | 589 | my $shared_fileroot = $self->{'shared_fileroot'};
|
---|
| 590 | if (!defined $shared_fileroot->{$file_prefix}) {
|
---|
[11122] | 591 | my $file_prefix_rec = { 'tie_to' => undef,
|
---|
| 592 | 'exts' => {} };
|
---|
[8892] | 593 | $shared_fileroot->{$file_prefix} = $file_prefix_rec;
|
---|
| 594 | }
|
---|
| 595 |
|
---|
| 596 | my $file_prefix_rec = $shared_fileroot->{$file_prefix};
|
---|
| 597 |
|
---|
| 598 | my $process_exp = $self->{'process_exp'};
|
---|
| 599 |
|
---|
[11122] | 600 | if ($file =~ m/$process_exp/) {
|
---|
[8892] | 601 | # This is the document the others should be tied to
|
---|
| 602 | $file_prefix_rec->{'tie_to'} = $file_ext;
|
---|
| 603 | }
|
---|
| 604 | else {
|
---|
[11122] | 605 | if ($file_ext =~ m/$associate_tail_re$/) {
|
---|
[9351] | 606 | $file_prefix_rec->{'exts'}->{$file_ext} = 1;
|
---|
| 607 | }
|
---|
[8892] | 608 | }
|
---|
[11122] | 609 |
|
---|
[8892] | 610 | }
|
---|
| 611 | }
|
---|
[11122] | 612 |
|
---|
[9067] | 613 | # now check whether we are actually processing this
|
---|
| 614 | my $filename = $file;
|
---|
| 615 | $filename = &util::filename_cat ($base_dir, $file) if $base_dir =~ /\w/;
|
---|
| 616 | if ($self->{'process_exp'} eq "" || $filename !~ /$self->{'process_exp'}/ || !-f $filename) {
|
---|
| 617 | return undef; # can't recognise
|
---|
| 618 | }
|
---|
[8892] | 619 |
|
---|
[9067] | 620 | # do smart blocking if appropriate
|
---|
[11089] | 621 | if ($self->{'smart_block'}) {
|
---|
[9067] | 622 | $self->store_block_files($filename);
|
---|
| 623 | }
|
---|
[11089] | 624 | # block the cover image if there is one
|
---|
| 625 | if ($self->{'cover_image'}) {
|
---|
| 626 | $self->block_cover_image($filename);
|
---|
| 627 | }
|
---|
[9067] | 628 |
|
---|
| 629 | return 1;
|
---|
[8510] | 630 | }
|
---|
| 631 |
|
---|
[8892] | 632 | sub tie_to_filename
|
---|
| 633 | {
|
---|
| 634 | my $self = shift (@_);
|
---|
| 635 |
|
---|
| 636 | my ($file_ext,$file_prefix_rec) = @_;
|
---|
| 637 |
|
---|
| 638 | if (defined $file_prefix_rec) {
|
---|
| 639 | my $tie_to = $file_prefix_rec->{'tie_to'};
|
---|
[9351] | 640 |
|
---|
[8892] | 641 | if (defined $tie_to) {
|
---|
| 642 | if ($tie_to eq $file_ext) {
|
---|
| 643 | return 1;
|
---|
| 644 | }
|
---|
| 645 | }
|
---|
| 646 | }
|
---|
| 647 |
|
---|
| 648 | return 0;
|
---|
| 649 | }
|
---|
| 650 |
|
---|
| 651 | sub tie_to_assoc_file
|
---|
| 652 | {
|
---|
| 653 | my $self = shift (@_);
|
---|
| 654 | my ($file_ext,$file_prefix_rec) = @_;
|
---|
| 655 |
|
---|
| 656 | if (defined $file_prefix_rec) {
|
---|
| 657 | my $tie_to = $file_prefix_rec->{'tie_to'};
|
---|
| 658 | if (defined $tie_to) {
|
---|
| 659 |
|
---|
| 660 | my $exts = $file_prefix_rec->{'exts'};
|
---|
| 661 |
|
---|
| 662 | my $has_file_ext = $exts->{$file_ext};
|
---|
[11122] | 663 |
|
---|
[8892] | 664 | if ($has_file_ext) {
|
---|
| 665 | return 1;
|
---|
| 666 | }
|
---|
| 667 | }
|
---|
| 668 | }
|
---|
| 669 |
|
---|
| 670 | return 0;
|
---|
| 671 | }
|
---|
| 672 |
|
---|
| 673 |
|
---|
| 674 | sub associate_with
|
---|
| 675 | {
|
---|
| 676 | my $self = shift (@_);
|
---|
| 677 | my ($file, $filename, $metadata) = @_;
|
---|
| 678 |
|
---|
[11122] | 679 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 680 | return 0 if (!$associate_tail_re);
|
---|
[8892] | 681 |
|
---|
| 682 | # If file, see if matches with "tie_to" doc or is one of the
|
---|
| 683 | # associated filename extensions.
|
---|
| 684 |
|
---|
[11122] | 685 | my ($file_prefix,$file_ext) = $self->root_ext_split($file,$associate_tail_re);
|
---|
| 686 |
|
---|
[8892] | 687 | if ((defined $file_prefix) && (defined $file_ext)) {
|
---|
| 688 |
|
---|
| 689 | my $file_prefix_rec = $self->{'shared_fileroot'}->{$file_prefix};
|
---|
[9351] | 690 |
|
---|
[8892] | 691 | if ($self->tie_to_filename($file_ext,$file_prefix_rec)) {
|
---|
| 692 |
|
---|
| 693 | # Set up gsdlassocfile_tobe
|
---|
| 694 |
|
---|
| 695 | my $exts = $file_prefix_rec->{'exts'};
|
---|
| 696 |
|
---|
| 697 | if (!defined $metadata->{'gsdlassocfile_tobe'}) {
|
---|
| 698 | $metadata->{'gsdlassocfile_tobe'} = [];
|
---|
| 699 | }
|
---|
| 700 |
|
---|
| 701 | my $assoc_tobe = $metadata->{'gsdlassocfile_tobe'};
|
---|
| 702 |
|
---|
| 703 | my ($full_prefix) = ($filename =~ m/^(.*)\..*?$/);
|
---|
| 704 | foreach my $e (keys %$exts) {
|
---|
[11122] | 705 | my $assoc_file = "$full_prefix$e";
|
---|
| 706 | print STDERR " $self->{'plugin_type'}: Associating $file_prefix$e with $file_prefix_rec->{'tie_to'} version\n";
|
---|
[8892] | 707 | my $mime_type = ""; # let system auto detect this
|
---|
| 708 | push(@$assoc_tobe,"$assoc_file:$mime_type:");
|
---|
| 709 | }
|
---|
[11122] | 710 |
|
---|
[8892] | 711 | }
|
---|
| 712 | elsif ($self->tie_to_assoc_file($file_ext,$file_prefix_rec)) {
|
---|
[11122] | 713 |
|
---|
| 714 |
|
---|
| 715 | # a form of smart block
|
---|
[8892] | 716 | return 1;
|
---|
| 717 | }
|
---|
| 718 | }
|
---|
| 719 |
|
---|
| 720 | return 0;
|
---|
| 721 | }
|
---|
| 722 |
|
---|
| 723 |
|
---|
[10280] | 724 | sub read_block {
|
---|
[1954] | 725 | my $self = shift (@_);
|
---|
| 726 |
|
---|
[9853] | 727 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
[4] | 728 |
|
---|
[8908] | 729 |
|
---|
[2795] | 730 | my $filename = $file;
|
---|
| 731 | $filename = &util::filename_cat ($base_dir, $file) if $base_dir =~ /\w/;
|
---|
| 732 |
|
---|
[8892] | 733 | if ($self->associate_with($file,$filename,$metadata)) {
|
---|
| 734 | # a form of smart block
|
---|
| 735 | $self->{'num_blocked'} ++;
|
---|
[10280] | 736 | return (0,undef); # blocked
|
---|
[8892] | 737 | }
|
---|
[8510] | 738 |
|
---|
[10280] | 739 | my $smart_block = $self->{'smart_block'};
|
---|
| 740 | my $smart_block_BN = $self->{'smart_block_BN'};
|
---|
[11089] | 741 |
|
---|
[8915] | 742 | if ($smart_block || $smart_block_BN) {
|
---|
[8510] | 743 | if (defined $self->{'file_blocks'}->{$filename} && $self->{'file_blocks'}->{$filename} == 1){
|
---|
[8892] | 744 | $self->{'num_blocked'} ++;
|
---|
[10280] | 745 | return (0,undef); # blocked
|
---|
[8510] | 746 | }
|
---|
[11089] | 747 | } else {
|
---|
| 748 | if ($self->{'block_exp'} ne "" && $filename =~ /$self->{'block_exp'}/) {
|
---|
| 749 | $self->{'num_blocked'} ++;
|
---|
| 750 | return (0,undef); # blocked
|
---|
| 751 | }
|
---|
| 752 | if ($self->{'cover_image'}) {
|
---|
| 753 | if (defined $self->{'file_blocks'}->{$filename} && $self->{'file_blocks'}->{$filename} == 1){
|
---|
| 754 | $self->{'num_blocked'} ++;
|
---|
| 755 | return (0,undef); # blocked
|
---|
| 756 | }
|
---|
| 757 | }
|
---|
[2785] | 758 | }
|
---|
[11122] | 759 |
|
---|
[1242] | 760 | if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) {
|
---|
[10280] | 761 | return (undef,undef); # can't recognise
|
---|
[1242] | 762 | }
|
---|
[10280] | 763 |
|
---|
| 764 | return (1,$filename);
|
---|
| 765 | }
|
---|
| 766 |
|
---|
| 767 | sub read_tidy_file {
|
---|
| 768 |
|
---|
| 769 | my $self = shift (@_);
|
---|
| 770 |
|
---|
| 771 | my ($file) = @_;
|
---|
| 772 |
|
---|
[1242] | 773 | $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
|
---|
[1844] | 774 |
|
---|
[10280] | 775 | return $file;
|
---|
| 776 | }
|
---|
| 777 |
|
---|
| 778 |
|
---|
| 779 |
|
---|
| 780 | # The BasPlug read_into_doc_obj() function. This function does all the
|
---|
| 781 | # right things to make general options work for a given plugin. It reads in
|
---|
| 782 | # a file and sets up a slew of metadata all saved in doc_obj, which
|
---|
| 783 | # it then returns as part of a tuple (process_status,doc_obj)
|
---|
| 784 | #
|
---|
| 785 | # Much of this functionality used to reside in read, but it was broken
|
---|
| 786 | # down into a supporting routine to make the code more flexible.
|
---|
| 787 | #
|
---|
| 788 | # recursive plugins (e.g. RecPlug) and specialized plugins like those
|
---|
| 789 | # capable of processing many documents within a single file (e.g.
|
---|
| 790 | # GMLPlug) will normally want to implement their own version of
|
---|
| 791 | # read_into_doc_obj()
|
---|
| 792 | #
|
---|
| 793 | # Note that $base_dir might be "" and that $file might
|
---|
| 794 | # include directories
|
---|
| 795 | sub read_into_doc_obj {
|
---|
| 796 | my $self = shift (@_);
|
---|
| 797 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
| 798 |
|
---|
| 799 | if ($self->is_recursive()) {
|
---|
| 800 | gsprintf(STDERR, "{BasPlug.read_must_be_implemented}") && die "\n";
|
---|
| 801 | }
|
---|
| 802 |
|
---|
| 803 | my $outhandle = $self->{'outhandle'};
|
---|
| 804 |
|
---|
| 805 | my ($block_status,$filename) = $self->read_block(@_);
|
---|
| 806 | return $block_status if ((!defined $block_status) || ($block_status==0));
|
---|
| 807 | $file = $self->read_tidy_file($file);
|
---|
| 808 |
|
---|
[2811] | 809 | # Do encoding stuff
|
---|
| 810 | my ($language, $encoding) = $self->textcat_get_language_encoding ($filename);
|
---|
[11389] | 811 | if ($self->{'verbosity'} > 2) {
|
---|
| 812 | print $outhandle "BasPlug: reading $file as ($encoding,$language)\n";
|
---|
| 813 | }
|
---|
[1844] | 814 |
|
---|
[1242] | 815 | # create a new document
|
---|
[1379] | 816 | my $doc_obj = new doc ($filename, "indexed_doc");
|
---|
[12270] | 817 | $doc_obj->set_OIDtype ($processor->{'OIDtype'}, $processor->{'OIDmetadata'});
|
---|
[1844] | 818 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Language", $language);
|
---|
[1868] | 819 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Encoding", $encoding);
|
---|
[7508] | 820 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Plugin", "$self->{'plugin_type'}");
|
---|
[8166] | 821 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "FileSize", (-s $filename));
|
---|
| 822 |
|
---|
[2796] | 823 | my ($filemeta) = $file =~ /([^\\\/]+)$/;
|
---|
[4845] | 824 | # how do we know what encoding the filename is in?
|
---|
[11389] | 825 | # assume it is in the same encoding as its contents
|
---|
| 826 | if ($encoding !~ /(?:ascii|utf8|unicode)/) {
|
---|
| 827 | $filemeta = unicode::unicode2utf8(
|
---|
| 828 | unicode::convert2unicode($encoding, \$filemeta)
|
---|
| 829 | );
|
---|
| 830 | }
|
---|
| 831 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Source", &ghtml::dmsafe($filemeta));
|
---|
[2816] | 832 | if ($self->{'cover_image'}) {
|
---|
| 833 | $self->associate_cover_image($doc_obj, $filename);
|
---|
| 834 | }
|
---|
[1242] | 835 |
|
---|
| 836 | # read in file ($text will be in utf8)
|
---|
| 837 | my $text = "";
|
---|
[2734] | 838 | $self->read_file ($filename, $encoding, $language, \$text);
|
---|
[1242] | 839 |
|
---|
[1844] | 840 | if (!length ($text)) {
|
---|
[2811] | 841 | my $plugin_name = ref ($self);
|
---|
[9586] | 842 | if ($gli) {
|
---|
| 843 | print STDERR "<ProcessingError n='$file' r='File contains no text'>\n";
|
---|
| 844 | }
|
---|
[9703] | 845 | gsprintf($outhandle, "$plugin_name: {BasPlug.file_has_no_text}\n", $filename) if $self->{'verbosity'};
|
---|
[2785] | 846 |
|
---|
| 847 | my $failhandle = $self->{'failhandle'};
|
---|
[9413] | 848 | gsprintf($failhandle, "$file: " . ref($self) . ": {BasPlug.empty_file}\n");
|
---|
[5681] | 849 | # print $failhandle "$file: " . ref($self) . ": file contains no text\n";
|
---|
[2785] | 850 | $self->{'num_not_processed'} ++;
|
---|
| 851 |
|
---|
[10280] | 852 | return (0,undef); # what should we return here?? error but don't want to pass it on
|
---|
[1242] | 853 | }
|
---|
[1954] | 854 |
|
---|
[1242] | 855 | # include any metadata passed in from previous plugins
|
---|
| 856 | # note that this metadata is associated with the top level section
|
---|
[8510] | 857 |
|
---|
[11122] | 858 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 859 |
|
---|
[1242] | 860 | $self->extra_metadata ($doc_obj, $doc_obj->get_top_section(), $metadata);
|
---|
[9398] | 861 |
|
---|
[8716] | 862 | # do plugin specific processing of doc_obj
|
---|
| 863 | unless (defined ($self->process (\$text, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli))) {
|
---|
| 864 | $text = '';
|
---|
| 865 | undef $text;
|
---|
[9584] | 866 | print STDERR "<ProcessingError n='$file'>\n" if ($gli);
|
---|
[10280] | 867 | return (-1,undef);
|
---|
[8716] | 868 | }
|
---|
| 869 | $text='';
|
---|
| 870 | undef $text;
|
---|
[9398] | 871 |
|
---|
[1242] | 872 | # do any automatic metadata extraction
|
---|
| 873 | $self->auto_extract_metadata ($doc_obj);
|
---|
[1954] | 874 |
|
---|
[1242] | 875 | # add an OID
|
---|
[3515] | 876 | # see if there is a plugin-specific set_OID function...
|
---|
[8716] | 877 | if (defined ($self->can('set_OID'))) {
|
---|
[3515] | 878 | # it will need $doc_obj to set the Identifier metadata...
|
---|
| 879 | $self->set_OID($doc_obj);
|
---|
| 880 | } else {
|
---|
| 881 | # use the default set_OID() in doc.pm
|
---|
| 882 | $doc_obj->set_OID();
|
---|
| 883 | }
|
---|
[10280] | 884 |
|
---|
| 885 | return (1,$doc_obj);
|
---|
| 886 | }
|
---|
[1242] | 887 |
|
---|
| 888 |
|
---|
[10280] | 889 | # The BasPlug read() function. This function calls read_into_doc_obj()
|
---|
| 890 | # to ensure all the right things to make general options work for a
|
---|
| 891 | # given plugin are done. It then calls the process() function which
|
---|
| 892 | # does all the work specific to a plugin (like the old read functions
|
---|
| 893 | # used to do). Most plugins should define their own process() function
|
---|
| 894 | # and let this read() function keep control.
|
---|
| 895 | #
|
---|
| 896 | # recursive plugins (e.g. RecPlug) and specialized plugins like those
|
---|
| 897 | # capable of processing many documents within a single file (e.g.
|
---|
| 898 | # GMLPlug) might want to implement their own version of read(), but
|
---|
| 899 | # more likely need to implement their own version of read_into_doc_obj()
|
---|
| 900 | #
|
---|
| 901 | # Return number of files processed, undef if can't recognise, -1 if can't
|
---|
| 902 | # process
|
---|
| 903 |
|
---|
| 904 | sub read {
|
---|
| 905 | my $self = shift (@_);
|
---|
| 906 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
| 907 |
|
---|
| 908 | my ($process_status,$doc_obj) = $self->read_into_doc_obj(@_);
|
---|
| 909 |
|
---|
| 910 | if ((defined $process_status) && ($process_status == 1)) {
|
---|
| 911 | # process the document
|
---|
| 912 | $processor->process($doc_obj);
|
---|
| 913 |
|
---|
| 914 | if(defined($self->{'places_filename'})){
|
---|
| 915 | &util::rm($self->{'places_filename'});
|
---|
| 916 | $self->{'places_filename'} = undef;
|
---|
| 917 | }
|
---|
| 918 |
|
---|
| 919 | $self->{'num_processed'} ++;
|
---|
| 920 | undef $doc_obj;
|
---|
[9398] | 921 | }
|
---|
| 922 |
|
---|
[10280] | 923 | # if process_status == 1, then the file has been processed.
|
---|
| 924 | return $process_status;
|
---|
| 925 |
|
---|
[4] | 926 | }
|
---|
| 927 |
|
---|
[1244] | 928 | # returns undef if file is rejected by the plugin
|
---|
[1242] | 929 | sub process {
|
---|
| 930 | my $self = shift (@_);
|
---|
[11089] | 931 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
---|
[1242] | 932 |
|
---|
[9413] | 933 | gsprintf(STDERR, "BasPlug::process {common.must_be_implemented}\n") && die "\n";
|
---|
[5681] | 934 | # die "Basplug::process function must be implemented in sub-class\n";
|
---|
[1244] | 935 |
|
---|
| 936 | return undef; # never gets here
|
---|
[1242] | 937 | }
|
---|
| 938 |
|
---|
[1219] | 939 | # uses the multiread package to read in the entire file pointed to
|
---|
| 940 | # by filename and loads the resulting text into $$textref. Input text
|
---|
| 941 | # may be in any of the encodings handled by multiread, output text
|
---|
| 942 | # will be in utf8
|
---|
| 943 | sub read_file {
|
---|
| 944 | my $self = shift (@_);
|
---|
[2734] | 945 | my ($filename, $encoding, $language, $textref) = @_;
|
---|
[4] | 946 |
|
---|
[1756] | 947 | if (!-r $filename)
|
---|
| 948 | {
|
---|
[1844] | 949 | my $outhandle = $self->{'outhandle'};
|
---|
[9413] | 950 | gsprintf($outhandle, "{BasPlug.read_denied}\n", $filename) if $self->{'verbosity'};
|
---|
[5681] | 951 | # print $outhandle "Read permission denied for $filename\n" if $self->{'verbosity'};
|
---|
[1756] | 952 | return;
|
---|
| 953 | }
|
---|
[1219] | 954 | $$textref = "";
|
---|
[9413] | 955 | if (!open (FILE, $filename)) {
|
---|
| 956 | gsprintf(STDERR, "BasPlug::read_file {BasPlug.could_not_open_for_reading} ($!)\n", $filename);
|
---|
[10280] | 957 | die "\n";
|
---|
| 958 | }
|
---|
| 959 |
|
---|
[1844] | 960 | if ($encoding eq "ascii") {
|
---|
[1219] | 961 | undef $/;
|
---|
| 962 | $$textref = <FILE>;
|
---|
| 963 | $/ = "\n";
|
---|
| 964 | } else {
|
---|
| 965 | my $reader = new multiread();
|
---|
| 966 | $reader->set_handle ('BasPlug::FILE');
|
---|
[1844] | 967 | $reader->set_encoding ($encoding);
|
---|
[1219] | 968 | $reader->read_file ($textref);
|
---|
[10280] | 969 | #Now segments chinese if the separate_cjk option is set
|
---|
[6584] | 970 | if ($self->{'separate_cjk'}) {
|
---|
[1219] | 971 | # segment the Chinese words
|
---|
| 972 | $$textref = &cnseg::segment($$textref);
|
---|
| 973 | }
|
---|
| 974 | }
|
---|
[10280] | 975 | close FILE;
|
---|
| 976 | }
|
---|
[1219] | 977 |
|
---|
[10280] | 978 | # write_file -- used by ConvertToPlug, for example in post processing
|
---|
| 979 | #
|
---|
| 980 | sub utf8_write_file {
|
---|
| 981 | my $self = shift (@_);
|
---|
| 982 | my ($textref, $filename) = @_;
|
---|
| 983 |
|
---|
| 984 | if (!open (FILE, ">$filename")) {
|
---|
| 985 | gsprintf(STDERR, "ConvertToPlug::write_file {ConvertToPlug.could_not_open_for_writing} ($!)\n", $filename);
|
---|
| 986 | die "\n";
|
---|
| 987 | }
|
---|
| 988 | print FILE $$textref;
|
---|
| 989 |
|
---|
[1219] | 990 | close FILE;
|
---|
| 991 | }
|
---|
| 992 |
|
---|
[10280] | 993 |
|
---|
[7504] | 994 | sub filename_based_title
|
---|
| 995 | {
|
---|
| 996 | my $self = shift (@_);
|
---|
| 997 | my ($file) = @_;
|
---|
| 998 |
|
---|
| 999 | my $file_derived_title = $file;
|
---|
| 1000 | $file_derived_title =~ s/_/ /g;
|
---|
| 1001 | $file_derived_title =~ s/\..*?$//;
|
---|
| 1002 |
|
---|
| 1003 | return $file_derived_title;
|
---|
| 1004 | }
|
---|
| 1005 |
|
---|
[9398] | 1006 |
|
---|
[7504] | 1007 | sub title_fallback
|
---|
| 1008 | {
|
---|
| 1009 | my $self = shift (@_);
|
---|
| 1010 | my ($doc_obj,$section,$file) = @_;
|
---|
| 1011 |
|
---|
| 1012 | if (!defined $doc_obj->get_metadata_element ($section, "Title")) {
|
---|
| 1013 |
|
---|
| 1014 | my $file_derived_title = $self->filename_based_title($file);
|
---|
| 1015 | $doc_obj->add_metadata ($section, "Title", $file_derived_title);
|
---|
| 1016 | }
|
---|
| 1017 | }
|
---|
| 1018 |
|
---|
[2811] | 1019 | sub textcat_get_language_encoding {
|
---|
| 1020 | my $self = shift (@_);
|
---|
| 1021 | my ($filename) = @_;
|
---|
| 1022 |
|
---|
[10280] | 1023 |
|
---|
[2811] | 1024 | my ($language, $encoding, $extracted_encoding);
|
---|
| 1025 | if ($self->{'input_encoding'} eq "auto") {
|
---|
| 1026 | # use textcat to automatically work out the input encoding and language
|
---|
| 1027 | ($language, $encoding) = $self->get_language_encoding ($filename);
|
---|
| 1028 | } elsif ($self->{'extract_language'}) {
|
---|
| 1029 | # use textcat to get language metadata
|
---|
| 1030 | ($language, $extracted_encoding) = $self->get_language_encoding ($filename);
|
---|
| 1031 | $encoding = $self->{'input_encoding'};
|
---|
[7644] | 1032 | # don't print this message for english... english in utf8 is identical
|
---|
| 1033 | # to english in iso-8859-1 (except for some punctuation). We don't have
|
---|
| 1034 | # a language model for en_utf8, so textcat always says iso-8859-1!
|
---|
| 1035 | if ($extracted_encoding ne $encoding && $language ne "en"
|
---|
| 1036 | && $self->{'verbosity'}) {
|
---|
[2811] | 1037 | my $plugin_name = ref ($self);
|
---|
| 1038 | my $outhandle = $self->{'outhandle'};
|
---|
[9413] | 1039 | gsprintf($outhandle, "$plugin_name: {BasPlug.wrong_encoding}\n", $filename, $encoding, $extracted_encoding);
|
---|
[2811] | 1040 | }
|
---|
| 1041 | } else {
|
---|
| 1042 | $language = $self->{'default_language'};
|
---|
| 1043 | $encoding = $self->{'input_encoding'};
|
---|
| 1044 | }
|
---|
[10280] | 1045 |
|
---|
[2811] | 1046 | return ($language, $encoding);
|
---|
| 1047 | }
|
---|
| 1048 |
|
---|
[1844] | 1049 | # Uses textcat to work out the encoding and language of the text in
|
---|
| 1050 | # $filename. All html tags are removed before processing.
|
---|
| 1051 | # returns an array containing "language" and "encoding"
|
---|
| 1052 | sub get_language_encoding {
|
---|
| 1053 | my $self = shift (@_);
|
---|
| 1054 | my ($filename) = @_;
|
---|
| 1055 | my $outhandle = $self->{'outhandle'};
|
---|
[9413] | 1056 | my $unicode_format = "";
|
---|
[11389] | 1057 | my $best_language = "";
|
---|
| 1058 | my $best_encoding = "";
|
---|
| 1059 |
|
---|
[1844] | 1060 | # read in file
|
---|
[11389] | 1061 | if (!open (FILE, $filename)) {
|
---|
| 1062 | gsprintf(STDERR, "BasPlug::get_language_encoding {BasPlug.could_not_open_for_reading} ($!)\n", $filename);
|
---|
| 1063 | # this is a pretty bad error, but try to continue anyway
|
---|
| 1064 | return ($self->{'default_language'}, $self->{'input_encoding'});
|
---|
| 1065 | }
|
---|
[1844] | 1066 | undef $/;
|
---|
| 1067 | my $text = <FILE>;
|
---|
| 1068 | $/ = "\n";
|
---|
| 1069 | close FILE;
|
---|
| 1070 |
|
---|
[9413] | 1071 | # check if first few bytes have a Byte Order Marker
|
---|
| 1072 | my $bom=substr($text,0,2); # check 16bit unicode
|
---|
| 1073 | if ($bom eq "\xff\xfe") { # little endian 16bit unicode
|
---|
| 1074 | $unicode_format="unicode";
|
---|
| 1075 | } elsif ($bom eq "\xfe\xff") { # big endian 16bit unicode
|
---|
| 1076 | $unicode_format="unicode";
|
---|
| 1077 | } else {
|
---|
| 1078 | $bom=substr($text,0,3); # check utf-8
|
---|
| 1079 | if ($bom eq "\xef\xbb\xbf") { # utf-8 coded FEFF bom
|
---|
| 1080 | $unicode_format="utf8";
|
---|
| 1081 | # } elsif ($bom eq "\xef\xbf\xbe") { # utf-8 coded FFFE bom. Error!?
|
---|
| 1082 | # $unicode_format="utf8";
|
---|
| 1083 | }
|
---|
| 1084 | }
|
---|
[10442] | 1085 |
|
---|
[1999] | 1086 |
|
---|
[11389] | 1087 | # handle html files specially
|
---|
[9413] | 1088 | # XXX this doesn't match plugins derived from HTMLPlug (except ConvertTo)
|
---|
| 1089 | if (ref($self) eq 'HTMLPlug' ||
|
---|
| 1090 | (exists $self->{'converted_to'} && $self->{'converted_to'} eq 'HTML')){
|
---|
[11389] | 1091 |
|
---|
| 1092 | # remove <title>stuff</title> -- as titles tend often to be in English
|
---|
| 1093 | # for foreign language documents
|
---|
| 1094 | $text =~ s!<title>.*?</title>!!si;
|
---|
| 1095 |
|
---|
| 1096 | # see if this html file specifies its encoding
|
---|
| 1097 | if ($text =~ /^<\?xml.*encoding="(.+?)"/) {
|
---|
| 1098 | $best_encoding = $1;
|
---|
[11880] | 1099 | } elsif ($text =~ /<meta http-equiv.*content-type.*charset=(.+?)"/i) {#"
|
---|
[11389] | 1100 | $best_encoding = $1;
|
---|
| 1101 | }
|
---|
| 1102 | if ($best_encoding) { # we extracted an encoding
|
---|
| 1103 | $best_encoding =~ s/-+/_/g;
|
---|
| 1104 | $best_encoding = lc($best_encoding); # lowercase
|
---|
| 1105 | if ($best_encoding eq "utf_8") { $best_encoding = "utf8" }
|
---|
| 1106 | $self->{'input_encoding'} = $best_encoding;
|
---|
| 1107 | }
|
---|
| 1108 |
|
---|
| 1109 | # remove all HTML tags
|
---|
[9413] | 1110 | $text =~ s/<[^>]*>//sg;
|
---|
| 1111 | }
|
---|
[1844] | 1112 |
|
---|
| 1113 | # get the language/encoding
|
---|
[11966] | 1114 | $self->{'textcat'} = new textcat() if (!defined($self->{'textcat'}));
|
---|
[2235] | 1115 | my $results = $self->{'textcat'}->classify(\$text);
|
---|
[1844] | 1116 |
|
---|
[1903] | 1117 | # if textcat returns 3 or less possibilities we'll use the
|
---|
| 1118 | # first one in the list - otherwise use the defaults
|
---|
[2235] | 1119 | if (scalar @$results > 3) {
|
---|
[9413] | 1120 | if ($unicode_format) { # in case the first had a BOM
|
---|
| 1121 | $best_encoding=$unicode_format;
|
---|
| 1122 | } else {
|
---|
| 1123 | my %guessed_encodings = ();
|
---|
| 1124 | foreach my $result (@$results) {
|
---|
| 1125 | $result =~ /([^\-]+)$/;
|
---|
| 1126 | my $enc=$1;
|
---|
| 1127 | if (!defined($guessed_encodings{$enc})) {
|
---|
| 1128 | $guessed_encodings{$enc}=0;
|
---|
| 1129 | }
|
---|
| 1130 | $guessed_encodings{$enc}++;
|
---|
[3731] | 1131 | }
|
---|
[9413] | 1132 |
|
---|
| 1133 | $guessed_encodings{""}=-1; # for default best_encoding of ""
|
---|
| 1134 | foreach my $enc (keys %guessed_encodings) {
|
---|
| 1135 | if ($guessed_encodings{$enc} >
|
---|
| 1136 | $guessed_encodings{$best_encoding}){
|
---|
| 1137 | $best_encoding=$enc;
|
---|
| 1138 | }
|
---|
| 1139 | }
|
---|
[3731] | 1140 | }
|
---|
| 1141 |
|
---|
[1844] | 1142 | if ($self->{'input_encoding'} ne 'auto') {
|
---|
[9961] | 1143 | if ($self->{'extract_language'} && ($self->{'verbosity'}>2)) {
|
---|
[9413] | 1144 | gsprintf($outhandle,
|
---|
| 1145 | "BasPlug: {BasPlug.could_not_extract_language}\n",
|
---|
| 1146 | $filename, $self->{'default_language'});
|
---|
[1844] | 1147 | }
|
---|
[11389] | 1148 | $best_language = $self->{'default_language'};
|
---|
| 1149 | $best_encoding = $self->{'input_encoding'};
|
---|
[1844] | 1150 |
|
---|
| 1151 | } else {
|
---|
[9961] | 1152 | if ($self->{'verbosity'}>2) {
|
---|
[9413] | 1153 | gsprintf($outhandle,
|
---|
| 1154 | "BasPlug: {BasPlug.could_not_extract_language}\n",
|
---|
| 1155 | $filename, $self->{'default_language'});
|
---|
[1844] | 1156 | }
|
---|
[11389] | 1157 | $best_language = $self->{'default_language'};
|
---|
[1844] | 1158 | }
|
---|
[11389] | 1159 | } else { # <= 3 suggestions
|
---|
| 1160 | my ($language, $encoding) = $results->[0] =~ /^([^-]*)(?:-(.*))?$/;
|
---|
| 1161 | if (!defined $language) {
|
---|
| 1162 | if ($self->{'verbosity'}>2) {
|
---|
| 1163 | gsprintf($outhandle,
|
---|
| 1164 | "BasPlug: {BasPlug.could_not_extract_language}\n",
|
---|
| 1165 | $filename, $self->{'default_language'});
|
---|
| 1166 | }
|
---|
| 1167 | $language = $self->{'default_language'};
|
---|
[1870] | 1168 | }
|
---|
[11389] | 1169 | if (!defined $encoding) {
|
---|
| 1170 | if ($self->{'verbosity'}>2) {
|
---|
| 1171 | gsprintf($outhandle,
|
---|
| 1172 | "BasPlug: {BasPlug.could_not_extract_encoding}\n",
|
---|
| 1173 | $filename, $self->{'default_encoding'});
|
---|
| 1174 | }
|
---|
| 1175 | $encoding = $self->{'default_encoding'};
|
---|
[1870] | 1176 | }
|
---|
[11389] | 1177 | $best_language = $language;
|
---|
| 1178 | if (! $best_encoding ) { # may already be set... eg from html meta tag
|
---|
| 1179 | $best_encoding = $encoding;
|
---|
| 1180 | }
|
---|
[1844] | 1181 | }
|
---|
| 1182 |
|
---|
[11389] | 1183 | my $text_copy = $text;
|
---|
| 1184 | if ($best_encoding =~ /^iso_8859/ && unicode::ensure_utf8(\$text_copy)==0) {
|
---|
| 1185 | # the text is valid utf8, so assume that's the real encoding
|
---|
| 1186 | # (since textcat is based on probabilities)
|
---|
| 1187 | $best_encoding = 'utf8';
|
---|
| 1188 | }
|
---|
[7818] | 1189 |
|
---|
| 1190 | # check for equivalents where textcat doesn't have some encodings...
|
---|
| 1191 | # eg MS versions of standard encodings
|
---|
[11389] | 1192 | if ($best_encoding =~ /^iso_8859_(\d+)/) {
|
---|
[7818] | 1193 | my $iso = $1; # which variant of the iso standard?
|
---|
| 1194 | # iso-8859 sets don't use chars 0x80-0x9f, windows codepages do
|
---|
| 1195 | if ($text =~ /[\x80-\x9f]/) {
|
---|
| 1196 | # Western Europe
|
---|
[11389] | 1197 | if ($iso == 1 or $iso == 15) { $best_encoding = 'windows_1252' }
|
---|
| 1198 | elsif ($iso == 2) {$best_encoding = 'windows_1250'} # Central Europe
|
---|
| 1199 | elsif ($iso == 5) {$best_encoding = 'windows_1251'} # Cyrillic
|
---|
| 1200 | elsif ($iso == 6) {$best_encoding = 'windows_1256'} # Arabic
|
---|
| 1201 | elsif ($iso == 7) {$best_encoding = 'windows_1253'} # Greek
|
---|
| 1202 | elsif ($iso == 8) {$best_encoding = 'windows_1255'} # Hebrew
|
---|
| 1203 | elsif ($iso == 9) {$best_encoding = 'windows_1254'} # Turkish
|
---|
[7818] | 1204 | }
|
---|
| 1205 | }
|
---|
| 1206 |
|
---|
[11389] | 1207 | if ($best_encoding !~ /^(ascii|utf8|unicode)$/ &&
|
---|
| 1208 | !defined $encodings::encodings->{$best_encoding}) {
|
---|
[1844] | 1209 | if ($self->{'verbosity'}) {
|
---|
[7818] | 1210 | gsprintf($outhandle, "BasPlug: {BasPlug.unsupported_encoding}\n",
|
---|
[11389] | 1211 | $filename, $best_encoding, $self->{'default_encoding'});
|
---|
[1844] | 1212 | }
|
---|
[11389] | 1213 | $best_encoding = $self->{'default_encoding'};
|
---|
[1844] | 1214 | }
|
---|
| 1215 |
|
---|
[11389] | 1216 | return ($best_language, $best_encoding);
|
---|
[1844] | 1217 | }
|
---|
| 1218 |
|
---|
[1219] | 1219 | # add any extra metadata that's been passed around from one
|
---|
| 1220 | # plugin to another.
|
---|
| 1221 | # extra_metadata uses add_utf8_metadata so it expects metadata values
|
---|
| 1222 | # to already be in utf8
|
---|
| 1223 | sub extra_metadata {
|
---|
| 1224 | my $self = shift (@_);
|
---|
| 1225 | my ($doc_obj, $cursection, $metadata) = @_;
|
---|
| 1226 |
|
---|
[11122] | 1227 | my $associate_tail_re = $self->{'associate_tail_re'};
|
---|
| 1228 |
|
---|
[1219] | 1229 | foreach my $field (keys(%$metadata)) {
|
---|
[839] | 1230 | # $metadata->{$field} may be an array reference
|
---|
[8510] | 1231 | if ($field eq "gsdlassocfile_tobe") {
|
---|
| 1232 | # 'gsdlassocfile_tobe' is artificially introduced metadata
|
---|
| 1233 | # that is used to signal that certain additional files should
|
---|
| 1234 | # be tied to this document. Useful in situations where a
|
---|
| 1235 | # metadata pass in the plugin pipeline works out some files
|
---|
| 1236 | # need to be associated with a document, but the document hasn't
|
---|
| 1237 | # been formed yet.
|
---|
| 1238 |
|
---|
| 1239 | my $equiv_form = "";
|
---|
| 1240 | foreach my $gaf (@{$metadata->{$field}}) {
|
---|
| 1241 | my ($full_filename,$mimetype) = ($gaf =~ m/^(.*):(.*):$/);
|
---|
| 1242 | my ($tail_filename) = ($full_filename =~ /^.*[\/\\](.+?)$/);
|
---|
| 1243 | my $filename = $full_filename;
|
---|
| 1244 |
|
---|
| 1245 | $doc_obj->associate_file($full_filename,$tail_filename,$mimetype);
|
---|
[11122] | 1246 |
|
---|
| 1247 | # work out extended tail extension (i.e. matching tail re)
|
---|
| 1248 |
|
---|
| 1249 | my ($file_prefix,$file_extended_ext)
|
---|
| 1250 | = $self->root_ext_split($tail_filename,$associate_tail_re);
|
---|
| 1251 | my ($pre_doc_ext) = ($file_extended_ext =~ m/^(.*)\..*$/);
|
---|
| 1252 |
|
---|
[8510] | 1253 | my ($doc_ext) = ($tail_filename =~ m/^.*\.(.*)$/);
|
---|
[11834] | 1254 | my $start_doclink = "<a href=\"_httpprefix_/collect/[collection]/index/assoc/{Or}{[parent(Top):archivedir],[archivedir]}/$tail_filename\">";
|
---|
[8510] | 1255 | my $srcicon = "_icon".$doc_ext."_";
|
---|
| 1256 | my $end_doclink = "</a>";
|
---|
| 1257 |
|
---|
[11122] | 1258 | my $assoc_form = "$start_doclink\{If\}{$srcicon,$srcicon,$doc_ext\}$end_doclink";
|
---|
| 1259 |
|
---|
| 1260 | if (defined $pre_doc_ext) {
|
---|
| 1261 | # for metadata such as [mp3._edited] [mp3._full] ...
|
---|
| 1262 | $doc_obj->add_utf8_metadata ($cursection, "$doc_ext.$pre_doc_ext", $assoc_form);
|
---|
| 1263 | }
|
---|
| 1264 |
|
---|
| 1265 | # for multiple metadata such as [mp3.assoclink]
|
---|
| 1266 | $doc_obj->add_utf8_metadata ($cursection, "$doc_ext.assoclink", $assoc_form);
|
---|
| 1267 |
|
---|
| 1268 | $equiv_form .= " $assoc_form";
|
---|
[8510] | 1269 | }
|
---|
| 1270 | $doc_obj->add_utf8_metadata ($cursection, "equivlink", $equiv_form);
|
---|
| 1271 | }
|
---|
| 1272 | elsif (ref ($metadata->{$field}) eq "ARRAY") {
|
---|
[839] | 1273 | map {
|
---|
[1219] | 1274 | $doc_obj->add_utf8_metadata ($cursection, $field, $_);
|
---|
[839] | 1275 | } @{$metadata->{$field}};
|
---|
| 1276 | } else {
|
---|
[1219] | 1277 | $doc_obj->add_utf8_metadata ($cursection, $field, $metadata->{$field});
|
---|
[839] | 1278 | }
|
---|
| 1279 | }
|
---|
| 1280 | }
|
---|
| 1281 |
|
---|
[1396] | 1282 | # initialise metadata extractors
|
---|
| 1283 | sub initialise_extractors {
|
---|
| 1284 | my $self = shift (@_);
|
---|
| 1285 |
|
---|
| 1286 | if ($self->{'extract_acronyms'} || $self->{'markup_acronyms'}) {
|
---|
| 1287 | &acronym::initialise_acronyms();
|
---|
| 1288 | }
|
---|
| 1289 | }
|
---|
| 1290 |
|
---|
| 1291 | # finalise metadata extractors
|
---|
| 1292 | sub finalise_extractors {
|
---|
| 1293 | my $self = shift (@_);
|
---|
| 1294 |
|
---|
| 1295 | if ($self->{'extract_acronyms'} || $self->{'markup_acronyms'}) {
|
---|
| 1296 | &acronym::finalise_acronyms();
|
---|
| 1297 | }
|
---|
| 1298 | }
|
---|
| 1299 |
|
---|
[1602] | 1300 | # FIRSTNNN: extract the first NNN characters as metadata
|
---|
| 1301 | sub extract_first_NNNN_characters {
|
---|
| 1302 | my $self = shift (@_);
|
---|
| 1303 | my ($textref, $doc_obj, $thissection) = @_;
|
---|
| 1304 |
|
---|
| 1305 | foreach my $size (split /,/, $self->{'first'}) {
|
---|
| 1306 | my $tmptext = $$textref;
|
---|
| 1307 | $tmptext =~ s/^\s+//;
|
---|
| 1308 | $tmptext =~ s/\s+$//;
|
---|
| 1309 | $tmptext =~ s/\s+/ /gs;
|
---|
| 1310 | $tmptext = substr ($tmptext, 0, $size);
|
---|
| 1311 | $tmptext =~ s/\s\S*$/…/;
|
---|
| 1312 | $doc_obj->add_utf8_metadata ($thissection, "First$size", $tmptext);
|
---|
| 1313 | }
|
---|
| 1314 | }
|
---|
| 1315 |
|
---|
| 1316 | sub extract_email {
|
---|
| 1317 | my $self = shift (@_);
|
---|
| 1318 | my ($textref, $doc_obj, $thissection) = @_;
|
---|
| 1319 | my $outhandle = $self->{'outhandle'};
|
---|
| 1320 |
|
---|
[9413] | 1321 | gsprintf($outhandle, " {BasPlug.extracting_emails}...\n")
|
---|
[1844] | 1322 | if ($self->{'verbosity'} > 2);
|
---|
[1602] | 1323 |
|
---|
[2604] | 1324 | my @email = ($$textref =~ m/([-a-z0-9\.@+_=]+@(?:[-a-z0-9]+\.)+(?:com|org|edu|mil|int|net|[a-z][a-z]))/g);
|
---|
[1602] | 1325 | @email = sort @email;
|
---|
| 1326 |
|
---|
[10218] | 1327 | # if($self->{"new_extract_email"} == 0)
|
---|
| 1328 | # {
|
---|
| 1329 | # my @email2 = ();
|
---|
| 1330 | # foreach my $address (@email)
|
---|
| 1331 | # {
|
---|
| 1332 | # if (!(join(" ",@email2) =~ m/(^| )$address( |$)/ ))
|
---|
| 1333 | # {
|
---|
| 1334 | # push @email2, $address;
|
---|
| 1335 | # $doc_obj->add_utf8_metadata ($thissection, "emailAddress", $address);
|
---|
| 1336 | # # print $outhandle " extracting $address\n"
|
---|
| 1337 | # &gsprintf($outhandle, " {BasPlug.extracting} $address\n")
|
---|
| 1338 | # if ($self->{'verbosity'} > 3);
|
---|
| 1339 | # }
|
---|
| 1340 | # }
|
---|
| 1341 | # }
|
---|
| 1342 | # else
|
---|
| 1343 | # {
|
---|
| 1344 | my $hashExistMail = {};
|
---|
[1602] | 1345 | foreach my $address (@email) {
|
---|
[10218] | 1346 | if (!(defined $hashExistMail->{$address}))
|
---|
| 1347 | {
|
---|
| 1348 | $hashExistMail->{$address} = 1;
|
---|
[1602] | 1349 | $doc_obj->add_utf8_metadata ($thissection, "emailAddress", $address);
|
---|
[9413] | 1350 | gsprintf($outhandle, " {BasPlug.extracting} $address\n")
|
---|
[1844] | 1351 | if ($self->{'verbosity'} > 3);
|
---|
[1602] | 1352 | }
|
---|
| 1353 | }
|
---|
[9413] | 1354 | gsprintf($outhandle, " {BasPlug.done_email_extract}\n")
|
---|
[1844] | 1355 | if ($self->{'verbosity'} > 2);
|
---|
[1602] | 1356 | }
|
---|
| 1357 |
|
---|
| 1358 | # extract metadata
|
---|
[5681] | 1359 | sub auto_extract_metadata {
|
---|
[1954] | 1360 |
|
---|
[1242] | 1361 | my $self = shift (@_);
|
---|
| 1362 | my ($doc_obj) = @_;
|
---|
[1602] | 1363 |
|
---|
| 1364 | if ($self->{'extract_email'}) {
|
---|
| 1365 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1366 | while (defined $thissection) {
|
---|
| 1367 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1368 | $self->extract_email (\$text, $doc_obj, $thissection) if $text =~ /./;
|
---|
| 1369 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1370 | }
|
---|
[1954] | 1371 | }
|
---|
[9398] | 1372 | if ($self->{'extract_placenames'}) {
|
---|
| 1373 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1374 | while (defined $thissection) {
|
---|
| 1375 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1376 | $self->extract_placenames (\$text, $doc_obj, $thissection) if $text =~ /./;
|
---|
| 1377 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1378 | }
|
---|
| 1379 | }
|
---|
[1954] | 1380 |
|
---|
[11069] | 1381 | if ($self->{'extract_keyphrases'} || $self->{'extract_keyphrases_kea4'}) {
|
---|
| 1382 | $self->extract_keyphrases($doc_obj);
|
---|
| 1383 | }
|
---|
[1954] | 1384 |
|
---|
[1602] | 1385 | if ($self->{'first'}) {
|
---|
| 1386 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1387 | while (defined $thissection) {
|
---|
| 1388 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1389 | $self->extract_first_NNNN_characters (\$text, $doc_obj, $thissection) if $text =~ /./;
|
---|
| 1390 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1391 | }
|
---|
| 1392 | }
|
---|
| 1393 |
|
---|
[1242] | 1394 | if ($self->{'extract_acronyms'}) {
|
---|
| 1395 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1396 | while (defined $thissection) {
|
---|
| 1397 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1398 | $self->extract_acronyms (\$text, $doc_obj, $thissection) if $text =~ /./;
|
---|
| 1399 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1400 | }
|
---|
| 1401 | }
|
---|
[1602] | 1402 |
|
---|
[1393] | 1403 | if ($self->{'markup_acronyms'}) {
|
---|
| 1404 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1405 | while (defined $thissection) {
|
---|
| 1406 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1407 | $text = $self->markup_acronyms ($text, $doc_obj, $thissection);
|
---|
| 1408 | $doc_obj->delete_text($thissection);
|
---|
| 1409 | $doc_obj->add_text($thissection, $text);
|
---|
| 1410 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1411 | }
|
---|
| 1412 | }
|
---|
| 1413 |
|
---|
[10218] | 1414 | if($self->{'extract_historical_years'}) {
|
---|
[1317] | 1415 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1416 | while (defined $thissection) {
|
---|
[10218] | 1417 |
|
---|
[1317] | 1418 | my $text = $doc_obj->get_text($thissection);
|
---|
[1846] | 1419 | &DateExtract::get_date_metadata($text, $doc_obj,
|
---|
| 1420 | $thissection,
|
---|
[10218] | 1421 | $self->{'no_bibliography'},
|
---|
| 1422 | $self->{'maximum_year'},
|
---|
| 1423 | $self->{'maximum_century'});
|
---|
[1317] | 1424 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1425 | }
|
---|
| 1426 | }
|
---|
[1242] | 1427 | }
|
---|
| 1428 |
|
---|
[11069] | 1429 |
|
---|
| 1430 | #adding kea keyphrases
|
---|
| 1431 | sub extract_keyphrases
|
---|
| 1432 | {
|
---|
| 1433 | my $self = shift(@_);
|
---|
| 1434 | my $doc_obj = shift(@_);
|
---|
| 1435 |
|
---|
| 1436 | # Use Kea 3.0 unless 4.0 has been specified
|
---|
| 1437 | my $kea_version = "3.0";
|
---|
| 1438 | if ($self->{'extract_keyphrases_kea4'}) {
|
---|
| 1439 | $kea_version = "4.0";
|
---|
| 1440 | }
|
---|
| 1441 |
|
---|
| 1442 | # Check that Kea exists, and tell the user where to get it if not
|
---|
| 1443 | my $keahome = &Kea::get_Kea_directory($kea_version);
|
---|
| 1444 | if (!-e $keahome) {
|
---|
| 1445 | gsprintf(STDERR, "{BasPlug.missing_kea}\n", $keahome, $kea_version);
|
---|
| 1446 | return;
|
---|
| 1447 | }
|
---|
| 1448 |
|
---|
| 1449 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1450 | my $text = "";
|
---|
| 1451 | my $list;
|
---|
| 1452 |
|
---|
| 1453 | #loop through sections to gather whole doc
|
---|
| 1454 | while (defined $thissection) {
|
---|
| 1455 | my $sectiontext = $doc_obj->get_text($thissection);
|
---|
| 1456 | $text = $text.$sectiontext;
|
---|
| 1457 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1458 | }
|
---|
| 1459 |
|
---|
| 1460 | if($self->{'extract_keyphrase_options'}) { #if kea options flag is set, call Kea with specified options
|
---|
| 1461 | $list = &Kea::extract_KeyPhrases ($kea_version, $text, $self->{'extract_keyphrase_options'});
|
---|
| 1462 | } else { #otherwise call Kea with no options
|
---|
| 1463 | $list = &Kea::extract_KeyPhrases ($kea_version, $text);
|
---|
| 1464 | }
|
---|
| 1465 |
|
---|
| 1466 | if ($list){
|
---|
| 1467 | # if a list of kea keyphrases was returned (ie not empty)
|
---|
| 1468 | if ($self->{'verbosity'}) {
|
---|
| 1469 | gsprintf(STDERR, "{BasPlug.keyphrases}: $list\n");
|
---|
| 1470 | }
|
---|
| 1471 |
|
---|
| 1472 | #add metadata to top section
|
---|
| 1473 | $thissection = $doc_obj->get_top_section();
|
---|
| 1474 |
|
---|
| 1475 | # add all key phrases as one metadata
|
---|
| 1476 | $doc_obj->add_metadata($thissection, "Keyphrases", $list);
|
---|
| 1477 |
|
---|
| 1478 | # add individual key phrases as multiple metadata
|
---|
| 1479 | foreach my $keyphrase (split(',', $list)) {
|
---|
| 1480 | $keyphrase =~ s/^\s+|\s+$//g;
|
---|
| 1481 | $doc_obj->add_metadata($thissection, "Keyphrase", $keyphrase);
|
---|
| 1482 | }
|
---|
| 1483 | }
|
---|
| 1484 | }
|
---|
| 1485 |
|
---|
| 1486 |
|
---|
[1335] | 1487 | # extract acronyms from a section in a document. progress is
|
---|
[1424] | 1488 | # reported to outhandle based on the verbosity. both the Acronym
|
---|
[1335] | 1489 | # and the AcronymKWIC metadata items are created.
|
---|
| 1490 |
|
---|
[1242] | 1491 | sub extract_acronyms {
|
---|
| 1492 | my $self = shift (@_);
|
---|
| 1493 | my ($textref, $doc_obj, $thissection) = @_;
|
---|
[1424] | 1494 | my $outhandle = $self->{'outhandle'};
|
---|
[1242] | 1495 |
|
---|
[5681] | 1496 | # print $outhandle " extracting acronyms ...\n"
|
---|
[9413] | 1497 | gsprintf($outhandle, " {BasPlug.extracting_acronyms}...\n")
|
---|
[1844] | 1498 | if ($self->{'verbosity'} > 2);
|
---|
[1335] | 1499 |
|
---|
[1242] | 1500 | my $acro_array = &acronym::acronyms($textref);
|
---|
[1360] | 1501 |
|
---|
[1242] | 1502 | foreach my $acro (@$acro_array) {
|
---|
| 1503 |
|
---|
[1335] | 1504 | #check that this is the first time ...
|
---|
| 1505 | my $seen_before = "false";
|
---|
| 1506 | my $previous_data = $doc_obj->get_metadata($thissection, "Acronym");
|
---|
| 1507 | foreach my $thisAcro (@$previous_data) {
|
---|
[1602] | 1508 | if ($thisAcro eq $acro->to_string()) {
|
---|
[1335] | 1509 | $seen_before = "true";
|
---|
[9413] | 1510 | if ($self->{'verbosity'} >= 4) {
|
---|
| 1511 | gsprintf($outhandle, " {BasPlug.already_seen} " .
|
---|
| 1512 | $acro->to_string() . "\n");
|
---|
| 1513 | }
|
---|
[5681] | 1514 | }
|
---|
[1242] | 1515 | }
|
---|
[1335] | 1516 |
|
---|
[1602] | 1517 | if ($seen_before eq "false") {
|
---|
[1393] | 1518 | #write it to the file ...
|
---|
| 1519 | $acro->write_to_file();
|
---|
| 1520 |
|
---|
[1335] | 1521 | #do the normal acronym
|
---|
| 1522 | $doc_obj->add_utf8_metadata($thissection, "Acronym", $acro->to_string());
|
---|
[9413] | 1523 | gsprintf($outhandle, " {BasPlug.adding} ".$acro->to_string()."\n")
|
---|
[1844] | 1524 | if ($self->{'verbosity'} > 3);
|
---|
[1335] | 1525 | }
|
---|
[1242] | 1526 | }
|
---|
[5681] | 1527 |
|
---|
[9413] | 1528 | gsprintf($outhandle, " {BasPlug.done_acronym_extract}\n")
|
---|
[1844] | 1529 | if ($self->{'verbosity'} > 2);
|
---|
[1242] | 1530 | }
|
---|
| 1531 |
|
---|
[1393] | 1532 | sub markup_acronyms {
|
---|
| 1533 | my $self = shift (@_);
|
---|
| 1534 | my ($text, $doc_obj, $thissection) = @_;
|
---|
[1424] | 1535 | my $outhandle = $self->{'outhandle'};
|
---|
[1393] | 1536 |
|
---|
[9413] | 1537 | gsprintf($outhandle, " {BasPlug.marking_up_acronyms}...\n")
|
---|
[1844] | 1538 | if ($self->{'verbosity'} > 2);
|
---|
[1393] | 1539 |
|
---|
| 1540 | #self is passed in to check for verbosity ...
|
---|
| 1541 | $text = &acronym::markup_acronyms($text, $self);
|
---|
| 1542 |
|
---|
[9413] | 1543 | gsprintf($outhandle, " {BasPlug.done_acronym_markup}\n")
|
---|
[1844] | 1544 | if ($self->{'verbosity'} > 2);
|
---|
[1393] | 1545 |
|
---|
| 1546 | return $text;
|
---|
| 1547 | }
|
---|
| 1548 |
|
---|
[2785] | 1549 | sub compile_stats {
|
---|
| 1550 | my $self = shift(@_);
|
---|
| 1551 | my ($stats) = @_;
|
---|
| 1552 |
|
---|
| 1553 | $stats->{'num_processed'} += $self->{'num_processed'};
|
---|
| 1554 | $stats->{'num_not_processed'} += $self->{'num_not_processed'};
|
---|
[2796] | 1555 | $stats->{'num_archives'} += $self->{'num_archives'};
|
---|
[2785] | 1556 |
|
---|
| 1557 | }
|
---|
| 1558 |
|
---|
[2816] | 1559 | sub associate_cover_image {
|
---|
[10833] | 1560 | my $self = shift;
|
---|
[2816] | 1561 | my ($doc_obj, $filename) = @_;
|
---|
| 1562 |
|
---|
[10833] | 1563 | $filename =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
| 1564 | if (exists $self->{'covers_missing_cache'}->{$filename}) {
|
---|
| 1565 | # don't stat() for existence eg for multiple document input files
|
---|
| 1566 | # (eg SplitPlug)
|
---|
| 1567 | return;
|
---|
| 1568 | }
|
---|
| 1569 |
|
---|
[9413] | 1570 | my $top_section=$doc_obj->get_top_section();
|
---|
| 1571 |
|
---|
[2816] | 1572 | if (-e $filename) {
|
---|
| 1573 | $doc_obj->associate_file($filename, "cover.jpg", "image/jpeg");
|
---|
[9413] | 1574 | $doc_obj->add_utf8_metadata($top_section, "hascover", 1);
|
---|
[3086] | 1575 | } else {
|
---|
[10833] | 1576 | my $upper_filename = $filename;
|
---|
| 1577 | $upper_filename =~ s/jpg$/JPG/;
|
---|
| 1578 | if (-e $upper_filename) {
|
---|
| 1579 | $doc_obj->associate_file($upper_filename, "cover.jpg",
|
---|
| 1580 | "image/jpeg");
|
---|
[9413] | 1581 | $doc_obj->add_utf8_metadata($top_section, "hascover", 1);
|
---|
[10833] | 1582 | } else {
|
---|
| 1583 | # file doesn't exist, so record the fact that it's missing so
|
---|
| 1584 | # we don't stat() again (stat is slow)
|
---|
| 1585 | $self->{'covers_missing_cache'}->{$filename} = 1;
|
---|
[3086] | 1586 | }
|
---|
[2816] | 1587 | }
|
---|
[10833] | 1588 |
|
---|
[2816] | 1589 | }
|
---|
| 1590 |
|
---|
[11332] | 1591 |
|
---|
| 1592 | # Overridden by exploding plugins (eg. ISISPlug)
|
---|
| 1593 | sub clean_up_after_exploding
|
---|
| 1594 | {
|
---|
| 1595 | my $self = shift(@_);
|
---|
| 1596 | }
|
---|
| 1597 |
|
---|
| 1598 |
|
---|
[4] | 1599 | 1;
|
---|