| 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 | #
|
|---|
| 8 | # Copyright (C) 1999 New Zealand Digital Library Project
|
|---|
| 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 | ###########################################################################
|
|---|
| 25 |
|
|---|
| 26 | package BasPlug;
|
|---|
| 27 |
|
|---|
| 28 | eval {require bytes};
|
|---|
| 29 |
|
|---|
| 30 | # suppress the annoying "subroutine redefined" warning that various
|
|---|
| 31 | # plugins cause under perl 5.6
|
|---|
| 32 | $SIG{__WARN__} = sub {warn($_[0]) unless ($_[0] =~ /Subroutine\s+\S+\sredefined/)};
|
|---|
| 33 |
|
|---|
| 34 | use Kea;
|
|---|
| 35 | use parsargv;
|
|---|
| 36 | use multiread;
|
|---|
| 37 | use encodings;
|
|---|
| 38 | use cnseg;
|
|---|
| 39 | use acronym;
|
|---|
| 40 | use textcat;
|
|---|
| 41 | use doc;
|
|---|
| 42 | eval "require diagnostics"; # some perl distros (eg mac) don't have this
|
|---|
| 43 | use DateExtract;
|
|---|
| 44 | use ghtml;
|
|---|
| 45 | use gsprintf;
|
|---|
| 46 | use printusage;
|
|---|
| 47 |
|
|---|
| 48 |
|
|---|
| 49 | my $unicode_list =
|
|---|
| 50 | [ { 'name' => "auto",
|
|---|
| 51 | 'desc' => "{BasPlug.input_encoding.auto}" },
|
|---|
| 52 | { 'name' => "ascii",
|
|---|
| 53 | 'desc' => "{BasPlug.input_encoding.ascii}" },
|
|---|
| 54 | { 'name' => "utf8",
|
|---|
| 55 | 'desc' => "{BasPlug.input_encoding.utf8}" },
|
|---|
| 56 | { 'name' => "unicode",
|
|---|
| 57 | 'desc' => "{BasPlug.input_encoding.unicode}" } ];
|
|---|
| 58 |
|
|---|
| 59 | my $arguments =
|
|---|
| 60 | [ { 'name' => "process_exp",
|
|---|
| 61 | 'desc' => "{BasPlug.process_exp}",
|
|---|
| 62 | 'type' => "regexp",
|
|---|
| 63 | 'deft' => "",
|
|---|
| 64 | 'reqd' => "no" },
|
|---|
| 65 | { 'name' => "block_exp",
|
|---|
| 66 | 'desc' => "{BasPlug.block_exp}",
|
|---|
| 67 | 'type' => "regexp",
|
|---|
| 68 | 'deft' => "",
|
|---|
| 69 | 'reqd' => "no" },
|
|---|
| 70 | { 'name' => "input_encoding",
|
|---|
| 71 | 'desc' => "{BasPlug.input_encoding}",
|
|---|
| 72 | 'type' => "enum",
|
|---|
| 73 | 'list' => $unicode_list,
|
|---|
| 74 | 'reqd' => "no" ,
|
|---|
| 75 | 'deft' => "auto" } ,
|
|---|
| 76 | { 'name' => "default_encoding",
|
|---|
| 77 | 'desc' => "{BasPlug.default_encoding}",
|
|---|
| 78 | 'type' => "enum",
|
|---|
| 79 | 'list' => $unicode_list,
|
|---|
| 80 | 'reqd' => "no",
|
|---|
| 81 | 'deft' => "utf8" },
|
|---|
| 82 | { 'name' => "extract_language",
|
|---|
| 83 | 'desc' => "{BasPlug.extract_language}",
|
|---|
| 84 | 'type' => "flag",
|
|---|
| 85 | 'reqd' => "no" },
|
|---|
| 86 | { 'name' => "default_language",
|
|---|
| 87 | 'desc' => "{BasPlug.default_language}",
|
|---|
| 88 | 'type' => "language",
|
|---|
| 89 | 'deft' => "en",
|
|---|
| 90 | 'reqd' => "no" },
|
|---|
| 91 | { 'name' => "extract_acronyms",
|
|---|
| 92 | 'desc' => "{BasPlug.extract_acronyms}",
|
|---|
| 93 | 'type' => "flag",
|
|---|
| 94 | 'reqd' => "no" },
|
|---|
| 95 | { 'name' => "markup_acronyms",
|
|---|
| 96 | 'desc' => "{BasPlug.markup_acronyms}",
|
|---|
| 97 | 'type' => "flag",
|
|---|
| 98 | 'reqd' => "no" },
|
|---|
| 99 | { 'name' => "first",
|
|---|
| 100 | 'desc' => "{BasPlug.first}",
|
|---|
| 101 | 'type' => "string",
|
|---|
| 102 | 'reqd' => "no" },
|
|---|
| 103 | { 'name' => "extract_email",
|
|---|
| 104 | 'desc' => "{BasPlug.extract_email}",
|
|---|
| 105 | 'type' => "flag",
|
|---|
| 106 | 'reqd' => "no" },
|
|---|
| 107 | { 'name' => "extract_historical_years",
|
|---|
| 108 | 'desc' => "{BasPlug.extract_historical_years}",
|
|---|
| 109 | 'type' => "flag",
|
|---|
| 110 | 'reqd' => "no" },
|
|---|
| 111 | { 'name' => "maximum_year",
|
|---|
| 112 | 'desc' => "{BasPlug.maximum_year}",
|
|---|
| 113 | 'type' => "int",
|
|---|
| 114 | 'deft' => (localtime)[5]+1900,
|
|---|
| 115 | 'reqd' => "no"},
|
|---|
| 116 | { 'name' => "maximum_century",
|
|---|
| 117 | 'desc' => "{BasPlug.maximum_century}",
|
|---|
| 118 | 'type' => "string",
|
|---|
| 119 | 'deft' => "",
|
|---|
| 120 | 'reqd' => "no" },
|
|---|
| 121 | { 'name' => "no_bibliography",
|
|---|
| 122 | 'desc' => "{BasPlug.no_bibliography}",
|
|---|
| 123 | 'type' => "flag",
|
|---|
| 124 | 'reqd' => "no"},
|
|---|
| 125 | { 'name' => "cover_image",
|
|---|
| 126 | 'desc' => "{BasPlug.cover_image}",
|
|---|
| 127 | 'type' => "flag",
|
|---|
| 128 | 'reqd' => "no" } ];
|
|---|
| 129 |
|
|---|
| 130 | my $options = { 'name' => "BasPlug",
|
|---|
| 131 | 'desc' => "{BasPlug.desc}",
|
|---|
| 132 | 'abstract' => "yes",
|
|---|
| 133 | 'inherits' => "no",
|
|---|
| 134 | 'args' => $arguments };
|
|---|
| 135 |
|
|---|
| 136 |
|
|---|
| 137 | sub gsprintf
|
|---|
| 138 | {
|
|---|
| 139 | return &gsprintf::gsprintf(@_);
|
|---|
| 140 | }
|
|---|
| 141 |
|
|---|
| 142 |
|
|---|
| 143 | sub get_arguments
|
|---|
| 144 | {
|
|---|
| 145 | local $self = shift(@_);
|
|---|
| 146 | local $optionlistref = $self->{'option_list'};
|
|---|
| 147 | local @optionlist = @$optionlistref;
|
|---|
| 148 | local $pluginoptions = pop(@$optionlistref);
|
|---|
| 149 | local $pluginarguments = $pluginoptions->{'args'};
|
|---|
| 150 | return $pluginarguments;
|
|---|
| 151 | }
|
|---|
| 152 |
|
|---|
| 153 |
|
|---|
| 154 | sub print_xml_usage
|
|---|
| 155 | {
|
|---|
| 156 | local $self = shift(@_);
|
|---|
| 157 |
|
|---|
| 158 | # XML output is always in UTF-8
|
|---|
| 159 | &gsprintf::output_strings_in_UTF8;
|
|---|
| 160 |
|
|---|
| 161 | &PrintUsage::print_xml_header();
|
|---|
| 162 | $self->print_xml();
|
|---|
| 163 | }
|
|---|
| 164 |
|
|---|
| 165 |
|
|---|
| 166 | sub print_xml
|
|---|
| 167 | {
|
|---|
| 168 | local $self = shift(@_);
|
|---|
| 169 |
|
|---|
| 170 | local $optionlistref = $self->{'option_list'};
|
|---|
| 171 | local @optionlist = @$optionlistref;
|
|---|
| 172 | local $pluginoptions = pop(@$optionlistref);
|
|---|
| 173 | return if (!defined($pluginoptions));
|
|---|
| 174 |
|
|---|
| 175 | &gsprintf(STDERR, "<PlugInfo>\n");
|
|---|
| 176 | &gsprintf(STDERR, " <Name>$pluginoptions->{'name'}</Name>\n");
|
|---|
| 177 | my $desc = &gsprintf::lookup_string($pluginoptions->{'desc'});
|
|---|
| 178 | $desc =~ s/</&lt;/g; # doubly escaped
|
|---|
| 179 | $desc =~ s/>/&gt;/g;
|
|---|
| 180 |
|
|---|
| 181 | &gsprintf(STDERR, " <Desc>$desc</Desc>\n");
|
|---|
| 182 | &gsprintf(STDERR, " <Abstract>$pluginoptions->{'abstract'}</Abstract>\n");
|
|---|
| 183 | &gsprintf(STDERR, " <Inherits>$pluginoptions->{'inherits'}</Inherits>\n");
|
|---|
| 184 | &gsprintf(STDERR, " <Arguments>\n");
|
|---|
| 185 | if (defined($pluginoptions->{'args'})) {
|
|---|
| 186 | &PrintUsage::print_options_xml($pluginoptions->{'args'});
|
|---|
| 187 | }
|
|---|
| 188 |
|
|---|
| 189 | # Recurse up the plugin hierarchy
|
|---|
| 190 | $self->print_xml();
|
|---|
| 191 |
|
|---|
| 192 | &gsprintf(STDERR, " </Arguments>\n");
|
|---|
| 193 | &gsprintf(STDERR, "</PlugInfo>\n");
|
|---|
| 194 | }
|
|---|
| 195 |
|
|---|
| 196 |
|
|---|
| 197 | sub print_txt_usage
|
|---|
| 198 | {
|
|---|
| 199 | local $self = shift(@_);
|
|---|
| 200 |
|
|---|
| 201 | # Print the usage message for a plugin (recursively)
|
|---|
| 202 | local $descoffset = $self->determine_description_offset(0);
|
|---|
| 203 | $self->print_plugin_usage($descoffset, 1);
|
|---|
| 204 | }
|
|---|
| 205 |
|
|---|
| 206 |
|
|---|
| 207 | sub determine_description_offset
|
|---|
| 208 | {
|
|---|
| 209 | local $self = shift(@_);
|
|---|
| 210 | local $maxoffset = shift(@_);
|
|---|
| 211 |
|
|---|
| 212 | local $optionlistref = $self->{'option_list'};
|
|---|
| 213 | local @optionlist = @$optionlistref;
|
|---|
| 214 | local $pluginoptions = pop(@$optionlistref);
|
|---|
| 215 | return $maxoffset if (!defined($pluginoptions));
|
|---|
| 216 |
|
|---|
| 217 | # Find the length of the longest option string of this plugin
|
|---|
| 218 | local $pluginargs = $pluginoptions->{'args'};
|
|---|
| 219 | if (defined($pluginargs)) {
|
|---|
| 220 | local $longest = &PrintUsage::find_longest_option_string($pluginargs);
|
|---|
| 221 | if ($longest > $maxoffset) {
|
|---|
| 222 | $maxoffset = $longest;
|
|---|
| 223 | }
|
|---|
| 224 | }
|
|---|
| 225 |
|
|---|
| 226 | # Recurse up the plugin hierarchy
|
|---|
| 227 | $maxoffset = $self->determine_description_offset($maxoffset);
|
|---|
| 228 | $self->{'option_list'} = \@optionlist;
|
|---|
| 229 | return $maxoffset;
|
|---|
| 230 | }
|
|---|
| 231 |
|
|---|
| 232 |
|
|---|
| 233 | sub print_plugin_usage
|
|---|
| 234 | {
|
|---|
| 235 | local $self = shift(@_);
|
|---|
| 236 | local $descoffset = shift(@_);
|
|---|
| 237 | local $isleafclass = shift(@_);
|
|---|
| 238 |
|
|---|
| 239 | local $optionlistref = $self->{'option_list'};
|
|---|
| 240 | local @optionlist = @$optionlistref;
|
|---|
| 241 | local $pluginoptions = pop(@$optionlistref);
|
|---|
| 242 | return if (!defined($pluginoptions));
|
|---|
| 243 |
|
|---|
| 244 | local $pluginname = $pluginoptions->{'name'};
|
|---|
| 245 | local $pluginargs = $pluginoptions->{'args'};
|
|---|
| 246 | local $plugindesc = $pluginoptions->{'desc'};
|
|---|
| 247 |
|
|---|
| 248 | # Produce the usage information using the data structure above
|
|---|
| 249 | if ($isleafclass) {
|
|---|
| 250 | if (defined($plugindesc)) {
|
|---|
| 251 | &gsprintf(STDERR, "$plugindesc\n\n");
|
|---|
| 252 | }
|
|---|
| 253 | &gsprintf(STDERR, " {common.usage}: plugin $pluginname [{common.options}]\n\n");
|
|---|
| 254 | }
|
|---|
| 255 |
|
|---|
| 256 | # Display the plugin options, if there are some
|
|---|
| 257 | if (defined($pluginargs)) {
|
|---|
| 258 | # Calculate the column offset of the option descriptions
|
|---|
| 259 | local $optiondescoffset = $descoffset + 2; # 2 spaces between options & descriptions
|
|---|
| 260 |
|
|---|
| 261 | if ($isleafclass) {
|
|---|
| 262 | &gsprintf(STDERR, " {common.specific_options}:\n");
|
|---|
| 263 | }
|
|---|
| 264 | else {
|
|---|
| 265 | &gsprintf(STDERR, " {common.general_options}:\n", $pluginname);
|
|---|
| 266 | }
|
|---|
| 267 |
|
|---|
| 268 | # Display the plugin options
|
|---|
| 269 | &PrintUsage::print_options_txt($pluginargs, $optiondescoffset);
|
|---|
| 270 | }
|
|---|
| 271 |
|
|---|
| 272 | # Recurse up the plugin hierarchy
|
|---|
| 273 | $self->print_plugin_usage($descoffset, 0);
|
|---|
| 274 | $self->{'option_list'} = \@optionlist;
|
|---|
| 275 | }
|
|---|
| 276 |
|
|---|
| 277 |
|
|---|
| 278 | sub new {
|
|---|
| 279 | my $class = shift (@_);
|
|---|
| 280 | my $plugin_name = shift (@_);
|
|---|
| 281 | my $self = {};
|
|---|
| 282 | $self->{'plugin_type'} = "BasPlug";
|
|---|
| 283 | my $enc = "^(";
|
|---|
| 284 | map {$enc .= "$_|";} keys %$encodings::encodings;
|
|---|
| 285 | my $denc = $enc . "ascii|utf8|unicode)\$";
|
|---|
| 286 | $enc .= "ascii|utf8|unicode|auto)\$";
|
|---|
| 287 |
|
|---|
| 288 | $self->{'outhandle'} = STDERR;
|
|---|
| 289 | my $year = (localtime)[5]+1900;
|
|---|
| 290 |
|
|---|
| 291 | $self->{'textcat'} = new textcat();
|
|---|
| 292 |
|
|---|
| 293 | $self->{'num_processed'} = 0;
|
|---|
| 294 | $self->{'num_not_processed'} = 0;
|
|---|
| 295 | $self->{'num_blocked'} = 0;
|
|---|
| 296 | $self->{'num_archives'} = 0;
|
|---|
| 297 |
|
|---|
| 298 | # 14-05-02 To allow for proper inheritance of arguments - John Thompson
|
|---|
| 299 | $self->{'option_list'} = [ $options ];
|
|---|
| 300 |
|
|---|
| 301 | # general options available to all plugins
|
|---|
| 302 | if (!parsargv::parse(\@_,
|
|---|
| 303 | q^process_exp/.*/^, \$self->{'process_exp'},
|
|---|
| 304 | q^block_exp/.*/^, \$self->{'block_exp'},
|
|---|
| 305 | q^extract_language^, \$self->{'extract_language'},
|
|---|
| 306 | q^extract_acronyms^, \$self->{'extract_acronyms'},
|
|---|
| 307 | q^extract_keyphrases^, \$self->{'kea'}, #with extra options (UNDOCUMENTED)
|
|---|
| 308 | q^extract_keyphrase_options/.*/^, \$self->{'kea_options'}, #no extra options (UNDOCUMENTED)
|
|---|
| 309 | qq^input_encoding/$enc/auto^, \$self->{'input_encoding'},
|
|---|
| 310 | qq^default_encoding/$denc/utf8^, \$self->{'default_encoding'},
|
|---|
| 311 | q^extract_email^, \$self->{'extract_email'},
|
|---|
| 312 | q^markup_acronyms^, \$self->{'markup_acronyms'},
|
|---|
| 313 | q^default_language/.{2}/en^, \$self->{'default_language'},
|
|---|
| 314 | q^first/.*/^, \$self->{'first'},
|
|---|
| 315 | q^extract_historical_years^, \$self->{'date_extract'},
|
|---|
| 316 | qq^maximum_year/\\d{4}/$year^, \$self->{'max_year'},
|
|---|
| 317 | q^no_bibliography^, \$self->{'no_biblio'},
|
|---|
| 318 | qq^maximum_century/-?\\d{1,2}( ?B\\.C\\.E\\.)?/-1^, \$self->{'max_century'},
|
|---|
| 319 | q^cover_image^, \$self->{'cover_image'},
|
|---|
| 320 | q^separate_cjk^, \$self->{'separate_cjk'},
|
|---|
| 321 | "allow_extra_options")) {
|
|---|
| 322 |
|
|---|
| 323 | &gsprintf(STDERR, "\n{BasPlug.bad_general_option}\n", $plugin_name);
|
|---|
| 324 | $self->print_txt_usage(""); # Use default resource bundle
|
|---|
| 325 | die "\n";
|
|---|
| 326 | }
|
|---|
| 327 |
|
|---|
| 328 | return bless $self, $class;
|
|---|
| 329 | }
|
|---|
| 330 |
|
|---|
| 331 | # initialize BasPlug options
|
|---|
| 332 | # if init() is overridden in a sub-class, remember to call BasPlug::init()
|
|---|
| 333 | sub init {
|
|---|
| 334 | my $self = shift (@_);
|
|---|
| 335 | my ($verbosity, $outhandle, $failhandle) = @_;
|
|---|
| 336 |
|
|---|
| 337 | # verbosity is passed through from the processor
|
|---|
| 338 | $self->{'verbosity'} = $verbosity;
|
|---|
| 339 |
|
|---|
| 340 | # as are the outhandle and failhandle
|
|---|
| 341 | $self->{'outhandle'} = $outhandle if defined $outhandle;
|
|---|
| 342 | $self->{'failhandle'} = $failhandle;
|
|---|
| 343 |
|
|---|
| 344 | # set process_exp and block_exp to defaults unless they were
|
|---|
| 345 | # explicitly set
|
|---|
| 346 |
|
|---|
| 347 | if ((!$self->is_recursive()) and
|
|---|
| 348 | (!defined $self->{'process_exp'}) || ($self->{'process_exp'} eq "")) {
|
|---|
| 349 |
|
|---|
| 350 | $self->{'process_exp'} = $self->get_default_process_exp ();
|
|---|
| 351 | if ($self->{'process_exp'} eq "") {
|
|---|
| 352 | warn ref($self) . " Warning: Non-recursive plugin has no process_exp\n";
|
|---|
| 353 | }
|
|---|
| 354 | }
|
|---|
| 355 |
|
|---|
| 356 | if ((!defined $self->{'block_exp'}) || ($self->{'block_exp'} eq "")) {
|
|---|
| 357 | $self->{'block_exp'} = $self->get_default_block_exp ();
|
|---|
| 358 | }
|
|---|
| 359 | }
|
|---|
| 360 |
|
|---|
| 361 | sub begin {
|
|---|
| 362 | my $self = shift (@_);
|
|---|
| 363 | my ($pluginfo, $base_dir, $processor, $maxdocs) = @_;
|
|---|
| 364 | $self->initialise_extractors();
|
|---|
| 365 | }
|
|---|
| 366 |
|
|---|
| 367 | sub end {
|
|---|
| 368 | my ($self) = @_;
|
|---|
| 369 | $self->finalise_extractors();
|
|---|
| 370 | }
|
|---|
| 371 |
|
|---|
| 372 | # this function should be overridden to return 1
|
|---|
| 373 | # in recursive plugins
|
|---|
| 374 | sub is_recursive {
|
|---|
| 375 | my $self = shift (@_);
|
|---|
| 376 |
|
|---|
| 377 | return 0;
|
|---|
| 378 | }
|
|---|
| 379 |
|
|---|
| 380 | sub get_default_block_exp {
|
|---|
| 381 | my $self = shift (@_);
|
|---|
| 382 |
|
|---|
| 383 | return "";
|
|---|
| 384 | }
|
|---|
| 385 |
|
|---|
| 386 | sub get_default_process_exp {
|
|---|
| 387 | my $self = shift (@_);
|
|---|
| 388 |
|
|---|
| 389 | return "";
|
|---|
| 390 | }
|
|---|
| 391 |
|
|---|
| 392 | # The BasPlug read() function. This function does all the right things
|
|---|
| 393 | # to make general options work for a given plugin. It calls the process()
|
|---|
| 394 | # function which does all the work specific to a plugin (like the old
|
|---|
| 395 | # read functions used to do). Most plugins should define their own
|
|---|
| 396 | # process() function and let this read() function keep control.
|
|---|
| 397 | #
|
|---|
| 398 | # recursive plugins (e.g. RecPlug) and specialized plugins like those
|
|---|
| 399 | # capable of processing many documents within a single file (e.g.
|
|---|
| 400 | # GMLPlug) should normally implement their own version of read()
|
|---|
| 401 | #
|
|---|
| 402 | # Return number of files processed, undef if can't recognise, -1 if can't
|
|---|
| 403 | # process
|
|---|
| 404 | # Note that $base_dir might be "" and that $file might
|
|---|
| 405 | # include directories
|
|---|
| 406 |
|
|---|
| 407 | sub read {
|
|---|
| 408 | my $self = shift (@_);
|
|---|
| 409 |
|
|---|
| 410 | my ($pluginfo, $base_dir, $file, $metadata, $processor, $maxdocs, $gli) = @_;
|
|---|
| 411 |
|
|---|
| 412 | if ($self->is_recursive()) {
|
|---|
| 413 | &gsprintf(STDERR, "{BasPlug.read_must_be_implemented}") && die "\n";
|
|---|
| 414 | }
|
|---|
| 415 |
|
|---|
| 416 | my $outhandle = $self->{'outhandle'};
|
|---|
| 417 |
|
|---|
| 418 | my $filename = $file;
|
|---|
| 419 | $filename = &util::filename_cat ($base_dir, $file) if $base_dir =~ /\w/;
|
|---|
| 420 |
|
|---|
| 421 | if ($self->{'block_exp'} ne "" && $filename =~ /$self->{'block_exp'}/) {
|
|---|
| 422 | $self->{'num_blocked'} ++;
|
|---|
| 423 | return 0; # blocked
|
|---|
| 424 | }
|
|---|
| 425 | if ($filename !~ /$self->{'process_exp'}/ || !-f $filename) {
|
|---|
| 426 | return undef; # can't recognise
|
|---|
| 427 | }
|
|---|
| 428 | $file =~ s/^[\/\\]+//; # $file often begins with / so we'll tidy it up
|
|---|
| 429 |
|
|---|
| 430 | # Do encoding stuff
|
|---|
| 431 | my ($language, $encoding) = $self->textcat_get_language_encoding ($filename);
|
|---|
| 432 |
|
|---|
| 433 | # create a new document
|
|---|
| 434 | my $doc_obj = new doc ($filename, "indexed_doc");
|
|---|
| 435 | $doc_obj->set_OIDtype ($processor->{'OIDtype'});
|
|---|
| 436 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Language", $language);
|
|---|
| 437 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Encoding", $encoding);
|
|---|
| 438 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "Plugin", "$self->{'plugin_type'}");
|
|---|
| 439 | $doc_obj->add_utf8_metadata($doc_obj->get_top_section(), "FileSize", (-s $filename));
|
|---|
| 440 |
|
|---|
| 441 | my ($filemeta) = $file =~ /([^\\\/]+)$/;
|
|---|
| 442 | # how do we know what encoding the filename is in?
|
|---|
| 443 | $doc_obj->add_metadata($doc_obj->get_top_section(), "Source", &ghtml::dmsafe($filemeta));
|
|---|
| 444 | if ($self->{'cover_image'}) {
|
|---|
| 445 | $self->associate_cover_image($doc_obj, $filename);
|
|---|
| 446 | }
|
|---|
| 447 |
|
|---|
| 448 | # read in file ($text will be in utf8)
|
|---|
| 449 | my $text = "";
|
|---|
| 450 | $self->read_file ($filename, $encoding, $language, \$text);
|
|---|
| 451 |
|
|---|
| 452 | if (!length ($text)) {
|
|---|
| 453 | my $plugin_name = ref ($self);
|
|---|
| 454 | &gsprintf($outhandle, "$plugin_name: {BasPlug.file_has_no_text}\n", $filename) if $self->{'verbosity'};
|
|---|
| 455 |
|
|---|
| 456 | my $failhandle = $self->{'failhandle'};
|
|---|
| 457 | &gsprintf($failhandle, "$file: " . ref($self) . ": {BasPlug.empty_file}\n");
|
|---|
| 458 | # print $failhandle "$file: " . ref($self) . ": file contains no text\n";
|
|---|
| 459 | $self->{'num_not_processed'} ++;
|
|---|
| 460 |
|
|---|
| 461 | return 0; # what should we return here?? error but don't want to pass it on
|
|---|
| 462 | }
|
|---|
| 463 |
|
|---|
| 464 | # include any metadata passed in from previous plugins
|
|---|
| 465 | # note that this metadata is associated with the top level section
|
|---|
| 466 | $self->extra_metadata ($doc_obj, $doc_obj->get_top_section(), $metadata);
|
|---|
| 467 |
|
|---|
| 468 | # do plugin specific processing of doc_obj
|
|---|
| 469 | return -1 unless defined ($self->process (\$text, $pluginfo, $base_dir, $file, $metadata, $doc_obj, $gli));
|
|---|
| 470 |
|
|---|
| 471 | # do any automatic metadata extraction
|
|---|
| 472 | $self->auto_extract_metadata ($doc_obj);
|
|---|
| 473 |
|
|---|
| 474 | # add an OID
|
|---|
| 475 | # see if there is a plugin-specific set_OID function...
|
|---|
| 476 | if (defined ($self->can(set_OID))) {
|
|---|
| 477 | # it will need $doc_obj to set the Identifier metadata...
|
|---|
| 478 | $self->set_OID($doc_obj);
|
|---|
| 479 | } else {
|
|---|
| 480 | # use the default set_OID() in doc.pm
|
|---|
| 481 | $doc_obj->set_OID();
|
|---|
| 482 | }
|
|---|
| 483 |
|
|---|
| 484 | # process the document
|
|---|
| 485 | $processor->process($doc_obj);
|
|---|
| 486 |
|
|---|
| 487 | $self->{'num_processed'} ++;
|
|---|
| 488 |
|
|---|
| 489 | return 1; # processed the file
|
|---|
| 490 | }
|
|---|
| 491 |
|
|---|
| 492 | # returns undef if file is rejected by the plugin
|
|---|
| 493 | sub process {
|
|---|
| 494 | my $self = shift (@_);
|
|---|
| 495 | my ($textref, $pluginfo, $base_dir, $file, $metadata, $doc_obj) = @_;
|
|---|
| 496 |
|
|---|
| 497 | &gsprintf(STDERR, "BasPlug::process {common.must_be_implemented}\n") && die "\n";
|
|---|
| 498 | # die "Basplug::process function must be implemented in sub-class\n";
|
|---|
| 499 |
|
|---|
| 500 | return undef; # never gets here
|
|---|
| 501 | }
|
|---|
| 502 |
|
|---|
| 503 | # uses the multiread package to read in the entire file pointed to
|
|---|
| 504 | # by filename and loads the resulting text into $$textref. Input text
|
|---|
| 505 | # may be in any of the encodings handled by multiread, output text
|
|---|
| 506 | # will be in utf8
|
|---|
| 507 | sub read_file {
|
|---|
| 508 | my $self = shift (@_);
|
|---|
| 509 | my ($filename, $encoding, $language, $textref) = @_;
|
|---|
| 510 |
|
|---|
| 511 | if (!-r $filename)
|
|---|
| 512 | {
|
|---|
| 513 | my $outhandle = $self->{'outhandle'};
|
|---|
| 514 | &gsprintf($outhandle, "{BasPlug.read_denied}\n", $filename) if $self->{'verbosity'};
|
|---|
| 515 | # print $outhandle "Read permission denied for $filename\n" if $self->{'verbosity'};
|
|---|
| 516 | return;
|
|---|
| 517 | }
|
|---|
| 518 |
|
|---|
| 519 | $$textref = "";
|
|---|
| 520 |
|
|---|
| 521 | open (FILE, $filename) || (&gsprintf(STDERR, "BasPlug::read_file {BasPlug.could_not_open_for_reading} ($!)\n", $filename) && die "\n");
|
|---|
| 522 | # open (FILE, $filename) || die "BasPlug::read_file could not open $filename for reading ($!)\n";
|
|---|
| 523 |
|
|---|
| 524 | if ($encoding eq "ascii") {
|
|---|
| 525 | undef $/;
|
|---|
| 526 | $$textref = <FILE>;
|
|---|
| 527 | $/ = "\n";
|
|---|
| 528 | } else {
|
|---|
| 529 | my $reader = new multiread();
|
|---|
| 530 | $reader->set_handle ('BasPlug::FILE');
|
|---|
| 531 | $reader->set_encoding ($encoding);
|
|---|
| 532 | $reader->read_file ($textref);
|
|---|
| 533 |
|
|---|
| 534 | #Now segments chinese if the separate_cjk option is set
|
|---|
| 535 | if ($self->{'separate_cjk'}) {
|
|---|
| 536 | # segment the Chinese words
|
|---|
| 537 | $$textref = &cnseg::segment($$textref);
|
|---|
| 538 | }
|
|---|
| 539 | }
|
|---|
| 540 |
|
|---|
| 541 | close FILE;
|
|---|
| 542 | }
|
|---|
| 543 |
|
|---|
| 544 | sub filename_based_title
|
|---|
| 545 | {
|
|---|
| 546 | my $self = shift (@_);
|
|---|
| 547 | my ($file) = @_;
|
|---|
| 548 |
|
|---|
| 549 | my $file_derived_title = $file;
|
|---|
| 550 | $file_derived_title =~ s/_/ /g;
|
|---|
| 551 | $file_derived_title =~ s/\..*?$//;
|
|---|
| 552 |
|
|---|
| 553 | return $file_derived_title;
|
|---|
| 554 | }
|
|---|
| 555 |
|
|---|
| 556 |
|
|---|
| 557 | sub title_fallback
|
|---|
| 558 | {
|
|---|
| 559 | my $self = shift (@_);
|
|---|
| 560 | my ($doc_obj,$section,$file) = @_;
|
|---|
| 561 |
|
|---|
| 562 | if (!defined $doc_obj->get_metadata_element ($section, "Title")) {
|
|---|
| 563 |
|
|---|
| 564 | my $file_derived_title = $self->filename_based_title($file);
|
|---|
| 565 | $doc_obj->add_metadata ($section, "Title", $file_derived_title);
|
|---|
| 566 | }
|
|---|
| 567 | }
|
|---|
| 568 |
|
|---|
| 569 | sub textcat_get_language_encoding {
|
|---|
| 570 | my $self = shift (@_);
|
|---|
| 571 | my ($filename) = @_;
|
|---|
| 572 |
|
|---|
| 573 | my ($language, $encoding, $extracted_encoding);
|
|---|
| 574 | if ($self->{'input_encoding'} eq "auto") {
|
|---|
| 575 | # use textcat to automatically work out the input encoding and language
|
|---|
| 576 | ($language, $encoding) = $self->get_language_encoding ($filename);
|
|---|
| 577 | } elsif ($self->{'extract_language'}) {
|
|---|
| 578 | # use textcat to get language metadata
|
|---|
| 579 | ($language, $extracted_encoding) = $self->get_language_encoding ($filename);
|
|---|
| 580 | $encoding = $self->{'input_encoding'};
|
|---|
| 581 | # don't print this message for english... english in utf8 is identical
|
|---|
| 582 | # to english in iso-8859-1 (except for some punctuation). We don't have
|
|---|
| 583 | # a language model for en_utf8, so textcat always says iso-8859-1!
|
|---|
| 584 | if ($extracted_encoding ne $encoding && $language ne "en"
|
|---|
| 585 | && $self->{'verbosity'}) {
|
|---|
| 586 | my $plugin_name = ref ($self);
|
|---|
| 587 | my $outhandle = $self->{'outhandle'};
|
|---|
| 588 | &gsprintf($outhandle, "$plugin_name: {BasPlug.wrong_encoding}\n", $filename, $encoding, $extracted_encoding);
|
|---|
| 589 | # print $outhandle "$plugin_name: WARNING: $filename was read using $encoding encoding but ";
|
|---|
| 590 | # print $outhandle "appears to be encoded as $extracted_encoding.\n";
|
|---|
| 591 | }
|
|---|
| 592 | } else {
|
|---|
| 593 | $language = $self->{'default_language'};
|
|---|
| 594 | $encoding = $self->{'input_encoding'};
|
|---|
| 595 | }
|
|---|
| 596 | return ($language, $encoding);
|
|---|
| 597 | }
|
|---|
| 598 |
|
|---|
| 599 | # Uses textcat to work out the encoding and language of the text in
|
|---|
| 600 | # $filename. All html tags are removed before processing.
|
|---|
| 601 | # returns an array containing "language" and "encoding"
|
|---|
| 602 | sub get_language_encoding {
|
|---|
| 603 | my $self = shift (@_);
|
|---|
| 604 | my ($filename) = @_;
|
|---|
| 605 | my $outhandle = $self->{'outhandle'};
|
|---|
| 606 |
|
|---|
| 607 | # read in file
|
|---|
| 608 | 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";
|
|---|
| 609 | undef $/;
|
|---|
| 610 | my $text = <FILE>;
|
|---|
| 611 | $/ = "\n";
|
|---|
| 612 | close FILE;
|
|---|
| 613 |
|
|---|
| 614 | # remove <title>stuff</title> -- as titles tend often to be in English
|
|---|
| 615 | # for foreign language documents
|
|---|
| 616 | $text =~ s/<title>.*?<\/title>//i;
|
|---|
| 617 |
|
|---|
| 618 | # remove all HTML tags
|
|---|
| 619 | $text =~ s/<[^>]*>//sg;
|
|---|
| 620 |
|
|---|
| 621 | # get the language/encoding
|
|---|
| 622 | my $results = $self->{'textcat'}->classify(\$text);
|
|---|
| 623 |
|
|---|
| 624 | # if textcat returns 3 or less possibilities we'll use the
|
|---|
| 625 | # first one in the list - otherwise use the defaults
|
|---|
| 626 | if (scalar @$results > 3) {
|
|---|
| 627 | # changed 12 Feb 2003 by jrm21
|
|---|
| 628 | # use the most popular encoding at least... otherwise we might
|
|---|
| 629 | # generate invalid archive files!
|
|---|
| 630 | my %guessed_encodings = ();
|
|---|
| 631 | foreach my $result (@$results) {
|
|---|
| 632 | $result =~ /([^\-]+)$/;
|
|---|
| 633 | my $enc=$1;
|
|---|
| 634 | if (!defined($guessed_encodings{$enc})) {
|
|---|
| 635 | $guessed_encodings{$enc}=0;
|
|---|
| 636 | }
|
|---|
| 637 | $guessed_encodings{$enc}++;
|
|---|
| 638 | }
|
|---|
| 639 | my $best_encoding="";
|
|---|
| 640 | $guessed_encodings{""}=-1;
|
|---|
| 641 | foreach my $enc (keys %guessed_encodings) {
|
|---|
| 642 | if ($guessed_encodings{$enc} > $guessed_encodings{$best_encoding}){
|
|---|
| 643 | $best_encoding=$enc;
|
|---|
| 644 | }
|
|---|
| 645 | }
|
|---|
| 646 |
|
|---|
| 647 | if ($self->{'input_encoding'} ne 'auto') {
|
|---|
| 648 | if ($self->{'extract_language'} && $self->{'verbosity'}) {
|
|---|
| 649 | &gsprintf($outhandle, "BasPlug: {BasPlug.could_not_extract_language}\n", $filename, $self->{'default_language'});
|
|---|
| 650 | # print $outhandle "BasPlug: WARNING: language could not be extracted from $filename - ";
|
|---|
| 651 | # print $outhandle "defaulting to $self->{'default_language'}\n";
|
|---|
| 652 | }
|
|---|
| 653 | return ($self->{'default_language'}, $self->{'input_encoding'});
|
|---|
| 654 |
|
|---|
| 655 | } else {
|
|---|
| 656 | if ($self->{'verbosity'}) {
|
|---|
| 657 | &gsprintf($outhandle, "BasPlug: {BasPlug.could_not_extract_language}\n", $filename, $self->{'default_language'});
|
|---|
| 658 | # print $outhandle "BASPlug: WARNING: language could not be extracted from $filename - ";
|
|---|
| 659 | # print $outhandle "defaulting to $self->{'default_language'}.\n";
|
|---|
| 660 | }
|
|---|
| 661 | return ($self->{'default_language'}, $best_encoding);
|
|---|
| 662 | }
|
|---|
| 663 | }
|
|---|
| 664 |
|
|---|
| 665 | # format language/encoding
|
|---|
| 666 | my ($language, $encoding) = $results->[0] =~ /^([^-]*)(?:-(.*))?$/;
|
|---|
| 667 | if (!defined $language) {
|
|---|
| 668 | if ($self->{'verbosity'}) {
|
|---|
| 669 | &gsprintf($outhandle, "BasPlug: {BasPlug.could_not_extract_language}\n", $filename, $self->{'default_language'});
|
|---|
| 670 | # print $outhandle "BasPlug: WARNING: language could not be extracted from $filename - ";
|
|---|
| 671 | # print $outhandle "defaulting to $self->{'default_language'}\n";
|
|---|
| 672 | }
|
|---|
| 673 | $language = $self->{'default_language'};
|
|---|
| 674 | }
|
|---|
| 675 | if (!defined $encoding) {
|
|---|
| 676 | if ($self->{'verbosity'}) {
|
|---|
| 677 | &gsprintf($outhandle, "BasPlug: {BasPlug.could_not_extract_encoding}\n", $filename, $self->{'default_encoding'});
|
|---|
| 678 | # print $outhandle "BasPlug: WARNING: encoding could not be extracted from $filename - ";
|
|---|
| 679 | # print $outhandle "defaulting to $self->{'default_encoding'}\n";
|
|---|
| 680 | }
|
|---|
| 681 | $encoding = $self->{'default_encoding'};
|
|---|
| 682 | }
|
|---|
| 683 |
|
|---|
| 684 |
|
|---|
| 685 | # check for equivalents where textcat doesn't have some encodings...
|
|---|
| 686 | # eg MS versions of standard encodings
|
|---|
| 687 | if ($encoding =~ /^iso_8859_(\d+)/) {
|
|---|
| 688 | my $iso = $1; # which variant of the iso standard?
|
|---|
| 689 | # iso-8859 sets don't use chars 0x80-0x9f, windows codepages do
|
|---|
| 690 | if ($text =~ /[\x80-\x9f]/) {
|
|---|
| 691 | # Western Europe
|
|---|
| 692 | if ($iso == 1 or $iso == 15) { $encoding = 'windows_1252' }
|
|---|
| 693 | elsif ($iso == 2) { $encoding = 'windows_1250' } # Central Europe
|
|---|
| 694 | elsif ($iso == 5) { $encoding = 'windows_1251' } # Cyrillic
|
|---|
| 695 | elsif ($iso == 6) { $encoding = 'windows_1256' } # Arabic
|
|---|
| 696 | elsif ($iso == 7) { $encoding = 'windows_1253' } # Greek
|
|---|
| 697 | elsif ($iso == 8) { $encoding = 'windows_1255' } # Hebrew
|
|---|
| 698 | elsif ($iso == 9) { $encoding = 'windows_1254' } # Turkish
|
|---|
| 699 | }
|
|---|
| 700 | }
|
|---|
| 701 |
|
|---|
| 702 | if ($encoding !~ /^(ascii|utf8|unicode)$/ &&
|
|---|
| 703 | !defined $encodings::encodings->{$encoding}) {
|
|---|
| 704 | if ($self->{'verbosity'}) {
|
|---|
| 705 | gsprintf($outhandle, "BasPlug: {BasPlug.unsupported_encoding}\n",
|
|---|
| 706 | $filename, $encoding, $self->{'default_encoding'});
|
|---|
| 707 | }
|
|---|
| 708 | $encoding = $self->{'default_encoding'};
|
|---|
| 709 | }
|
|---|
| 710 |
|
|---|
| 711 | return ($language, $encoding);
|
|---|
| 712 | }
|
|---|
| 713 |
|
|---|
| 714 | # add any extra metadata that's been passed around from one
|
|---|
| 715 | # plugin to another.
|
|---|
| 716 | # extra_metadata uses add_utf8_metadata so it expects metadata values
|
|---|
| 717 | # to already be in utf8
|
|---|
| 718 | sub extra_metadata {
|
|---|
| 719 | my $self = shift (@_);
|
|---|
| 720 | my ($doc_obj, $cursection, $metadata) = @_;
|
|---|
| 721 |
|
|---|
| 722 | foreach my $field (keys(%$metadata)) {
|
|---|
| 723 | # $metadata->{$field} may be an array reference
|
|---|
| 724 | if (ref ($metadata->{$field}) eq "ARRAY") {
|
|---|
| 725 | map {
|
|---|
| 726 | $doc_obj->add_utf8_metadata ($cursection, $field, $_);
|
|---|
| 727 | } @{$metadata->{$field}};
|
|---|
| 728 | } else {
|
|---|
| 729 | $doc_obj->add_utf8_metadata ($cursection, $field, $metadata->{$field});
|
|---|
| 730 | }
|
|---|
| 731 | }
|
|---|
| 732 | }
|
|---|
| 733 |
|
|---|
| 734 | # initialise metadata extractors
|
|---|
| 735 | sub initialise_extractors {
|
|---|
| 736 | my $self = shift (@_);
|
|---|
| 737 |
|
|---|
| 738 | if ($self->{'extract_acronyms'} || $self->{'markup_acronyms'}) {
|
|---|
| 739 | &acronym::initialise_acronyms();
|
|---|
| 740 | }
|
|---|
| 741 | }
|
|---|
| 742 |
|
|---|
| 743 | # finalise metadata extractors
|
|---|
| 744 | sub finalise_extractors {
|
|---|
| 745 | my $self = shift (@_);
|
|---|
| 746 |
|
|---|
| 747 | if ($self->{'extract_acronyms'} || $self->{'markup_acronyms'}) {
|
|---|
| 748 | &acronym::finalise_acronyms();
|
|---|
| 749 | }
|
|---|
| 750 | }
|
|---|
| 751 |
|
|---|
| 752 | # FIRSTNNN: extract the first NNN characters as metadata
|
|---|
| 753 | sub extract_first_NNNN_characters {
|
|---|
| 754 | my $self = shift (@_);
|
|---|
| 755 | my ($textref, $doc_obj, $thissection) = @_;
|
|---|
| 756 |
|
|---|
| 757 | foreach my $size (split /,/, $self->{'first'}) {
|
|---|
| 758 | my $tmptext = $$textref;
|
|---|
| 759 | $tmptext =~ s/^\s+//;
|
|---|
| 760 | $tmptext =~ s/\s+$//;
|
|---|
| 761 | $tmptext =~ s/\s+/ /gs;
|
|---|
| 762 | $tmptext = substr ($tmptext, 0, $size);
|
|---|
| 763 | $tmptext =~ s/\s\S*$/…/;
|
|---|
| 764 | $doc_obj->add_utf8_metadata ($thissection, "First$size", $tmptext);
|
|---|
| 765 | }
|
|---|
| 766 | }
|
|---|
| 767 |
|
|---|
| 768 | sub extract_email {
|
|---|
| 769 | my $self = shift (@_);
|
|---|
| 770 | my ($textref, $doc_obj, $thissection) = @_;
|
|---|
| 771 | my $outhandle = $self->{'outhandle'};
|
|---|
| 772 |
|
|---|
| 773 | # print $outhandle " extracting email addresses ...\n"
|
|---|
| 774 | &gsprintf($outhandle, " {BasPlug.extracting_emails}...\n")
|
|---|
| 775 | if ($self->{'verbosity'} > 2);
|
|---|
| 776 |
|
|---|
| 777 | my @email = ($$textref =~ m/([-a-z0-9\.@+_=]+@(?:[-a-z0-9]+\.)+(?:com|org|edu|mil|int|net|[a-z][a-z]))/g);
|
|---|
| 778 | @email = sort @email;
|
|---|
| 779 |
|
|---|
| 780 | my @email2 = ();
|
|---|
| 781 | foreach my $address (@email) {
|
|---|
| 782 | if (!(join(" ",@email2) =~ m/$address/ )) {
|
|---|
| 783 | push @email2, $address;
|
|---|
| 784 | $doc_obj->add_utf8_metadata ($thissection, "emailAddress", $address);
|
|---|
| 785 | # print $outhandle " extracting $address\n"
|
|---|
| 786 | &gsprintf($outhandle, " {BasPlug.extracting} $address\n")
|
|---|
| 787 | if ($self->{'verbosity'} > 3);
|
|---|
| 788 | }
|
|---|
| 789 | }
|
|---|
| 790 | # print $outhandle " done extracting email addresses.\n"
|
|---|
| 791 | &gsprintf($outhandle, " {BasPlug.done_email_extract}\n")
|
|---|
| 792 | if ($self->{'verbosity'} > 2);
|
|---|
| 793 | }
|
|---|
| 794 |
|
|---|
| 795 | # extract metadata
|
|---|
| 796 | sub auto_extract_metadata {
|
|---|
| 797 |
|
|---|
| 798 | my $self = shift (@_);
|
|---|
| 799 | my ($doc_obj) = @_;
|
|---|
| 800 |
|
|---|
| 801 | if ($self->{'extract_email'}) {
|
|---|
| 802 | my $thissection = $doc_obj->get_top_section();
|
|---|
| 803 | while (defined $thissection) {
|
|---|
| 804 | my $text = $doc_obj->get_text($thissection);
|
|---|
| 805 | $self->extract_email (\$text, $doc_obj, $thissection) if $text =~ /./;
|
|---|
| 806 | $thissection = $doc_obj->get_next_section ($thissection);
|
|---|
| 807 | }
|
|---|
| 808 | }
|
|---|
| 809 |
|
|---|
| 810 |
|
|---|
| 811 | #adding kea keyphrases
|
|---|
| 812 | if ($self->{'kea'}) {
|
|---|
| 813 |
|
|---|
| 814 | my $thissection = $doc_obj->get_top_section();
|
|---|
| 815 | my $text = "";
|
|---|
| 816 | my @list;
|
|---|
| 817 |
|
|---|
| 818 | while (defined $thissection) { #loop through sections to gather whole doc
|
|---|
| 819 | my $sectiontext = $doc_obj->get_text($thissection);
|
|---|
| 820 | $text = $text.$sectiontext;
|
|---|
| 821 | $thissection = $doc_obj->get_next_section ($thissection);
|
|---|
| 822 | }
|
|---|
| 823 |
|
|---|
| 824 | if($self->{'kea_options'}) { #if kea options flag is set, call Kea with specified options
|
|---|
| 825 | @list = &Kea::extract_KeyPhrases ($text, $self->{'kea_options'});
|
|---|
| 826 | } else { #otherwise call Kea with no options
|
|---|
| 827 | @list = &Kea::extract_KeyPhrases ($text);
|
|---|
| 828 | }
|
|---|
| 829 |
|
|---|
| 830 | if(@list){ #if a list of kea keyphrases was returned (ie not empty)
|
|---|
| 831 | my $keyphrases = $list[0]; #first arg is keyphrase list
|
|---|
| 832 | my $stems = $list[1]; #second arg is stemmed keyphrase list
|
|---|
| 833 | &gsprintf(STDERR, "{BasPlug.keyphrases}: $keyphrases\n");
|
|---|
| 834 | # print STDERR "keyphrases: $keyphrases\n";
|
|---|
| 835 | &gsprintf(STDERR, "{BasPlug.stems}: $stems\n");
|
|---|
| 836 | # print STDERR "stems: $stems\n";
|
|---|
| 837 | $thissection = $doc_obj->get_top_section(); #add metadata to top section
|
|---|
| 838 | # add all key phrases as one metadata
|
|---|
| 839 | $doc_obj->add_metadata($thissection, "Keyphrases", $keyphrases);
|
|---|
| 840 | # add individual key phrases as multiple metadata
|
|---|
| 841 | foreach my $keyphrase (split(',', $keyphrases)) {
|
|---|
| 842 | $keyphrase =~ s/^\s*//; $keyphrase =~ s/\s*$//;
|
|---|
| 843 | $doc_obj->add_metadata($thissection, "Keyphrase", $keyphrase);
|
|---|
| 844 | }
|
|---|
| 845 | $doc_obj->add_metadata($thissection, "stems", $stems);
|
|---|
| 846 | }
|
|---|
| 847 | } #end of kea
|
|---|
| 848 |
|
|---|
| 849 | if ($self->{'first'}) {
|
|---|
| 850 | my $thissection = $doc_obj->get_top_section();
|
|---|
| 851 | while (defined $thissection) {
|
|---|
| 852 | my $text = $doc_obj->get_text($thissection);
|
|---|
| 853 | $self->extract_first_NNNN_characters (\$text, $doc_obj, $thissection) if $text =~ /./;
|
|---|
| 854 | $thissection = $doc_obj->get_next_section ($thissection);
|
|---|
| 855 | }
|
|---|
| 856 | }
|
|---|
| 857 |
|
|---|
| 858 | if ($self->{'extract_acronyms'}) {
|
|---|
| 859 | my $thissection = $doc_obj->get_top_section();
|
|---|
| 860 | while (defined $thissection) {
|
|---|
| 861 | my $text = $doc_obj->get_text($thissection);
|
|---|
| 862 | $self->extract_acronyms (\$text, $doc_obj, $thissection) if $text =~ /./;
|
|---|
| 863 | $thissection = $doc_obj->get_next_section ($thissection);
|
|---|
| 864 | }
|
|---|
| 865 | }
|
|---|
| 866 |
|
|---|
| 867 | if ($self->{'markup_acronyms'}) {
|
|---|
| 868 | my $thissection = $doc_obj->get_top_section();
|
|---|
| 869 | while (defined $thissection) {
|
|---|
| 870 | my $text = $doc_obj->get_text($thissection);
|
|---|
| 871 | $text = $self->markup_acronyms ($text, $doc_obj, $thissection);
|
|---|
| 872 | $doc_obj->delete_text($thissection);
|
|---|
| 873 | $doc_obj->add_text($thissection, $text);
|
|---|
| 874 | $thissection = $doc_obj->get_next_section ($thissection);
|
|---|
| 875 | }
|
|---|
| 876 | }
|
|---|
| 877 |
|
|---|
| 878 | if($self->{'date_extract'}) {
|
|---|
| 879 | my $thissection = $doc_obj->get_top_section();
|
|---|
| 880 | while (defined $thissection) {
|
|---|
| 881 |
|
|---|
| 882 | my $text = $doc_obj->get_text($thissection);
|
|---|
| 883 | &DateExtract::get_date_metadata($text, $doc_obj,
|
|---|
| 884 | $thissection,
|
|---|
| 885 | $self->{'no_biblio'},
|
|---|
| 886 | $self->{'max_year'},
|
|---|
| 887 | $self->{'max_century'});
|
|---|
| 888 | $thissection = $doc_obj->get_next_section ($thissection);
|
|---|
| 889 | }
|
|---|
| 890 | }
|
|---|
| 891 | }
|
|---|
| 892 |
|
|---|
| 893 | # extract acronyms from a section in a document. progress is
|
|---|
| 894 | # reported to outhandle based on the verbosity. both the Acronym
|
|---|
| 895 | # and the AcronymKWIC metadata items are created.
|
|---|
| 896 |
|
|---|
| 897 | sub extract_acronyms {
|
|---|
| 898 | my $self = shift (@_);
|
|---|
| 899 | my ($textref, $doc_obj, $thissection) = @_;
|
|---|
| 900 | my $outhandle = $self->{'outhandle'};
|
|---|
| 901 |
|
|---|
| 902 | # print $outhandle " extracting acronyms ...\n"
|
|---|
| 903 | &gsprintf($outhandle, " {BasPlug.extracting_acronyms}...\n")
|
|---|
| 904 | if ($self->{'verbosity'} > 2);
|
|---|
| 905 |
|
|---|
| 906 | my $acro_array = &acronym::acronyms($textref);
|
|---|
| 907 |
|
|---|
| 908 | foreach my $acro (@$acro_array) {
|
|---|
| 909 |
|
|---|
| 910 | #check that this is the first time ...
|
|---|
| 911 | my $seen_before = "false";
|
|---|
| 912 | my $previous_data = $doc_obj->get_metadata($thissection, "Acronym");
|
|---|
| 913 | foreach my $thisAcro (@$previous_data) {
|
|---|
| 914 | if ($thisAcro eq $acro->to_string()) {
|
|---|
| 915 | $seen_before = "true";
|
|---|
| 916 | # print $outhandle " already seen ". $acro->to_string() . "\n"
|
|---|
| 917 | &gsprintf($outhandle, " {BasPlug.already_seen} " . $acro->to_string() . "\n")
|
|---|
| 918 | if ($self->{'verbosity'} >= 4);
|
|---|
| 919 | }
|
|---|
| 920 | }
|
|---|
| 921 |
|
|---|
| 922 | if ($seen_before eq "false") {
|
|---|
| 923 | #write it to the file ...
|
|---|
| 924 | $acro->write_to_file();
|
|---|
| 925 |
|
|---|
| 926 | #do the normal acronym
|
|---|
| 927 | $doc_obj->add_utf8_metadata($thissection, "Acronym", $acro->to_string());
|
|---|
| 928 | # print $outhandle " adding ". $acro->to_string() . "\n"
|
|---|
| 929 | &gsprintf($outhandle, " {BasPlug.adding} " . $acro->to_string() . "\n")
|
|---|
| 930 | if ($self->{'verbosity'} > 3);
|
|---|
| 931 | }
|
|---|
| 932 | }
|
|---|
| 933 |
|
|---|
| 934 | # print $outhandle " done extracting acronyms. \n"
|
|---|
| 935 | &gsprintf($outhandle, " {BasPlug.done_acronym_extract}\n")
|
|---|
| 936 | if ($self->{'verbosity'} > 2);
|
|---|
| 937 | }
|
|---|
| 938 |
|
|---|
| 939 | sub markup_acronyms {
|
|---|
| 940 | my $self = shift (@_);
|
|---|
| 941 | my ($text, $doc_obj, $thissection) = @_;
|
|---|
| 942 | my $outhandle = $self->{'outhandle'};
|
|---|
| 943 |
|
|---|
| 944 | # print $outhandle " marking up acronyms ...\n"
|
|---|
| 945 | &gsprintf($outhandle, " {BasPlug.marking_up_acronyms}...\n")
|
|---|
| 946 | if ($self->{'verbosity'} > 2);
|
|---|
| 947 |
|
|---|
| 948 | #self is passed in to check for verbosity ...
|
|---|
| 949 | $text = &acronym::markup_acronyms($text, $self);
|
|---|
| 950 |
|
|---|
| 951 | # print $outhandle " done marking up acronyms. \n"
|
|---|
| 952 | &gsprintf($outhandle, " {BasPlug.done_acronym_markup}\n")
|
|---|
| 953 | if ($self->{'verbosity'} > 2);
|
|---|
| 954 |
|
|---|
| 955 | return $text;
|
|---|
| 956 | }
|
|---|
| 957 |
|
|---|
| 958 | sub compile_stats {
|
|---|
| 959 | my $self = shift(@_);
|
|---|
| 960 | my ($stats) = @_;
|
|---|
| 961 |
|
|---|
| 962 | $stats->{'num_processed'} += $self->{'num_processed'};
|
|---|
| 963 | $stats->{'num_not_processed'} += $self->{'num_not_processed'};
|
|---|
| 964 | $stats->{'num_archives'} += $self->{'num_archives'};
|
|---|
| 965 |
|
|---|
| 966 | }
|
|---|
| 967 |
|
|---|
| 968 | sub associate_cover_image {
|
|---|
| 969 | my $self = shift(@_);
|
|---|
| 970 | my ($doc_obj, $filename) = @_;
|
|---|
| 971 |
|
|---|
| 972 | $filename =~ s/\.[^\\\/\.]+$/\.jpg/;
|
|---|
| 973 | if (-e $filename) {
|
|---|
| 974 | $doc_obj->associate_file($filename, "cover.jpg", "image/jpeg");
|
|---|
| 975 | } else {
|
|---|
| 976 | $filename =~ s/jpg$/JPG/;
|
|---|
| 977 | if (-e $filename) {
|
|---|
| 978 | $doc_obj->associate_file($filename, "cover.jpg", "image/jpeg");
|
|---|
| 979 | }
|
|---|
| 980 | }
|
|---|
| 981 | }
|
|---|
| 982 |
|
|---|
| 983 | 1;
|
|---|
