[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;
|
---|
[1219] | 47 | use cnseg;
|
---|
[1242] | 48 | use acronym;
|
---|
[1317] | 49 | use textcat;
|
---|
[1242] | 50 | use doc;
|
---|
[7645] | 51 | eval "require diagnostics"; # some perl distros (eg mac) don't have this
|
---|
[1411] | 52 | use DateExtract;
|
---|
[2751] | 53 | use ghtml;
|
---|
[9413] | 54 | use gsprintf 'gsprintf';
|
---|
[4778] | 55 | use printusage;
|
---|
[10218] | 56 | use parse2;
|
---|
[4] | 57 |
|
---|
[10218] | 58 |
|
---|
[9398] | 59 | use GISBasPlug;
|
---|
[5681] | 60 |
|
---|
[9413] | 61 | @BasPlug::ISA = ( GISBasPlug );
|
---|
[9398] | 62 |
|
---|
[4873] | 63 | my $unicode_list =
|
---|
[10218] | 64 | [ { 'name' => "ascii",
|
---|
[4873] | 65 | 'desc' => "{BasPlug.input_encoding.ascii}" },
|
---|
[4744] | 66 | { 'name' => "utf8",
|
---|
[4873] | 67 | 'desc' => "{BasPlug.input_encoding.utf8}" },
|
---|
[4744] | 68 | { 'name' => "unicode",
|
---|
[4873] | 69 | 'desc' => "{BasPlug.input_encoding.unicode}" } ];
|
---|
[3540] | 70 |
|
---|
[10218] | 71 | my $auto_unicode_list =
|
---|
| 72 | [ { 'name' => "auto",
|
---|
| 73 | 'desc' => "{BasPlug.input_encoding.auto}" } ];
|
---|
| 74 |
|
---|
[10620] | 75 | my $e = $encodings::encodings;
|
---|
| 76 | foreach my $enc (sort {$e->{$a}->{'name'} cmp $e->{$b}->{'name'}} keys (%$e))
|
---|
| 77 | {
|
---|
| 78 | my $hashEncode =
|
---|
| 79 | {'name' => $enc,
|
---|
| 80 | 'desc' => $e->{$enc}->{'name'}};
|
---|
| 81 |
|
---|
| 82 | push(@{$unicode_list},$hashEncode);
|
---|
| 83 | }
|
---|
| 84 |
|
---|
| 85 | push(@{$auto_unicode_list},@{$unicode_list});
|
---|
| 86 |
|
---|
[4873] | 87 | my $arguments =
|
---|
[3540] | 88 | [ { 'name' => "process_exp",
|
---|
[4873] | 89 | 'desc' => "{BasPlug.process_exp}",
|
---|
[6408] | 90 | 'type' => "regexp",
|
---|
[3540] | 91 | 'deft' => "",
|
---|
| 92 | 'reqd' => "no" },
|
---|
| 93 | { 'name' => "block_exp",
|
---|
[4873] | 94 | 'desc' => "{BasPlug.block_exp}",
|
---|
[6408] | 95 | 'type' => "regexp",
|
---|
[3540] | 96 | 'deft' => "",
|
---|
| 97 | 'reqd' => "no" },
|
---|
[9067] | 98 | { 'name' => "smart_block",
|
---|
| 99 | 'desc' => "{BasPlug.smart_block}",
|
---|
| 100 | 'type' => "flag",
|
---|
[11089] | 101 | 'reqd' => "no",
|
---|
| 102 | 'hiddengli' => "yes" },
|
---|
[8892] | 103 | { 'name' => "associate_ext",
|
---|
| 104 | 'desc' => "{BasPlug.associate_ext}",
|
---|
| 105 | 'type' => "string",
|
---|
| 106 | 'reqd' => "no" },
|
---|
[3540] | 107 | { 'name' => "input_encoding",
|
---|
[4873] | 108 | 'desc' => "{BasPlug.input_encoding}",
|
---|
[3540] | 109 | 'type' => "enum",
|
---|
[10218] | 110 | 'list' => $auto_unicode_list,
|
---|
[3540] | 111 | 'reqd' => "no" ,
|
---|
| 112 | 'deft' => "auto" } ,
|
---|
| 113 | { 'name' => "default_encoding",
|
---|
[4873] | 114 | 'desc' => "{BasPlug.default_encoding}",
|
---|
[4744] | 115 | 'type' => "enum",
|
---|
[6332] | 116 | 'list' => $unicode_list,
|
---|
[4744] | 117 | 'reqd' => "no",
|
---|
| 118 | 'deft' => "utf8" },
|
---|
[3540] | 119 | { 'name' => "extract_language",
|
---|
[4873] | 120 | 'desc' => "{BasPlug.extract_language}",
|
---|
[3540] | 121 | 'type' => "flag",
|
---|
| 122 | 'reqd' => "no" },
|
---|
| 123 | { 'name' => "default_language",
|
---|
[4873] | 124 | 'desc' => "{BasPlug.default_language}",
|
---|
[10329] | 125 | 'type' => "string",
|
---|
[3540] | 126 | 'deft' => "en",
|
---|
| 127 | 'reqd' => "no" },
|
---|
| 128 | { 'name' => "extract_acronyms",
|
---|
[4873] | 129 | 'desc' => "{BasPlug.extract_acronyms}",
|
---|
[3540] | 130 | 'type' => "flag",
|
---|
| 131 | 'reqd' => "no" },
|
---|
| 132 | { 'name' => "markup_acronyms",
|
---|
[4873] | 133 | 'desc' => "{BasPlug.markup_acronyms}",
|
---|
[3540] | 134 | 'type' => "flag",
|
---|
[9398] | 135 | 'reqd' => "no" },
|
---|
[8789] | 136 | { 'name' => "extract_keyphrases",
|
---|
| 137 | 'desc' => "{BasPlug.extract_keyphrases}",
|
---|
| 138 | 'type' => "flag",
|
---|
[8814] | 139 | 'reqd' => "no" },
|
---|
[11069] | 140 | { 'name' => "extract_keyphrases_kea4",
|
---|
| 141 | 'desc' => "{BasPlug.extract_keyphrases_kea4}",
|
---|
| 142 | 'type' => "flag",
|
---|
| 143 | 'reqd' => "no" },
|
---|
[8789] | 144 | { 'name' => "extract_keyphrase_options",
|
---|
| 145 | 'desc' => "{BasPlug.extract_keyphrase_options}",
|
---|
| 146 | 'type' => "string",
|
---|
| 147 | 'deft' => "",
|
---|
[8814] | 148 | 'reqd' => "no" },
|
---|
[3540] | 149 | { 'name' => "first",
|
---|
[4873] | 150 | 'desc' => "{BasPlug.first}",
|
---|
[3540] | 151 | 'type' => "string",
|
---|
| 152 | 'reqd' => "no" },
|
---|
| 153 | { 'name' => "extract_email",
|
---|
[4873] | 154 | 'desc' => "{BasPlug.extract_email}",
|
---|
[3540] | 155 | 'type' => "flag",
|
---|
| 156 | 'reqd' => "no" },
|
---|
| 157 | { 'name' => "extract_historical_years",
|
---|
[4873] | 158 | 'desc' => "{BasPlug.extract_historical_years}",
|
---|
[3540] | 159 | 'type' => "flag",
|
---|
| 160 | 'reqd' => "no" },
|
---|
| 161 | { 'name' => "maximum_year",
|
---|
[4873] | 162 | 'desc' => "{BasPlug.maximum_year}",
|
---|
[3540] | 163 | 'type' => "int",
|
---|
[4744] | 164 | 'deft' => (localtime)[5]+1900,
|
---|
[10218] | 165 | 'char_length' => "4",
|
---|
| 166 | #'range' => "2,100",
|
---|
[3540] | 167 | 'reqd' => "no"},
|
---|
| 168 | { 'name' => "maximum_century",
|
---|
[4873] | 169 | 'desc' => "{BasPlug.maximum_century}",
|
---|
[7105] | 170 | 'type' => "string",
|
---|
[10218] | 171 | 'deft' => "-1",
|
---|
[3540] | 172 | 'reqd' => "no" },
|
---|
| 173 | { 'name' => "no_bibliography",
|
---|
[4873] | 174 | 'desc' => "{BasPlug.no_bibliography}",
|
---|
[3540] | 175 | 'type' => "flag",
|
---|
| 176 | 'reqd' => "no"},
|
---|
[8678] | 177 | { 'name' => "no_cover_image",
|
---|
| 178 | 'desc' => "{BasPlug.no_cover_image}",
|
---|
[3540] | 179 | 'type' => "flag",
|
---|
[10218] | 180 | 'reqd' => "no" },
|
---|
| 181 | { 'name' => "separate_cjk",
|
---|
| 182 | 'desc' => "{BasPlug.separate_cjk}",
|
---|
| 183 | 'type' => "flag",
|
---|
| 184 | 'reqd' => "no",
|
---|
| 185 | 'hiddengli' => "yes" },
|
---|
| 186 | { 'name' => "new_extract_email",
|
---|
| 187 | 'desc' => "",
|
---|
| 188 | 'type' => "flag",
|
---|
| 189 | 'reqd' => "no",
|
---|
| 190 | 'hiddengli' => "yes" } ];
|
---|
[3540] | 191 |
|
---|
[9398] | 192 | my $gis_arguments =
|
---|
| 193 | [ { 'name' => "extract_placenames",
|
---|
| 194 | 'desc' => "{GISBasPlug.extract_placenames}",
|
---|
| 195 | 'type' => "flag",
|
---|
| 196 | 'reqd' => "no" },
|
---|
| 197 | { 'name' => "gazetteer",
|
---|
| 198 | 'desc' => "{GISBasPlug.gazetteer}",
|
---|
| 199 | 'type' => "string",
|
---|
| 200 | 'reqd' => "no" },
|
---|
| 201 | { 'name' => "place_list",
|
---|
| 202 | 'desc' => "{GISBasPlug.place_list}",
|
---|
| 203 | 'type' => "flag",
|
---|
| 204 | 'reqd' => "no" } ];
|
---|
| 205 |
|
---|
| 206 |
|
---|
[3540] | 207 | my $options = { 'name' => "BasPlug",
|
---|
[5681] | 208 | 'desc' => "{BasPlug.desc}",
|
---|
[6408] | 209 | 'abstract' => "yes",
|
---|
| 210 | 'inherits' => "no",
|
---|
[4750] | 211 | 'args' => $arguments };
|
---|
[3540] | 212 |
|
---|
[4778] | 213 |
|
---|
[10478] | 214 | sub set_keepold {
|
---|
| 215 | my $self = shift(@_);
|
---|
| 216 | my ($keepold) = @_;
|
---|
| 217 |
|
---|
| 218 | $self->{'keepold'} = $keepold;
|
---|
| 219 | }
|
---|
| 220 |
|
---|
[4873] | 221 | sub get_arguments
|
---|
| 222 | {
|
---|
[8716] | 223 | my $self = shift(@_);
|
---|
| 224 | my $optionlistref = $self->{'option_list'};
|
---|
| 225 | my @optionlist = @$optionlistref;
|
---|
| 226 | my $pluginoptions = pop(@$optionlistref);
|
---|
| 227 | my $pluginarguments = $pluginoptions->{'args'};
|
---|
[4873] | 228 | return $pluginarguments;
|
---|
| 229 | }
|
---|
| 230 |
|
---|
| 231 |
|
---|
[4778] | 232 | sub print_xml_usage
|
---|
| 233 | {
|
---|
[8716] | 234 | my $self = shift(@_);
|
---|
[4778] | 235 |
|
---|
[6945] | 236 | # XML output is always in UTF-8
|
---|
[9413] | 237 | gsprintf::output_strings_in_UTF8;
|
---|
[6945] | 238 |
|
---|
[9413] | 239 | PrintUsage::print_xml_header();
|
---|
[6925] | 240 | $self->print_xml();
|
---|
[3540] | 241 | }
|
---|
| 242 |
|
---|
[4778] | 243 |
|
---|
| 244 | sub print_xml
|
---|
| 245 | {
|
---|
[8716] | 246 | my $self = shift(@_);
|
---|
[4778] | 247 |
|
---|
[8716] | 248 | my $optionlistref = $self->{'option_list'};
|
---|
| 249 | my @optionlist = @$optionlistref;
|
---|
[10229] | 250 | my $pluginoptions = shift(@$optionlistref);
|
---|
[4778] | 251 | return if (!defined($pluginoptions));
|
---|
| 252 |
|
---|
[9413] | 253 | gsprintf(STDERR, "<PlugInfo>\n");
|
---|
| 254 | gsprintf(STDERR, " <Name>$pluginoptions->{'name'}</Name>\n");
|
---|
| 255 | my $desc = gsprintf::lookup_string($pluginoptions->{'desc'});
|
---|
[7023] | 256 | $desc =~ s/</&lt;/g; # doubly escaped
|
---|
| 257 | $desc =~ s/>/&gt;/g;
|
---|
| 258 |
|
---|
[9413] | 259 | gsprintf(STDERR, " <Desc>$desc</Desc>\n");
|
---|
| 260 | gsprintf(STDERR, " <Abstract>$pluginoptions->{'abstract'}</Abstract>\n");
|
---|
| 261 | gsprintf(STDERR, " <Inherits>$pluginoptions->{'inherits'}</Inherits>\n");
|
---|
| 262 | gsprintf(STDERR, " <Explodes>" . ($pluginoptions->{'explodes'} || "no") . "</Explodes>\n");
|
---|
| 263 | gsprintf(STDERR, " <Arguments>\n");
|
---|
[4778] | 264 | if (defined($pluginoptions->{'args'})) {
|
---|
[6925] | 265 | &PrintUsage::print_options_xml($pluginoptions->{'args'});
|
---|
[3540] | 266 | }
|
---|
[4778] | 267 |
|
---|
| 268 | # Recurse up the plugin hierarchy
|
---|
[6925] | 269 | $self->print_xml();
|
---|
[4778] | 270 |
|
---|
[9413] | 271 | gsprintf(STDERR, " </Arguments>\n");
|
---|
| 272 | gsprintf(STDERR, "</PlugInfo>\n");
|
---|
[3540] | 273 | }
|
---|
| 274 |
|
---|
[4744] | 275 |
|
---|
[4778] | 276 | sub print_txt_usage
|
---|
[4744] | 277 | {
|
---|
[8716] | 278 | my $self = shift(@_);
|
---|
[4750] | 279 | # Print the usage message for a plugin (recursively)
|
---|
[8716] | 280 | my $descoffset = $self->determine_description_offset(0);
|
---|
[6925] | 281 | $self->print_plugin_usage($descoffset, 1);
|
---|
[4750] | 282 | }
|
---|
[4744] | 283 |
|
---|
| 284 |
|
---|
[4750] | 285 | sub determine_description_offset
|
---|
| 286 | {
|
---|
[8716] | 287 | my $self = shift(@_);
|
---|
| 288 | my $maxoffset = shift(@_);
|
---|
[4750] | 289 |
|
---|
[8716] | 290 | my $optionlistref = $self->{'option_list'};
|
---|
| 291 | my @optionlist = @$optionlistref;
|
---|
[10229] | 292 | my $pluginoptions = shift(@$optionlistref);
|
---|
[4750] | 293 | return $maxoffset if (!defined($pluginoptions));
|
---|
| 294 |
|
---|
| 295 | # Find the length of the longest option string of this plugin
|
---|
[8716] | 296 | my $pluginargs = $pluginoptions->{'args'};
|
---|
[4744] | 297 | if (defined($pluginargs)) {
|
---|
[8716] | 298 | my $longest = &PrintUsage::find_longest_option_string($pluginargs);
|
---|
[4778] | 299 | if ($longest > $maxoffset) {
|
---|
| 300 | $maxoffset = $longest;
|
---|
[4744] | 301 | }
|
---|
[4750] | 302 | }
|
---|
[4744] | 303 |
|
---|
[4750] | 304 | # Recurse up the plugin hierarchy
|
---|
| 305 | $maxoffset = $self->determine_description_offset($maxoffset);
|
---|
| 306 | $self->{'option_list'} = \@optionlist;
|
---|
| 307 | return $maxoffset;
|
---|
| 308 | }
|
---|
| 309 |
|
---|
| 310 |
|
---|
| 311 | sub print_plugin_usage
|
---|
| 312 | {
|
---|
[8716] | 313 | my $self = shift(@_);
|
---|
| 314 | my $descoffset = shift(@_);
|
---|
| 315 | my $isleafclass = shift(@_);
|
---|
[4750] | 316 |
|
---|
[8716] | 317 | my $optionlistref = $self->{'option_list'};
|
---|
| 318 | my @optionlist = @$optionlistref;
|
---|
[10229] | 319 | my $pluginoptions = shift(@$optionlistref);
|
---|
[4750] | 320 | return if (!defined($pluginoptions));
|
---|
| 321 |
|
---|
[8716] | 322 | my $pluginname = $pluginoptions->{'name'};
|
---|
| 323 | my $pluginargs = $pluginoptions->{'args'};
|
---|
| 324 | my $plugindesc = $pluginoptions->{'desc'};
|
---|
[4750] | 325 |
|
---|
| 326 | # Produce the usage information using the data structure above
|
---|
| 327 | if ($isleafclass) {
|
---|
[6932] | 328 | if (defined($plugindesc)) {
|
---|
[9413] | 329 | gsprintf(STDERR, "$plugindesc\n\n");
|
---|
[6932] | 330 | }
|
---|
[9413] | 331 | gsprintf(STDERR, " {common.usage}: plugin $pluginname [{common.options}]\n\n");
|
---|
[4750] | 332 | }
|
---|
| 333 |
|
---|
| 334 | # Display the plugin options, if there are some
|
---|
| 335 | if (defined($pluginargs)) {
|
---|
[4744] | 336 | # Calculate the column offset of the option descriptions
|
---|
[8716] | 337 | my $optiondescoffset = $descoffset + 2; # 2 spaces between options & descriptions
|
---|
[4744] | 338 |
|
---|
[4750] | 339 | if ($isleafclass) {
|
---|
[9413] | 340 | gsprintf(STDERR, " {common.specific_options}:\n");
|
---|
[4750] | 341 | }
|
---|
| 342 | else {
|
---|
[9413] | 343 | gsprintf(STDERR, " {common.general_options}:\n", $pluginname);
|
---|
[4750] | 344 | }
|
---|
| 345 |
|
---|
[4744] | 346 | # Display the plugin options
|
---|
[6925] | 347 | &PrintUsage::print_options_txt($pluginargs, $optiondescoffset);
|
---|
[4744] | 348 | }
|
---|
| 349 |
|
---|
[4750] | 350 | # Recurse up the plugin hierarchy
|
---|
[6925] | 351 | $self->print_plugin_usage($descoffset, 0);
|
---|
[4750] | 352 | $self->{'option_list'} = \@optionlist;
|
---|
[4744] | 353 | }
|
---|
| 354 |
|
---|
| 355 |
|
---|
[4] | 356 | sub new {
|
---|
[10218] | 357 | # Set Encodings to the list!!
|
---|
| 358 |
|
---|
| 359 |
|
---|
| 360 | # Start the BasPlug Constructor
|
---|
[1219] | 361 | my $class = shift (@_);
|
---|
[10218] | 362 | my ($pluginlist,$args,$hashArgOptLists) = @_;
|
---|
| 363 | push(@$pluginlist, $class);
|
---|
| 364 | my $plugin_name = (defined $pluginlist->[0]) ? $pluginlist->[0] : $class;
|
---|
[9398] | 365 |
|
---|
[10218] | 366 | if(defined $arguments){ push(@{$hashArgOptLists->{"ArgList"}},@{$arguments});}
|
---|
| 367 | if(defined $options) { push(@{$hashArgOptLists->{"OptList"}},$options)};
|
---|
| 368 |
|
---|
[9398] | 369 | if (GISBasPlug::has_mapdata()) {
|
---|
| 370 | push(@$arguments,@$gis_arguments);
|
---|
| 371 | }
|
---|
[10218] | 372 |
|
---|
| 373 | my $self = {};
|
---|
[10579] | 374 | $self->{'outhandle'} = STDERR;
|
---|
| 375 | $self->{'option_list'} = $hashArgOptLists->{"OptList"};
|
---|
| 376 | $self->{"info_only"} = 0;
|
---|
| 377 |
|
---|
| 378 | # Check if gsdlinfo is in the argument list or not - if it is, don't parse
|
---|
| 379 | # the args, just return the object.
|
---|
| 380 | foreach my $strArg (@{$args})
|
---|
| 381 | {
|
---|
| 382 | if($strArg eq "-gsdlinfo")
|
---|
| 383 | {
|
---|
| 384 | $self->{"info_only"} = 1;
|
---|
| 385 | return bless $self, $class;
|
---|
| 386 | }
|
---|
| 387 | }
|
---|
| 388 |
|
---|
[10218] | 389 | if(!parse2::parse($args,$hashArgOptLists->{"ArgList"},$self))
|
---|
| 390 | {
|
---|
| 391 | my $classTempClass = bless $self, $class;
|
---|
[10620] | 392 | print STDERR "<BadPlugin p=$plugin_name>\n";
|
---|
[10218] | 393 | &gsprintf(STDERR, "\n{BasPlug.bad_general_option}\n", $plugin_name);
|
---|
| 394 | $classTempClass->print_txt_usage(""); # Use default resource bundle
|
---|
| 395 | die "\n";
|
---|
| 396 | }
|
---|
[9398] | 397 |
|
---|
[10280] | 398 |
|
---|
[10579] | 399 | delete $self->{"info_only"};
|
---|
[10218] | 400 | # else parsing was successful.
|
---|
| 401 |
|
---|
| 402 | $self->{'plugin_type'} = $plugin_name;
|
---|
[10579] | 403 | #$self->{'outhandle'} = STDERR;
|
---|
[2235] | 404 | $self->{'textcat'} = new textcat();
|
---|
[2785] | 405 | $self->{'num_processed'} = 0;
|
---|
| 406 | $self->{'num_not_processed'} = 0;
|
---|
| 407 | $self->{'num_blocked'} = 0;
|
---|
| 408 | $self->{'num_archives'} = 0;
|
---|
[8678] | 409 | $self->{'cover_image'} = 1; # cover image is on by default
|
---|
[10218] | 410 | $self->{'cover_image'} = 0 if ($self->{'no_cover_image'});
|
---|
[10579] | 411 | #$self->{'option_list'} = $hashArgOptLists->{"OptList"};
|
---|
[3540] | 412 |
|
---|
[8892] | 413 | my $associate_ext = $self->{'associate_ext'};
|
---|
| 414 | if ((defined $associate_ext) && ($associate_ext ne "")) {
|
---|
| 415 | my @exts = split(/,/,$associate_ext);
|
---|
[9351] | 416 |
|
---|
[8892] | 417 | my %associate_ext_lookup = ();
|
---|
| 418 | foreach my $e (@exts) {
|
---|
[9351] | 419 | $associate_ext_lookup{$e} = 1;
|
---|
[8892] | 420 | }
|
---|
| 421 |
|
---|
| 422 | $self->{'associate_ext_lookup'} = \%associate_ext_lookup;
|
---|
| 423 | }
|
---|
| 424 |
|
---|
| 425 | $self->{'shared_fileroot'} = {};
|
---|
[8510] | 426 | $self->{'file_blocks'} = {};
|
---|
[1219] | 427 |
|
---|
[9398] | 428 | if ($self->{'extract_placenames'}) {
|
---|
| 429 |
|
---|
| 430 | my $outhandle = $self->{'outhandle'};
|
---|
[10218] | 431 |
|
---|
[9398] | 432 | my $places_ref
|
---|
| 433 | = GISBasPlug::loadGISDatabase($outhandle,$self->{'gazetteer'});
|
---|
[10218] | 434 |
|
---|
[9398] | 435 | if (!defined $places_ref) {
|
---|
| 436 | print $outhandle "Warning: Error loading mapdata gazetteer \"$self->{'gazetteer'}\"\n";
|
---|
| 437 | print $outhandle " No placename extraction will take place.\n";
|
---|
| 438 | $self->{'extract_placenames'} = undef;
|
---|
| 439 | }
|
---|
| 440 | else {
|
---|
| 441 | $self->{'places'} = $places_ref;
|
---|
| 442 | }
|
---|
[10280] | 443 | }
|
---|
[11089] | 444 |
|
---|
[1219] | 445 | return bless $self, $class;
|
---|
[10218] | 446 |
|
---|
[4] | 447 | }
|
---|
| 448 |
|
---|
[1242] | 449 | # initialize BasPlug options
|
---|
| 450 | # if init() is overridden in a sub-class, remember to call BasPlug::init()
|
---|
| 451 | sub init {
|
---|
| 452 | my $self = shift (@_);
|
---|
[2785] | 453 | my ($verbosity, $outhandle, $failhandle) = @_;
|
---|
[1242] | 454 |
|
---|
| 455 | # verbosity is passed through from the processor
|
---|
| 456 | $self->{'verbosity'} = $verbosity;
|
---|
| 457 |
|
---|
[2785] | 458 | # as are the outhandle and failhandle
|
---|
[1424] | 459 | $self->{'outhandle'} = $outhandle if defined $outhandle;
|
---|
[2785] | 460 | $self->{'failhandle'} = $failhandle;
|
---|
[1424] | 461 |
|
---|
[1242] | 462 | # set process_exp and block_exp to defaults unless they were
|
---|
| 463 | # explicitly set
|
---|
[1244] | 464 |
|
---|
| 465 | if ((!$self->is_recursive()) and
|
---|
[1242] | 466 | (!defined $self->{'process_exp'}) || ($self->{'process_exp'} eq "")) {
|
---|
[1244] | 467 |
|
---|
[1242] | 468 | $self->{'process_exp'} = $self->get_default_process_exp ();
|
---|
| 469 | if ($self->{'process_exp'} eq "") {
|
---|
[1244] | 470 | warn ref($self) . " Warning: Non-recursive plugin has no process_exp\n";
|
---|
[1242] | 471 | }
|
---|
| 472 | }
|
---|
| 473 |
|
---|
| 474 | if ((!defined $self->{'block_exp'}) || ($self->{'block_exp'} eq "")) {
|
---|
| 475 | $self->{'block_exp'} = $self->get_default_block_exp ();
|
---|
| 476 | }
|
---|
[11089] | 477 |
|
---|
[1242] | 478 | }
|
---|
| 479 |
|
---|
[839] | 480 | sub begin {
|
---|
| 481 | my $self = shift (@_);
|
---|
| 482 | my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
|
---|
[1396] | 483 | $self->initialise_extractors();
|
---|
[839] | 484 | }
|
---|
| 485 |
|
---|
| 486 | sub end {
|
---|
[10155] | 487 | # potentially called at the end of each plugin pass
|
---|
| 488 | # import.pl only has one plugin pass, but buildcol.pl has multiple ones
|
---|
| 489 |
|
---|
[839] | 490 | my ($self) = @_;
|
---|
[1396] | 491 | $self->finalise_extractors();
|
---|
[839] | 492 | }
|
---|
| 493 |
|
---|
[10155] | 494 | sub deinit {
|
---|
| 495 | # called only once, after all plugin passes have been done
|
---|
| 496 |
|
---|
| 497 | my ($self) = @_;
|
---|
| 498 | }
|
---|
| 499 |
|
---|
[1242] | 500 | # this function should be overridden to return 1
|
---|
| 501 | # in recursive plugins
|
---|
[4] | 502 | sub is_recursive {
|
---|
| 503 | my $self = shift (@_);
|
---|
| 504 |
|
---|
[1242] | 505 | return 0;
|
---|
[4] | 506 | }
|
---|
| 507 |
|
---|
[1242] | 508 | sub get_default_block_exp {
|
---|
| 509 | my $self = shift (@_);
|
---|
| 510 |
|
---|
| 511 | return "";
|
---|
| 512 | }
|
---|
| 513 |
|
---|
| 514 | sub get_default_process_exp {
|
---|
| 515 | my $self = shift (@_);
|
---|
| 516 |
|
---|
| 517 | return "";
|
---|
| 518 | }
|
---|
| 519 |
|
---|
[9067] | 520 | # default implementation is to do nothing.
|
---|
| 521 | sub store_block_files
|
---|
| 522 | {
|
---|
| 523 | my $self =shift (@_);
|
---|
| 524 | my ($filename) = @_;
|
---|
| 525 | return;
|
---|
| 526 | }
|
---|
| 527 |
|
---|
| 528 | #default implementation is to block a file with same name as this, but extension jpg or JPG, if cover_images is on.
|
---|
| 529 | sub block_cover_image
|
---|
| 530 | {
|
---|
[10833] | 531 | my $self =shift;
|
---|
| 532 | my $filename = shift;
|
---|
| 533 |
|
---|
[9067] | 534 | if ($self->{'cover_image'}) {
|
---|
| 535 | my $coverfile = $filename;
|
---|
| 536 | $coverfile =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
| 537 | if (!-e $coverfile) {
|
---|
| 538 | $coverfile =~ s/jpg$/JPG/;
|
---|
| 539 | }
|
---|
| 540 | if (-e $coverfile) {
|
---|
| 541 | $self->{'file_blocks'}->{$coverfile} = 1;
|
---|
[11089] | 542 | }
|
---|
[9067] | 543 | }
|
---|
| 544 |
|
---|
| 545 | return;
|
---|
| 546 | }
|
---|
| 547 |
|
---|
[8510] | 548 | sub metadata_read {
|
---|
| 549 | my $self = shift (@_);
|
---|
| 550 | my ($pluginfo, $base_dir, $file, $metadata, $extrametakeys, $extrametadata, $processor, $maxdocs, $gli) = @_;
|
---|
[8892] | 551 | # Keep track of filenames with same root but different extensions
|
---|
[9067] | 552 | # Used to support -associate_ext
|
---|
[8892] | 553 |
|
---|
| 554 | my $associate_ext = $self->{'associate_ext'};
|
---|
| 555 | if ((defined $associate_ext) && ($associate_ext ne "")) {
|
---|
| 556 |
|
---|
| 557 | my ($file_prefix,$file_ext) = ($file =~ m/^(.*)\.(.*?)$/);
|
---|
| 558 | if ((defined $file_prefix) && (defined $file_ext)) {
|
---|
| 559 |
|
---|
| 560 | my $shared_fileroot = $self->{'shared_fileroot'};
|
---|
| 561 | if (!defined $shared_fileroot->{$file_prefix}) {
|
---|
| 562 | my $file_prefix_rec = { 'tie_to' => undef, 'exts' => {} };
|
---|
| 563 | $shared_fileroot->{$file_prefix} = $file_prefix_rec;
|
---|
| 564 | }
|
---|
| 565 |
|
---|
| 566 | my $file_prefix_rec = $shared_fileroot->{$file_prefix};
|
---|
| 567 |
|
---|
| 568 | my $process_exp = $self->{'process_exp'};
|
---|
| 569 |
|
---|
| 570 | if ($file =~ m/$self->{'process_exp'}/) {
|
---|
| 571 | # This is the document the others should be tied to
|
---|
| 572 | $file_prefix_rec->{'tie_to'} = $file_ext;
|
---|
| 573 | }
|
---|
| 574 | else {
|
---|
[9351] | 575 | if (defined $self->{'associate_ext_lookup'}->{$file_ext}) {
|
---|
| 576 | $file_prefix_rec->{'exts'}->{$file_ext} = 1;
|
---|
| 577 | }
|
---|
[8892] | 578 | }
|
---|
| 579 | }
|
---|
| 580 | }
|
---|
[9067] | 581 |
|
---|
| 582 | # now check whether we are actually processing this
|
---|
| 583 | my $filename = $file;
|
---|
| 584 | $filename = &util::filename_cat ($base_dir, $file) if $base_dir =~ /\w/;
|
---|
| 585 | if ($self->{'process_exp'} eq "" || $filename !~ /$self->{'process_exp'}/ || !-f $filename) {
|
---|
| 586 | return undef; # can't recognise
|
---|
| 587 | }
|
---|
[8892] | 588 |
|
---|
[9067] | 589 | # do smart blocking if appropriate
|
---|
[11089] | 590 | if ($self->{'smart_block'}) {
|
---|
[9067] | 591 | $self->store_block_files($filename);
|
---|
| 592 | }
|
---|
[11089] | 593 | # block the cover image if there is one
|
---|
| 594 | if ($self->{'cover_image'}) {
|
---|
| 595 | $self->block_cover_image($filename);
|
---|
| 596 | }
|
---|
[9067] | 597 |
|
---|
| 598 | return 1;
|
---|
[8510] | 599 | }
|
---|
| 600 |
|
---|
[8892] | 601 | sub tie_to_filename
|
---|
| 602 | {
|
---|
| 603 | my $self = shift (@_);
|
---|
| 604 |
|
---|
| 605 | my ($file_ext,$file_prefix_rec) = @_;
|
---|
| 606 |
|
---|
| 607 | if (defined $file_prefix_rec) {
|
---|
| 608 | my $tie_to = $file_prefix_rec->{'tie_to'};
|
---|
[9351] | 609 |
|
---|
[8892] | 610 | if (defined $tie_to) {
|
---|
| 611 | if ($tie_to eq $file_ext) {
|
---|
| 612 | return 1;
|
---|
| 613 | }
|
---|
| 614 | }
|
---|
| 615 | }
|
---|
| 616 |
|
---|
| 617 | return 0;
|
---|
| 618 | }
|
---|
| 619 |
|
---|
| 620 | sub tie_to_assoc_file
|
---|
| 621 | {
|
---|
| 622 | my $self = shift (@_);
|
---|
| 623 | my ($file_ext,$file_prefix_rec) = @_;
|
---|
| 624 |
|
---|
| 625 | if (defined $file_prefix_rec) {
|
---|
| 626 | my $tie_to = $file_prefix_rec->{'tie_to'};
|
---|
| 627 | if (defined $tie_to) {
|
---|
| 628 |
|
---|
| 629 | my $exts = $file_prefix_rec->{'exts'};
|
---|
| 630 |
|
---|
| 631 | my $has_file_ext = $exts->{$file_ext};
|
---|
| 632 |
|
---|
| 633 | if ($has_file_ext) {
|
---|
| 634 | return 1;
|
---|
| 635 | }
|
---|
| 636 | }
|
---|
| 637 | }
|
---|
| 638 |
|
---|
| 639 | return 0;
|
---|
| 640 | }
|
---|
| 641 |
|
---|
| 642 |
|
---|
| 643 | sub associate_with
|
---|
| 644 | {
|
---|
| 645 | my $self = shift (@_);
|
---|
| 646 | my ($file, $filename, $metadata) = @_;
|
---|
| 647 |
|
---|
| 648 | my $associate_ext = $self->{'associate_ext'};
|
---|
| 649 |
|
---|
[9351] | 650 |
|
---|
[8892] | 651 | return 0 if (!$associate_ext);
|
---|
| 652 |
|
---|
| 653 | # If file, see if matches with "tie_to" doc or is one of the
|
---|
| 654 | # associated filename extensions.
|
---|
| 655 |
|
---|
| 656 | my ($file_prefix,$file_ext) = ($file =~ m/^(.*)\.(.*?)$/);
|
---|
| 657 | if ((defined $file_prefix) && (defined $file_ext)) {
|
---|
| 658 |
|
---|
| 659 | my $file_prefix_rec = $self->{'shared_fileroot'}->{$file_prefix};
|
---|
[9351] | 660 |
|
---|
[8892] | 661 | if ($self->tie_to_filename($file_ext,$file_prefix_rec)) {
|
---|
| 662 |
|
---|
| 663 | # Set up gsdlassocfile_tobe
|
---|
| 664 |
|
---|
| 665 | my $exts = $file_prefix_rec->{'exts'};
|
---|
| 666 |
|
---|
| 667 | if (!defined $metadata->{'gsdlassocfile_tobe'}) {
|
---|
| 668 | $metadata->{'gsdlassocfile_tobe'} = [];
|
---|
| 669 | }
|
---|
| 670 |
|
---|
| 671 | my $assoc_tobe = $metadata->{'gsdlassocfile_tobe'};
|
---|
| 672 |
|
---|
| 673 | my ($full_prefix) = ($filename =~ m/^(.*)\..*?$/);
|
---|
| 674 | foreach my $e (keys %$exts) {
|
---|
| 675 | my $assoc_file = "$full_prefix.$e";
|
---|
| 676 | my $mime_type = ""; # let system auto detect this
|
---|
| 677 | push(@$assoc_tobe,"$assoc_file:$mime_type:");
|
---|
| 678 | }
|
---|
| 679 | }
|
---|
| 680 | elsif ($self->tie_to_assoc_file($file_ext,$file_prefix_rec)) {
|
---|
| 681 | # a form of smart block
|
---|
| 682 |
|
---|
| 683 | return 1;
|
---|
| 684 | }
|
---|
| 685 | }
|
---|
| 686 |
|
---|
| 687 | return 0;
|
---|
| 688 | }
|
---|
| 689 |
|
---|
| 690 |
|
---|
[10280] | 691 | sub read_block {
|
---|
[1954] | 692 | my $self = shift (@_);
|
---|
| 693 |
|
---|
[9853] | 694 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
[4] | 695 |
|
---|
[8908] | 696 |
|
---|
[2795] | 697 | my $filename = $file;
|
---|
| 698 | $filename = &util::filename_cat ($base_dir, $file) if $base_dir =~ /\w/;
|
---|
| 699 |
|
---|
[8892] | 700 | if ($self->associate_with($file,$filename,$metadata)) {
|
---|
| 701 | # a form of smart block
|
---|
| 702 | $self->{'num_blocked'} ++;
|
---|
[10280] | 703 | return (0,undef); # blocked
|
---|
[8892] | 704 | }
|
---|
[8510] | 705 |
|
---|
[10280] | 706 | my $smart_block = $self->{'smart_block'};
|
---|
| 707 | my $smart_block_BN = $self->{'smart_block_BN'};
|
---|
[11089] | 708 |
|
---|
[8915] | 709 | if ($smart_block || $smart_block_BN) {
|
---|
[8510] | 710 | if (defined $self->{'file_blocks'}->{$filename} && $self->{'file_blocks'}->{$filename} == 1){
|
---|
[8892] | 711 | $self->{'num_blocked'} ++;
|
---|
[10280] | 712 | return (0,undef); # blocked
|
---|
[8510] | 713 | }
|
---|
[11089] | 714 | } else {
|
---|
| 715 | if ($self->{'block_exp'} ne "" && $filename =~ /$self->{'block_exp'}/) {
|
---|
| 716 | $self->{'num_blocked'} ++;
|
---|
| 717 | return (0,undef); # blocked
|
---|
| 718 | }
|
---|
| 719 | if ($self->{'cover_image'}) {
|
---|
| 720 | if (defined $self->{'file_blocks'}->{$filename} && $self->{'file_blocks'}->{$filename} == 1){
|
---|
| 721 | $self->{'num_blocked'} ++;
|
---|
| 722 | return (0,undef); # blocked
|
---|
| 723 | }
|
---|
| 724 | }
|
---|
[2785] | 725 | }
|
---|
[1242] | 726 | if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) {
|
---|
[10280] | 727 | return (undef,undef); # can't recognise
|
---|
[1242] | 728 | }
|
---|
[10280] | 729 |
|
---|
| 730 | return (1,$filename);
|
---|
| 731 | }
|
---|
| 732 |
|
---|
| 733 | sub read_tidy_file {
|
---|
| 734 |
|
---|
| 735 | my $self = shift (@_);
|
---|
| 736 |
|
---|
| 737 | my ($file) = @_;
|
---|
| 738 |
|
---|
[1242] | 739 | $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
|
---|
[1844] | 740 |
|
---|
[10280] | 741 | return $file;
|
---|
| 742 | }
|
---|
| 743 |
|
---|
| 744 |
|
---|
| 745 |
|
---|
| 746 | # The BasPlug read_into_doc_obj() function. This function does all the
|
---|
| 747 | # right things to make general options work for a given plugin. It reads in
|
---|
| 748 | # a file and sets up a slew of metadata all saved in doc_obj, which
|
---|
| 749 | # it then returns as part of a tuple (process_status,doc_obj)
|
---|
| 750 | #
|
---|
| 751 | # Much of this functionality used to reside in read, but it was broken
|
---|
| 752 | # down into a supporting routine to make the code more flexible.
|
---|
| 753 | #
|
---|
| 754 | # recursive plugins (e.g. RecPlug) and specialized plugins like those
|
---|
| 755 | # capable of processing many documents within a single file (e.g.
|
---|
| 756 | # GMLPlug) will normally want to implement their own version of
|
---|
| 757 | # read_into_doc_obj()
|
---|
| 758 | #
|
---|
| 759 | # Note that $base_dir might be "" and that $file might
|
---|
| 760 | # include directories
|
---|
| 761 | sub read_into_doc_obj {
|
---|
| 762 | my $self = shift (@_);
|
---|
| 763 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
| 764 |
|
---|
| 765 | if ($self->is_recursive()) {
|
---|
| 766 | gsprintf(STDERR, "{BasPlug.read_must_be_implemented}") && die "\n";
|
---|
| 767 | }
|
---|
| 768 |
|
---|
| 769 | my $outhandle = $self->{'outhandle'};
|
---|
| 770 |
|
---|
| 771 | my ($block_status,$filename) = $self->read_block(@_);
|
---|
| 772 | return $block_status if ((!defined $block_status) || ($block_status==0));
|
---|
| 773 | $file = $self->read_tidy_file($file);
|
---|
| 774 |
|
---|
[2811] | 775 | # Do encoding stuff
|
---|
| 776 | my ($language, $encoding) = $self->textcat_get_language_encoding ($filename);
|
---|
[1844] | 777 |
|
---|
[1242] | 778 | # create a new document
|
---|
[1379] | 779 | my $doc_obj = new doc ($filename, "indexed_doc");
|
---|
[2327] | 780 | $doc_obj->set_OIDtype ($processor->{'OIDtype'});
|
---|
[1844] | 781 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Language", $language);
|
---|
[1868] | 782 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Encoding", $encoding);
|
---|
[7508] | 783 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Plugin", "$self->{'plugin_type'}");
|
---|
[8166] | 784 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "FileSize", (-s $filename));
|
---|
| 785 |
|
---|
[2796] | 786 | my ($filemeta) = $file =~ /([^\\\/]+)$/;
|
---|
[4845] | 787 | # how do we know what encoding the filename is in?
|
---|
| 788 | $doc_obj->add_metadata($doc_obj->get_top_section(), "Source", &ghtml::dmsafe($filemeta));
|
---|
[2816] | 789 | if ($self->{'cover_image'}) {
|
---|
| 790 | $self->associate_cover_image($doc_obj, $filename);
|
---|
| 791 | }
|
---|
[1242] | 792 |
|
---|
| 793 | # read in file ($text will be in utf8)
|
---|
| 794 | my $text = "";
|
---|
[2734] | 795 | $self->read_file ($filename, $encoding, $language, \$text);
|
---|
[1242] | 796 |
|
---|
[1844] | 797 | if (!length ($text)) {
|
---|
[2811] | 798 | my $plugin_name = ref ($self);
|
---|
[9586] | 799 | if ($gli) {
|
---|
| 800 | print STDERR "<ProcessingError n='$file' r='File contains no text'>\n";
|
---|
| 801 | }
|
---|
[9703] | 802 | gsprintf($outhandle, "$plugin_name: {BasPlug.file_has_no_text}\n", $filename) if $self->{'verbosity'};
|
---|
[2785] | 803 |
|
---|
| 804 | my $failhandle = $self->{'failhandle'};
|
---|
[9413] | 805 | gsprintf($failhandle, "$file: " . ref($self) . ": {BasPlug.empty_file}\n");
|
---|
[5681] | 806 | # print $failhandle "$file: " . ref($self) . ": file contains no text\n";
|
---|
[2785] | 807 | $self->{'num_not_processed'} ++;
|
---|
| 808 |
|
---|
[10280] | 809 | return (0,undef); # what should we return here?? error but don't want to pass it on
|
---|
[1242] | 810 | }
|
---|
[1954] | 811 |
|
---|
[1242] | 812 | # include any metadata passed in from previous plugins
|
---|
| 813 | # note that this metadata is associated with the top level section
|
---|
[8510] | 814 |
|
---|
[1242] | 815 | $self->extra_metadata ($doc_obj, $doc_obj->get_top_section(), $metadata);
|
---|
[9398] | 816 |
|
---|
[8716] | 817 | # do plugin specific processing of doc_obj
|
---|
| 818 | unless (defined ($self->process (\$text, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli))) {
|
---|
| 819 | $text = '';
|
---|
| 820 | undef $text;
|
---|
[9584] | 821 | print STDERR "<ProcessingError n='$file'>\n" if ($gli);
|
---|
[10280] | 822 | return (-1,undef);
|
---|
[8716] | 823 | }
|
---|
| 824 | $text='';
|
---|
| 825 | undef $text;
|
---|
[9398] | 826 |
|
---|
[1242] | 827 | # do any automatic metadata extraction
|
---|
| 828 | $self->auto_extract_metadata ($doc_obj);
|
---|
[1954] | 829 |
|
---|
[1242] | 830 | # add an OID
|
---|
[3515] | 831 | # see if there is a plugin-specific set_OID function...
|
---|
[8716] | 832 | if (defined ($self->can('set_OID'))) {
|
---|
[3515] | 833 | # it will need $doc_obj to set the Identifier metadata...
|
---|
| 834 | $self->set_OID($doc_obj);
|
---|
| 835 | } else {
|
---|
| 836 | # use the default set_OID() in doc.pm
|
---|
| 837 | $doc_obj->set_OID();
|
---|
| 838 | }
|
---|
[10280] | 839 |
|
---|
| 840 | return (1,$doc_obj);
|
---|
| 841 | }
|
---|
[1242] | 842 |
|
---|
| 843 |
|
---|
[10280] | 844 | # The BasPlug read() function. This function calls read_into_doc_obj()
|
---|
| 845 | # to ensure all the right things to make general options work for a
|
---|
| 846 | # given plugin are done. It then calls the process() function which
|
---|
| 847 | # does all the work specific to a plugin (like the old read functions
|
---|
| 848 | # used to do). Most plugins should define their own process() function
|
---|
| 849 | # and let this read() function keep control.
|
---|
| 850 | #
|
---|
| 851 | # recursive plugins (e.g. RecPlug) and specialized plugins like those
|
---|
| 852 | # capable of processing many documents within a single file (e.g.
|
---|
| 853 | # GMLPlug) might want to implement their own version of read(), but
|
---|
| 854 | # more likely need to implement their own version of read_into_doc_obj()
|
---|
| 855 | #
|
---|
| 856 | # Return number of files processed, undef if can't recognise, -1 if can't
|
---|
| 857 | # process
|
---|
| 858 |
|
---|
| 859 | sub read {
|
---|
| 860 | my $self = shift (@_);
|
---|
| 861 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $total_count, $gli) = @_;
|
---|
| 862 |
|
---|
| 863 | my ($process_status,$doc_obj) = $self->read_into_doc_obj(@_);
|
---|
| 864 |
|
---|
| 865 | if ((defined $process_status) && ($process_status == 1)) {
|
---|
| 866 | # process the document
|
---|
| 867 | $processor->process($doc_obj);
|
---|
| 868 |
|
---|
| 869 | if(defined($self->{'places_filename'})){
|
---|
| 870 | &util::rm($self->{'places_filename'});
|
---|
| 871 | $self->{'places_filename'} = undef;
|
---|
| 872 | }
|
---|
| 873 |
|
---|
| 874 | $self->{'num_processed'} ++;
|
---|
| 875 | undef $doc_obj;
|
---|
[9398] | 876 | }
|
---|
| 877 |
|
---|
[10280] | 878 | # if process_status == 1, then the file has been processed.
|
---|
| 879 | return $process_status;
|
---|
| 880 |
|
---|
[4] | 881 | }
|
---|
| 882 |
|
---|
[1244] | 883 | # returns undef if file is rejected by the plugin
|
---|
[1242] | 884 | sub process {
|
---|
| 885 | my $self = shift (@_);
|
---|
[11089] | 886 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli) = @_;
|
---|
[1242] | 887 |
|
---|
[9413] | 888 | gsprintf(STDERR, "BasPlug::process {common.must_be_implemented}\n") && die "\n";
|
---|
[5681] | 889 | # die "Basplug::process function must be implemented in sub-class\n";
|
---|
[1244] | 890 |
|
---|
| 891 | return undef; # never gets here
|
---|
[1242] | 892 | }
|
---|
| 893 |
|
---|
[1219] | 894 | # uses the multiread package to read in the entire file pointed to
|
---|
| 895 | # by filename and loads the resulting text into $$textref. Input text
|
---|
| 896 | # may be in any of the encodings handled by multiread, output text
|
---|
| 897 | # will be in utf8
|
---|
| 898 | sub read_file {
|
---|
| 899 | my $self = shift (@_);
|
---|
[2734] | 900 | my ($filename, $encoding, $language, $textref) = @_;
|
---|
[4] | 901 |
|
---|
[1756] | 902 | if (!-r $filename)
|
---|
| 903 | {
|
---|
[1844] | 904 | my $outhandle = $self->{'outhandle'};
|
---|
[9413] | 905 | gsprintf($outhandle, "{BasPlug.read_denied}\n", $filename) if $self->{'verbosity'};
|
---|
[5681] | 906 | # print $outhandle "Read permission denied for $filename\n" if $self->{'verbosity'};
|
---|
[1756] | 907 | return;
|
---|
| 908 | }
|
---|
[1219] | 909 | $$textref = "";
|
---|
[9413] | 910 | if (!open (FILE, $filename)) {
|
---|
| 911 | gsprintf(STDERR, "BasPlug::read_file {BasPlug.could_not_open_for_reading} ($!)\n", $filename);
|
---|
[10280] | 912 | die "\n";
|
---|
| 913 | }
|
---|
| 914 |
|
---|
[1844] | 915 | if ($encoding eq "ascii") {
|
---|
[1219] | 916 | undef $/;
|
---|
| 917 | $$textref = <FILE>;
|
---|
| 918 | $/ = "\n";
|
---|
| 919 | } else {
|
---|
| 920 | my $reader = new multiread();
|
---|
| 921 | $reader->set_handle ('BasPlug::FILE');
|
---|
[1844] | 922 | $reader->set_encoding ($encoding);
|
---|
[1219] | 923 | $reader->read_file ($textref);
|
---|
[10280] | 924 | #Now segments chinese if the separate_cjk option is set
|
---|
[6584] | 925 | if ($self->{'separate_cjk'}) {
|
---|
[1219] | 926 | # segment the Chinese words
|
---|
| 927 | $$textref = &cnseg::segment($$textref);
|
---|
| 928 | }
|
---|
| 929 | }
|
---|
[10280] | 930 | close FILE;
|
---|
| 931 | }
|
---|
[1219] | 932 |
|
---|
[10280] | 933 | # write_file -- used by ConvertToPlug, for example in post processing
|
---|
| 934 | #
|
---|
| 935 | sub utf8_write_file {
|
---|
| 936 | my $self = shift (@_);
|
---|
| 937 | my ($textref, $filename) = @_;
|
---|
| 938 |
|
---|
| 939 | if (!open (FILE, ">$filename")) {
|
---|
| 940 | gsprintf(STDERR, "ConvertToPlug::write_file {ConvertToPlug.could_not_open_for_writing} ($!)\n", $filename);
|
---|
| 941 | die "\n";
|
---|
| 942 | }
|
---|
| 943 | print FILE $$textref;
|
---|
| 944 |
|
---|
[1219] | 945 | close FILE;
|
---|
| 946 | }
|
---|
| 947 |
|
---|
[10280] | 948 |
|
---|
[7504] | 949 | sub filename_based_title
|
---|
| 950 | {
|
---|
| 951 | my $self = shift (@_);
|
---|
| 952 | my ($file) = @_;
|
---|
| 953 |
|
---|
| 954 | my $file_derived_title = $file;
|
---|
| 955 | $file_derived_title =~ s/_/ /g;
|
---|
| 956 | $file_derived_title =~ s/\..*?$//;
|
---|
| 957 |
|
---|
| 958 | return $file_derived_title;
|
---|
| 959 | }
|
---|
| 960 |
|
---|
[9398] | 961 |
|
---|
[7504] | 962 | sub title_fallback
|
---|
| 963 | {
|
---|
| 964 | my $self = shift (@_);
|
---|
| 965 | my ($doc_obj,$section,$file) = @_;
|
---|
| 966 |
|
---|
| 967 | if (!defined $doc_obj->get_metadata_element ($section, "Title")) {
|
---|
| 968 |
|
---|
| 969 | my $file_derived_title = $self->filename_based_title($file);
|
---|
| 970 | $doc_obj->add_metadata ($section, "Title", $file_derived_title);
|
---|
| 971 | }
|
---|
| 972 | }
|
---|
| 973 |
|
---|
[2811] | 974 | sub textcat_get_language_encoding {
|
---|
| 975 | my $self = shift (@_);
|
---|
| 976 | my ($filename) = @_;
|
---|
| 977 |
|
---|
[10280] | 978 |
|
---|
[2811] | 979 | my ($language, $encoding, $extracted_encoding);
|
---|
| 980 | if ($self->{'input_encoding'} eq "auto") {
|
---|
| 981 | # use textcat to automatically work out the input encoding and language
|
---|
| 982 | ($language, $encoding) = $self->get_language_encoding ($filename);
|
---|
| 983 | } elsif ($self->{'extract_language'}) {
|
---|
| 984 | # use textcat to get language metadata
|
---|
| 985 | ($language, $extracted_encoding) = $self->get_language_encoding ($filename);
|
---|
| 986 | $encoding = $self->{'input_encoding'};
|
---|
[7644] | 987 | # don't print this message for english... english in utf8 is identical
|
---|
| 988 | # to english in iso-8859-1 (except for some punctuation). We don't have
|
---|
| 989 | # a language model for en_utf8, so textcat always says iso-8859-1!
|
---|
| 990 | if ($extracted_encoding ne $encoding && $language ne "en"
|
---|
| 991 | && $self->{'verbosity'}) {
|
---|
[2811] | 992 | my $plugin_name = ref ($self);
|
---|
| 993 | my $outhandle = $self->{'outhandle'};
|
---|
[9413] | 994 | gsprintf($outhandle, "$plugin_name: {BasPlug.wrong_encoding}\n", $filename, $encoding, $extracted_encoding);
|
---|
[5681] | 995 | # print $outhandle "$plugin_name: WARNING: $filename was read using $encoding encoding but ";
|
---|
| 996 | # print $outhandle "appears to be encoded as $extracted_encoding.\n";
|
---|
[2811] | 997 | }
|
---|
| 998 | } else {
|
---|
| 999 | $language = $self->{'default_language'};
|
---|
| 1000 | $encoding = $self->{'input_encoding'};
|
---|
| 1001 | }
|
---|
[10280] | 1002 |
|
---|
[2811] | 1003 | return ($language, $encoding);
|
---|
| 1004 | }
|
---|
| 1005 |
|
---|
[1844] | 1006 | # Uses textcat to work out the encoding and language of the text in
|
---|
| 1007 | # $filename. All html tags are removed before processing.
|
---|
| 1008 | # returns an array containing "language" and "encoding"
|
---|
| 1009 | sub get_language_encoding {
|
---|
| 1010 | my $self = shift (@_);
|
---|
| 1011 | my ($filename) = @_;
|
---|
| 1012 | my $outhandle = $self->{'outhandle'};
|
---|
[9413] | 1013 | my $unicode_format = "";
|
---|
[1844] | 1014 | # read in file
|
---|
[9413] | 1015 | open (FILE, $filename) || (gsprintf(STDERR, "BasPlug::get_language_encoding {BasPlug.could_not_open_for_reading} ($!)\n", $filename) && die "\n"); # die "BasPlug::get_language_encoding could not open $filename for reading ($!)\n";
|
---|
[1844] | 1016 | undef $/;
|
---|
| 1017 | my $text = <FILE>;
|
---|
| 1018 | $/ = "\n";
|
---|
| 1019 | close FILE;
|
---|
| 1020 |
|
---|
[9413] | 1021 | # check if first few bytes have a Byte Order Marker
|
---|
| 1022 | my $bom=substr($text,0,2); # check 16bit unicode
|
---|
| 1023 | if ($bom eq "\xff\xfe") { # little endian 16bit unicode
|
---|
| 1024 | $unicode_format="unicode";
|
---|
| 1025 | } elsif ($bom eq "\xfe\xff") { # big endian 16bit unicode
|
---|
| 1026 | $unicode_format="unicode";
|
---|
| 1027 | } else {
|
---|
| 1028 | $bom=substr($text,0,3); # check utf-8
|
---|
| 1029 | if ($bom eq "\xef\xbb\xbf") { # utf-8 coded FEFF bom
|
---|
| 1030 | $unicode_format="utf8";
|
---|
| 1031 | # } elsif ($bom eq "\xef\xbf\xbe") { # utf-8 coded FFFE bom. Error!?
|
---|
| 1032 | # $unicode_format="utf8";
|
---|
| 1033 | }
|
---|
| 1034 | }
|
---|
[10442] | 1035 |
|
---|
| 1036 | # VB scripting generated Word to HTML file
|
---|
[10446] | 1037 | if ($text =~ /charset=(windows.*)[\"]/ig){
|
---|
[10442] | 1038 | my $vbhtml_encoding = $1;
|
---|
| 1039 | $vbhtml_encoding =~ s/-+/_/g;
|
---|
| 1040 | $self->{'input_encoding'} = $vbhtml_encoding;
|
---|
| 1041 | }
|
---|
| 1042 |
|
---|
[1999] | 1043 | # remove <title>stuff</title> -- as titles tend often to be in English
|
---|
| 1044 | # for foreign language documents
|
---|
[8818] | 1045 | $text =~ s/<title>(.|\n)*?<\/title>//i;
|
---|
[1999] | 1046 |
|
---|
[1844] | 1047 | # remove all HTML tags
|
---|
[9413] | 1048 | # XXX this doesn't match plugins derived from HTMLPlug (except ConvertTo)
|
---|
| 1049 | if (ref($self) eq 'HTMLPlug' ||
|
---|
| 1050 | (exists $self->{'converted_to'} && $self->{'converted_to'} eq 'HTML')){
|
---|
| 1051 | $text =~ s/<[^>]*>//sg;
|
---|
| 1052 | }
|
---|
[1844] | 1053 |
|
---|
| 1054 | # get the language/encoding
|
---|
[2235] | 1055 | my $results = $self->{'textcat'}->classify(\$text);
|
---|
[1844] | 1056 |
|
---|
[1903] | 1057 | # if textcat returns 3 or less possibilities we'll use the
|
---|
| 1058 | # first one in the list - otherwise use the defaults
|
---|
[2235] | 1059 | if (scalar @$results > 3) {
|
---|
[3731] | 1060 | my $best_encoding="";
|
---|
[9413] | 1061 | if ($unicode_format) { # in case the first had a BOM
|
---|
| 1062 | $best_encoding=$unicode_format;
|
---|
| 1063 | } else {
|
---|
| 1064 | my %guessed_encodings = ();
|
---|
| 1065 | foreach my $result (@$results) {
|
---|
| 1066 | $result =~ /([^\-]+)$/;
|
---|
| 1067 | my $enc=$1;
|
---|
| 1068 | if (!defined($guessed_encodings{$enc})) {
|
---|
| 1069 | $guessed_encodings{$enc}=0;
|
---|
| 1070 | }
|
---|
| 1071 | $guessed_encodings{$enc}++;
|
---|
[3731] | 1072 | }
|
---|
[9413] | 1073 |
|
---|
| 1074 | $guessed_encodings{""}=-1; # for default best_encoding of ""
|
---|
| 1075 | foreach my $enc (keys %guessed_encodings) {
|
---|
| 1076 | if ($guessed_encodings{$enc} >
|
---|
| 1077 | $guessed_encodings{$best_encoding}){
|
---|
| 1078 | $best_encoding=$enc;
|
---|
| 1079 | }
|
---|
| 1080 | }
|
---|
[3731] | 1081 | }
|
---|
| 1082 |
|
---|
[1844] | 1083 | if ($self->{'input_encoding'} ne 'auto') {
|
---|
[9961] | 1084 | if ($self->{'extract_language'} && ($self->{'verbosity'}>2)) {
|
---|
[9413] | 1085 | gsprintf($outhandle,
|
---|
| 1086 | "BasPlug: {BasPlug.could_not_extract_language}\n",
|
---|
| 1087 | $filename, $self->{'default_language'});
|
---|
[1844] | 1088 | }
|
---|
| 1089 | return ($self->{'default_language'}, $self->{'input_encoding'});
|
---|
| 1090 |
|
---|
| 1091 | } else {
|
---|
[9961] | 1092 | if ($self->{'verbosity'}>2) {
|
---|
[9413] | 1093 | gsprintf($outhandle,
|
---|
| 1094 | "BasPlug: {BasPlug.could_not_extract_language}\n",
|
---|
| 1095 | $filename, $self->{'default_language'});
|
---|
[1844] | 1096 | }
|
---|
[3731] | 1097 | return ($self->{'default_language'}, $best_encoding);
|
---|
[1844] | 1098 | }
|
---|
| 1099 | }
|
---|
| 1100 |
|
---|
| 1101 | # format language/encoding
|
---|
[2235] | 1102 | my ($language, $encoding) = $results->[0] =~ /^([^-]*)(?:-(.*))?$/;
|
---|
[1870] | 1103 | if (!defined $language) {
|
---|
[9961] | 1104 | if ($self->{'verbosity'}>2) {
|
---|
[9413] | 1105 | gsprintf($outhandle,
|
---|
| 1106 | "BasPlug: {BasPlug.could_not_extract_language}\n",
|
---|
| 1107 | $filename, $self->{'default_language'});
|
---|
[1870] | 1108 | }
|
---|
| 1109 | $language = $self->{'default_language'};
|
---|
| 1110 | }
|
---|
[1844] | 1111 | if (!defined $encoding) {
|
---|
[9961] | 1112 | if ($self->{'verbosity'}>2) {
|
---|
[9413] | 1113 | gsprintf($outhandle,
|
---|
| 1114 | "BasPlug: {BasPlug.could_not_extract_encoding}\n",
|
---|
| 1115 | $filename, $self->{'default_encoding'});
|
---|
[1870] | 1116 | }
|
---|
| 1117 | $encoding = $self->{'default_encoding'};
|
---|
[1844] | 1118 | }
|
---|
| 1119 |
|
---|
[7818] | 1120 |
|
---|
| 1121 | # check for equivalents where textcat doesn't have some encodings...
|
---|
| 1122 | # eg MS versions of standard encodings
|
---|
| 1123 | if ($encoding =~ /^iso_8859_(\d+)/) {
|
---|
| 1124 | my $iso = $1; # which variant of the iso standard?
|
---|
| 1125 | # iso-8859 sets don't use chars 0x80-0x9f, windows codepages do
|
---|
| 1126 | if ($text =~ /[\x80-\x9f]/) {
|
---|
| 1127 | # Western Europe
|
---|
| 1128 | if ($iso == 1 or $iso == 15) { $encoding = 'windows_1252' }
|
---|
| 1129 | elsif ($iso == 2) { $encoding = 'windows_1250' } # Central Europe
|
---|
| 1130 | elsif ($iso == 5) { $encoding = 'windows_1251' } # Cyrillic
|
---|
| 1131 | elsif ($iso == 6) { $encoding = 'windows_1256' } # Arabic
|
---|
| 1132 | elsif ($iso == 7) { $encoding = 'windows_1253' } # Greek
|
---|
| 1133 | elsif ($iso == 8) { $encoding = 'windows_1255' } # Hebrew
|
---|
| 1134 | elsif ($iso == 9) { $encoding = 'windows_1254' } # Turkish
|
---|
| 1135 | }
|
---|
| 1136 | }
|
---|
| 1137 |
|
---|
[1870] | 1138 | if ($encoding !~ /^(ascii|utf8|unicode)$/ &&
|
---|
| 1139 | !defined $encodings::encodings->{$encoding}) {
|
---|
[1844] | 1140 | if ($self->{'verbosity'}) {
|
---|
[7818] | 1141 | gsprintf($outhandle, "BasPlug: {BasPlug.unsupported_encoding}\n",
|
---|
| 1142 | $filename, $encoding, $self->{'default_encoding'});
|
---|
[1844] | 1143 | }
|
---|
| 1144 | $encoding = $self->{'default_encoding'};
|
---|
| 1145 | }
|
---|
| 1146 |
|
---|
| 1147 | return ($language, $encoding);
|
---|
| 1148 | }
|
---|
| 1149 |
|
---|
[1219] | 1150 | # add any extra metadata that's been passed around from one
|
---|
| 1151 | # plugin to another.
|
---|
| 1152 | # extra_metadata uses add_utf8_metadata so it expects metadata values
|
---|
| 1153 | # to already be in utf8
|
---|
| 1154 | sub extra_metadata {
|
---|
| 1155 | my $self = shift (@_);
|
---|
| 1156 | my ($doc_obj, $cursection, $metadata) = @_;
|
---|
| 1157 |
|
---|
| 1158 | foreach my $field (keys(%$metadata)) {
|
---|
[839] | 1159 | # $metadata->{$field} may be an array reference
|
---|
[8510] | 1160 | if ($field eq "gsdlassocfile_tobe") {
|
---|
| 1161 | # 'gsdlassocfile_tobe' is artificially introduced metadata
|
---|
| 1162 | # that is used to signal that certain additional files should
|
---|
| 1163 | # be tied to this document. Useful in situations where a
|
---|
| 1164 | # metadata pass in the plugin pipeline works out some files
|
---|
| 1165 | # need to be associated with a document, but the document hasn't
|
---|
| 1166 | # been formed yet.
|
---|
| 1167 |
|
---|
| 1168 | my $equiv_form = "";
|
---|
| 1169 | foreach my $gaf (@{$metadata->{$field}}) {
|
---|
| 1170 | my ($full_filename,$mimetype) = ($gaf =~ m/^(.*):(.*):$/);
|
---|
| 1171 | my ($tail_filename) = ($full_filename =~ /^.*[\/\\](.+?)$/);
|
---|
| 1172 | my $filename = $full_filename;
|
---|
| 1173 |
|
---|
| 1174 | $doc_obj->associate_file($full_filename,$tail_filename,$mimetype);
|
---|
| 1175 |
|
---|
| 1176 | my ($doc_ext) = ($tail_filename =~ m/^.*\.(.*)$/);
|
---|
[9351] | 1177 | my $start_doclink = "<a href=\"_httpcollection_/index/assoc/{Or}{[parent(Top):archivedir],[archivedir]}/$tail_filename\">";
|
---|
[8510] | 1178 | my $srcicon = "_icon".$doc_ext."_";
|
---|
| 1179 | my $end_doclink = "</a>";
|
---|
| 1180 |
|
---|
| 1181 | $equiv_form .= " $start_doclink\{If\}{$srcicon,$srcicon,$doc_ext\}$end_doclink";
|
---|
| 1182 | }
|
---|
| 1183 | $doc_obj->add_utf8_metadata ($cursection, "equivlink", $equiv_form);
|
---|
| 1184 | }
|
---|
| 1185 | elsif (ref ($metadata->{$field}) eq "ARRAY") {
|
---|
[839] | 1186 | map {
|
---|
[1219] | 1187 | $doc_obj->add_utf8_metadata ($cursection, $field, $_);
|
---|
[839] | 1188 | } @{$metadata->{$field}};
|
---|
| 1189 | } else {
|
---|
[1219] | 1190 | $doc_obj->add_utf8_metadata ($cursection, $field, $metadata->{$field});
|
---|
[839] | 1191 | }
|
---|
| 1192 | }
|
---|
| 1193 | }
|
---|
| 1194 |
|
---|
[1396] | 1195 | # initialise metadata extractors
|
---|
| 1196 | sub initialise_extractors {
|
---|
| 1197 | my $self = shift (@_);
|
---|
| 1198 |
|
---|
| 1199 | if ($self->{'extract_acronyms'} || $self->{'markup_acronyms'}) {
|
---|
| 1200 | &acronym::initialise_acronyms();
|
---|
| 1201 | }
|
---|
| 1202 | }
|
---|
| 1203 |
|
---|
| 1204 | # finalise metadata extractors
|
---|
| 1205 | sub finalise_extractors {
|
---|
| 1206 | my $self = shift (@_);
|
---|
| 1207 |
|
---|
| 1208 | if ($self->{'extract_acronyms'} || $self->{'markup_acronyms'}) {
|
---|
| 1209 | &acronym::finalise_acronyms();
|
---|
| 1210 | }
|
---|
| 1211 | }
|
---|
| 1212 |
|
---|
[1602] | 1213 | # FIRSTNNN: extract the first NNN characters as metadata
|
---|
| 1214 | sub extract_first_NNNN_characters {
|
---|
| 1215 | my $self = shift (@_);
|
---|
| 1216 | my ($textref, $doc_obj, $thissection) = @_;
|
---|
| 1217 |
|
---|
| 1218 | foreach my $size (split /,/, $self->{'first'}) {
|
---|
| 1219 | my $tmptext = $$textref;
|
---|
| 1220 | $tmptext =~ s/^\s+//;
|
---|
| 1221 | $tmptext =~ s/\s+$//;
|
---|
| 1222 | $tmptext =~ s/\s+/ /gs;
|
---|
| 1223 | $tmptext = substr ($tmptext, 0, $size);
|
---|
| 1224 | $tmptext =~ s/\s\S*$/…/;
|
---|
| 1225 | $doc_obj->add_utf8_metadata ($thissection, "First$size", $tmptext);
|
---|
| 1226 | }
|
---|
| 1227 | }
|
---|
| 1228 |
|
---|
| 1229 | sub extract_email {
|
---|
| 1230 | my $self = shift (@_);
|
---|
| 1231 | my ($textref, $doc_obj, $thissection) = @_;
|
---|
| 1232 | my $outhandle = $self->{'outhandle'};
|
---|
| 1233 |
|
---|
[9413] | 1234 | gsprintf($outhandle, " {BasPlug.extracting_emails}...\n")
|
---|
[1844] | 1235 | if ($self->{'verbosity'} > 2);
|
---|
[1602] | 1236 |
|
---|
[2604] | 1237 | my @email = ($$textref =~ m/([-a-z0-9\.@+_=]+@(?:[-a-z0-9]+\.)+(?:com|org|edu|mil|int|net|[a-z][a-z]))/g);
|
---|
[1602] | 1238 | @email = sort @email;
|
---|
| 1239 |
|
---|
[10218] | 1240 | # if($self->{"new_extract_email"} == 0)
|
---|
| 1241 | # {
|
---|
| 1242 | # my @email2 = ();
|
---|
| 1243 | # foreach my $address (@email)
|
---|
| 1244 | # {
|
---|
| 1245 | # if (!(join(" ",@email2) =~ m/(^| )$address( |$)/ ))
|
---|
| 1246 | # {
|
---|
| 1247 | # push @email2, $address;
|
---|
| 1248 | # $doc_obj->add_utf8_metadata ($thissection, "emailAddress", $address);
|
---|
| 1249 | # # print $outhandle " extracting $address\n"
|
---|
| 1250 | # &gsprintf($outhandle, " {BasPlug.extracting} $address\n")
|
---|
| 1251 | # if ($self->{'verbosity'} > 3);
|
---|
| 1252 | # }
|
---|
| 1253 | # }
|
---|
| 1254 | # }
|
---|
| 1255 | # else
|
---|
| 1256 | # {
|
---|
| 1257 | my $hashExistMail = {};
|
---|
[1602] | 1258 | foreach my $address (@email) {
|
---|
[10218] | 1259 | if (!(defined $hashExistMail->{$address}))
|
---|
| 1260 | {
|
---|
| 1261 | $hashExistMail->{$address} = 1;
|
---|
[1602] | 1262 | $doc_obj->add_utf8_metadata ($thissection, "emailAddress", $address);
|
---|
[9413] | 1263 | gsprintf($outhandle, " {BasPlug.extracting} $address\n")
|
---|
[1844] | 1264 | if ($self->{'verbosity'} > 3);
|
---|
[1602] | 1265 | }
|
---|
| 1266 | }
|
---|
[9413] | 1267 | gsprintf($outhandle, " {BasPlug.done_email_extract}\n")
|
---|
[1844] | 1268 | if ($self->{'verbosity'} > 2);
|
---|
[1602] | 1269 | }
|
---|
| 1270 |
|
---|
| 1271 | # extract metadata
|
---|
[5681] | 1272 | sub auto_extract_metadata {
|
---|
[1954] | 1273 |
|
---|
[1242] | 1274 | my $self = shift (@_);
|
---|
| 1275 | my ($doc_obj) = @_;
|
---|
[1602] | 1276 |
|
---|
| 1277 | if ($self->{'extract_email'}) {
|
---|
| 1278 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1279 | while (defined $thissection) {
|
---|
| 1280 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1281 | $self->extract_email (\$text, $doc_obj, $thissection) if $text =~ /./;
|
---|
| 1282 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1283 | }
|
---|
[1954] | 1284 | }
|
---|
[9398] | 1285 | if ($self->{'extract_placenames'}) {
|
---|
| 1286 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1287 | while (defined $thissection) {
|
---|
| 1288 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1289 | $self->extract_placenames (\$text, $doc_obj, $thissection) if $text =~ /./;
|
---|
| 1290 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1291 | }
|
---|
| 1292 | }
|
---|
[1954] | 1293 |
|
---|
[11069] | 1294 | if ($self->{'extract_keyphrases'} || $self->{'extract_keyphrases_kea4'}) {
|
---|
| 1295 | $self->extract_keyphrases($doc_obj);
|
---|
| 1296 | }
|
---|
[1954] | 1297 |
|
---|
[1602] | 1298 | if ($self->{'first'}) {
|
---|
| 1299 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1300 | while (defined $thissection) {
|
---|
| 1301 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1302 | $self->extract_first_NNNN_characters (\$text, $doc_obj, $thissection) if $text =~ /./;
|
---|
| 1303 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1304 | }
|
---|
| 1305 | }
|
---|
| 1306 |
|
---|
[1242] | 1307 | if ($self->{'extract_acronyms'}) {
|
---|
| 1308 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1309 | while (defined $thissection) {
|
---|
| 1310 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1311 | $self->extract_acronyms (\$text, $doc_obj, $thissection) if $text =~ /./;
|
---|
| 1312 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1313 | }
|
---|
| 1314 | }
|
---|
[1602] | 1315 |
|
---|
[1393] | 1316 | if ($self->{'markup_acronyms'}) {
|
---|
| 1317 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1318 | while (defined $thissection) {
|
---|
| 1319 | my $text = $doc_obj->get_text($thissection);
|
---|
| 1320 | $text = $self->markup_acronyms ($text, $doc_obj, $thissection);
|
---|
| 1321 | $doc_obj->delete_text($thissection);
|
---|
| 1322 | $doc_obj->add_text($thissection, $text);
|
---|
| 1323 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1324 | }
|
---|
| 1325 | }
|
---|
| 1326 |
|
---|
[10218] | 1327 | if($self->{'extract_historical_years'}) {
|
---|
[1317] | 1328 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1329 | while (defined $thissection) {
|
---|
[10218] | 1330 |
|
---|
[1317] | 1331 | my $text = $doc_obj->get_text($thissection);
|
---|
[1846] | 1332 | &DateExtract::get_date_metadata($text, $doc_obj,
|
---|
| 1333 | $thissection,
|
---|
[10218] | 1334 | $self->{'no_bibliography'},
|
---|
| 1335 | $self->{'maximum_year'},
|
---|
| 1336 | $self->{'maximum_century'});
|
---|
[1317] | 1337 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1338 | }
|
---|
| 1339 | }
|
---|
[1242] | 1340 | }
|
---|
| 1341 |
|
---|
[11069] | 1342 |
|
---|
| 1343 | #adding kea keyphrases
|
---|
| 1344 | sub extract_keyphrases
|
---|
| 1345 | {
|
---|
| 1346 | my $self = shift(@_);
|
---|
| 1347 | my $doc_obj = shift(@_);
|
---|
| 1348 |
|
---|
| 1349 | # Use Kea 3.0 unless 4.0 has been specified
|
---|
| 1350 | my $kea_version = "3.0";
|
---|
| 1351 | if ($self->{'extract_keyphrases_kea4'}) {
|
---|
| 1352 | $kea_version = "4.0";
|
---|
| 1353 | }
|
---|
| 1354 |
|
---|
| 1355 | # Check that Kea exists, and tell the user where to get it if not
|
---|
| 1356 | my $keahome = &Kea::get_Kea_directory($kea_version);
|
---|
| 1357 | if (!-e $keahome) {
|
---|
| 1358 | gsprintf(STDERR, "{BasPlug.missing_kea}\n", $keahome, $kea_version);
|
---|
| 1359 | return;
|
---|
| 1360 | }
|
---|
| 1361 |
|
---|
| 1362 | my $thissection = $doc_obj->get_top_section();
|
---|
| 1363 | my $text = "";
|
---|
| 1364 | my $list;
|
---|
| 1365 |
|
---|
| 1366 | #loop through sections to gather whole doc
|
---|
| 1367 | while (defined $thissection) {
|
---|
| 1368 | my $sectiontext = $doc_obj->get_text($thissection);
|
---|
| 1369 | $text = $text.$sectiontext;
|
---|
| 1370 | $thissection = $doc_obj->get_next_section ($thissection);
|
---|
| 1371 | }
|
---|
| 1372 |
|
---|
| 1373 | if($self->{'extract_keyphrase_options'}) { #if kea options flag is set, call Kea with specified options
|
---|
| 1374 | $list = &Kea::extract_KeyPhrases ($kea_version, $text, $self->{'extract_keyphrase_options'});
|
---|
| 1375 | } else { #otherwise call Kea with no options
|
---|
| 1376 | $list = &Kea::extract_KeyPhrases ($kea_version, $text);
|
---|
| 1377 | }
|
---|
| 1378 |
|
---|
| 1379 | if ($list){
|
---|
| 1380 | # if a list of kea keyphrases was returned (ie not empty)
|
---|
| 1381 | if ($self->{'verbosity'}) {
|
---|
| 1382 | gsprintf(STDERR, "{BasPlug.keyphrases}: $list\n");
|
---|
| 1383 | }
|
---|
| 1384 |
|
---|
| 1385 | #add metadata to top section
|
---|
| 1386 | $thissection = $doc_obj->get_top_section();
|
---|
| 1387 |
|
---|
| 1388 | # add all key phrases as one metadata
|
---|
| 1389 | $doc_obj->add_metadata($thissection, "Keyphrases", $list);
|
---|
| 1390 |
|
---|
| 1391 | # add individual key phrases as multiple metadata
|
---|
| 1392 | foreach my $keyphrase (split(',', $list)) {
|
---|
| 1393 | $keyphrase =~ s/^\s+|\s+$//g;
|
---|
| 1394 | $doc_obj->add_metadata($thissection, "Keyphrase", $keyphrase);
|
---|
| 1395 | }
|
---|
| 1396 | }
|
---|
| 1397 | }
|
---|
| 1398 |
|
---|
| 1399 |
|
---|
[1335] | 1400 | # extract acronyms from a section in a document. progress is
|
---|
[1424] | 1401 | # reported to outhandle based on the verbosity. both the Acronym
|
---|
[1335] | 1402 | # and the AcronymKWIC metadata items are created.
|
---|
| 1403 |
|
---|
[1242] | 1404 | sub extract_acronyms {
|
---|
| 1405 | my $self = shift (@_);
|
---|
| 1406 | my ($textref, $doc_obj, $thissection) = @_;
|
---|
[1424] | 1407 | my $outhandle = $self->{'outhandle'};
|
---|
[1242] | 1408 |
|
---|
[5681] | 1409 | # print $outhandle " extracting acronyms ...\n"
|
---|
[9413] | 1410 | gsprintf($outhandle, " {BasPlug.extracting_acronyms}...\n")
|
---|
[1844] | 1411 | if ($self->{'verbosity'} > 2);
|
---|
[1335] | 1412 |
|
---|
[1242] | 1413 | my $acro_array = &acronym::acronyms($textref);
|
---|
[1360] | 1414 |
|
---|
[1242] | 1415 | foreach my $acro (@$acro_array) {
|
---|
| 1416 |
|
---|
[1335] | 1417 | #check that this is the first time ...
|
---|
| 1418 | my $seen_before = "false";
|
---|
| 1419 | my $previous_data = $doc_obj->get_metadata($thissection, "Acronym");
|
---|
| 1420 | foreach my $thisAcro (@$previous_data) {
|
---|
[1602] | 1421 | if ($thisAcro eq $acro->to_string()) {
|
---|
[1335] | 1422 | $seen_before = "true";
|
---|
[9413] | 1423 | if ($self->{'verbosity'} >= 4) {
|
---|
| 1424 | gsprintf($outhandle, " {BasPlug.already_seen} " .
|
---|
| 1425 | $acro->to_string() . "\n");
|
---|
| 1426 | }
|
---|
[5681] | 1427 | }
|
---|
[1242] | 1428 | }
|
---|
[1335] | 1429 |
|
---|
[1602] | 1430 | if ($seen_before eq "false") {
|
---|
[1393] | 1431 | #write it to the file ...
|
---|
| 1432 | $acro->write_to_file();
|
---|
| 1433 |
|
---|
[1335] | 1434 | #do the normal acronym
|
---|
| 1435 | $doc_obj->add_utf8_metadata($thissection, "Acronym", $acro->to_string());
|
---|
[9413] | 1436 | gsprintf($outhandle, " {BasPlug.adding} ".$acro->to_string()."\n")
|
---|
[1844] | 1437 | if ($self->{'verbosity'} > 3);
|
---|
[1335] | 1438 | }
|
---|
[1242] | 1439 | }
|
---|
[5681] | 1440 |
|
---|
[9413] | 1441 | gsprintf($outhandle, " {BasPlug.done_acronym_extract}\n")
|
---|
[1844] | 1442 | if ($self->{'verbosity'} > 2);
|
---|
[1242] | 1443 | }
|
---|
| 1444 |
|
---|
[1393] | 1445 | sub markup_acronyms {
|
---|
| 1446 | my $self = shift (@_);
|
---|
| 1447 | my ($text, $doc_obj, $thissection) = @_;
|
---|
[1424] | 1448 | my $outhandle = $self->{'outhandle'};
|
---|
[1393] | 1449 |
|
---|
[9413] | 1450 | gsprintf($outhandle, " {BasPlug.marking_up_acronyms}...\n")
|
---|
[1844] | 1451 | if ($self->{'verbosity'} > 2);
|
---|
[1393] | 1452 |
|
---|
| 1453 | #self is passed in to check for verbosity ...
|
---|
| 1454 | $text = &acronym::markup_acronyms($text, $self);
|
---|
| 1455 |
|
---|
[9413] | 1456 | gsprintf($outhandle, " {BasPlug.done_acronym_markup}\n")
|
---|
[1844] | 1457 | if ($self->{'verbosity'} > 2);
|
---|
[1393] | 1458 |
|
---|
| 1459 | return $text;
|
---|
| 1460 | }
|
---|
| 1461 |
|
---|
[2785] | 1462 | sub compile_stats {
|
---|
| 1463 | my $self = shift(@_);
|
---|
| 1464 | my ($stats) = @_;
|
---|
| 1465 |
|
---|
| 1466 | $stats->{'num_processed'} += $self->{'num_processed'};
|
---|
| 1467 | $stats->{'num_not_processed'} += $self->{'num_not_processed'};
|
---|
[2796] | 1468 | $stats->{'num_archives'} += $self->{'num_archives'};
|
---|
[2785] | 1469 |
|
---|
| 1470 | }
|
---|
| 1471 |
|
---|
[2816] | 1472 | sub associate_cover_image {
|
---|
[10833] | 1473 | my $self = shift;
|
---|
[2816] | 1474 | my ($doc_obj, $filename) = @_;
|
---|
| 1475 |
|
---|
[10833] | 1476 | $filename =~ s/\.[^\\\/\.]+$/\.jpg/;
|
---|
| 1477 | if (exists $self->{'covers_missing_cache'}->{$filename}) {
|
---|
| 1478 | # don't stat() for existence eg for multiple document input files
|
---|
| 1479 | # (eg SplitPlug)
|
---|
| 1480 | return;
|
---|
| 1481 | }
|
---|
| 1482 |
|
---|
[9413] | 1483 | my $top_section=$doc_obj->get_top_section();
|
---|
| 1484 |
|
---|
[2816] | 1485 | if (-e $filename) {
|
---|
| 1486 | $doc_obj->associate_file($filename, "cover.jpg", "image/jpeg");
|
---|
[9413] | 1487 | $doc_obj->add_utf8_metadata($top_section, "hascover", 1);
|
---|
[3086] | 1488 | } else {
|
---|
[10833] | 1489 | my $upper_filename = $filename;
|
---|
| 1490 | $upper_filename =~ s/jpg$/JPG/;
|
---|
| 1491 | if (-e $upper_filename) {
|
---|
| 1492 | $doc_obj->associate_file($upper_filename, "cover.jpg",
|
---|
| 1493 | "image/jpeg");
|
---|
[9413] | 1494 | $doc_obj->add_utf8_metadata($top_section, "hascover", 1);
|
---|
[10833] | 1495 | } else {
|
---|
| 1496 | # file doesn't exist, so record the fact that it's missing so
|
---|
| 1497 | # we don't stat() again (stat is slow)
|
---|
| 1498 | $self->{'covers_missing_cache'}->{$filename} = 1;
|
---|
[3086] | 1499 | }
|
---|
[2816] | 1500 | }
|
---|
[10833] | 1501 |
|
---|
[2816] | 1502 | }
|
---|
| 1503 |
|
---|
[4] | 1504 | 1;
|
---|