| 1 | ##################################################################### | ||
| 2 | # Copyright (C) 2004 Jörg Tiedemann <joerg@stp.ling.uu.se> | ||
| 3 | # | ||
| 4 | # This program is free software; you can redistribute it and/or modify | ||
| 5 | # it under the terms of the GNU General Public License as published by | ||
| 6 | # the Free Software Foundation; either version 2 of the License, or | ||
| 7 | # (at your option) any later version. | ||
| 8 | # | ||
| 9 | # This program is distributed in the hope that it will be useful, | ||
| 10 | # but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
| 11 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | ||
| 12 | # GNU General Public License for more details. | ||
| 13 | # | ||
| 14 | # You should have received a copy of the GNU General Public License | ||
| 15 | # along with this program; if not, write to the Free Software | ||
| 16 | # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA | ||
| 17 | # | ||
| 18 | # $Author$ | ||
| 19 | # $Id$ | ||
| 20 | # | ||
| 21 | ########################################################################### | ||
| 22 | # | ||
| 23 | # | ||
| 24 | # | ||
| 25 | # | ||
| 26 | ########################################################################### | ||
| 27 | |||
| 28 | |||
| 29 | package Uplug::Data::Lang; | ||
| 30 | |||
| 31 | use strict; | ||
| 32 | use vars qw( @ISA | ||
| 33 | $DEFAULTLANGUAGE $DEFAULTLANGUAGEFILE | ||
| 34 | $DEFAULTSPLITPATTERN $DEFAULTDELIMITER ); | ||
| 35 | |||
| 36 | use Uplug::Data; | ||
| 37 | use Uplug::Config; | ||
| 38 | |||
| 39 | #---------------------- | ||
| 40 | # Uplug::Data::Lang is derived from | ||
| 41 | # either Uplug::Data::XML | ||
| 42 | # either Uplug::Data::Tree | ||
| 43 | # or Uplug::Data::DOM (otherwise) | ||
| 44 | # depending on $Uplug::Data::DEFAULTDATATYPE | ||
| 45 | #---------------------- | ||
| 46 | |||
| 47 | @ISA=qw( Uplug::Data ); | ||
| 48 | |||
| 49 | $DEFAULTLANGUAGE = 'default'; | ||
| 50 | $DEFAULTLANGUAGEFILE = 'default.ini'; | ||
| 51 | $DEFAULTSPLITPATTERN = '\s+'; | ||
| 52 | $DEFAULTDELIMITER=$Uplug::Data::DEFAULTDELIMITER; | ||
| 53 | |||
| 54 | |||
| 55 | |||
| 56 | sub init{ | ||
| 57 | my $self=shift; | ||
| 58 | my $language=shift; | ||
| 59 | if ((defined $language) or (not ref($self->{LanguageData}))){ | ||
| 60 | $self->{LanguageData}={}; | ||
| 61 | if ((defined $language) and ($self->{language} ne $language)){ | ||
| 62 | $self->loadLanguageFile($language.'.ini'); | ||
| 63 | } | ||
| 64 | # if (not $language){$language=$DEFAULTLANGUAGE;} | ||
| 65 | # if ($self->{language} ne $language){ | ||
| 66 | # $self->{language}=$language; | ||
| 67 | # $self->loadLanguageFile($language.'.ini'); | ||
| 68 | # } | ||
| 69 | } | ||
| 70 | return $self->SUPER::init(@_); | ||
| 71 | } | ||
| 72 | |||
| 73 | sub clone{return Uplug::Data::Lang->new();} | ||
| 74 | |||
| 75 | sub loadLanguageFile{ | ||
| 76 | my $self=shift; | ||
| 77 | my $file=shift; | ||
| 78 | my $lang=$self->getLanguage(); | ||
| 79 | $self->{LanguageData}=&ReadConfig($file); | ||
| 80 | # &LoadIniData($self->{LanguageData},$file); | ||
| 81 | if (defined $self->{LanguageData}->{$lang}){ # we don't need | ||
| 82 | $self->{LanguageData}=$self->{LanguageData}->{$lang}; # a root node! | ||
| 83 | } | ||
| 84 | return $self->initLanguageData(); | ||
| 85 | } | ||
| 86 | |||
| 87 | sub setLanguage{ | ||
| 88 | my $self=shift; | ||
| 89 | my $lang=shift; | ||
| 90 | if ($lang ne $self->getLanguage()){ | ||
| 91 | $self->{language}=$lang; | ||
| 92 | return $self->loadLanguageFile($lang.'.ini'); | ||
| 93 | } | ||
| 94 | return 1; | ||
| 95 | } | ||
| 96 | |||
| 97 | sub getLanguage{return $_[0]->{language};} | ||
| 98 | sub language{return $_[0]->getLanguage();} | ||
| 99 | |||
| 100 | sub getLanguageData{ | ||
| 101 | my $self=shift; | ||
| 102 | my ($cat,$subcat,$attr)=@_; | ||
| 103 | my $data=$self->{LanguageData}; | ||
| 104 | if ((defined $cat) and (defined $data->{$cat})){ | ||
| 105 | if ((defined $subcat) and (defined $data->{$cat}->{$subcat})){ | ||
| 106 | if ((defined $attr) and | ||
| 107 | (defined $data->{$cat}->{$subcat}->{$attr})){ | ||
| 108 | return $data->{$cat}->{$subcat}->{$attr}; | ||
| 109 | } | ||
| 110 | elsif (defined $attr){return undef;} | ||
| 111 | return $data->{$cat}->{$subcat}; | ||
| 112 | } | ||
| 113 | elsif (defined $subcat){return undef;} | ||
| 114 | return $data->{$cat}; | ||
| 115 | } | ||
| 116 | elsif (defined $cat){return undef;} | ||
| 117 | return $data; | ||
| 118 | } | ||
| 119 | sub languageData{return $_[0]->getLanguageData(@_);} | ||
| 120 | |||
| 121 | |||
| 122 | |||
| 123 | #---------------------------------------------------------------------------- | ||
| 124 | # initLanguageData: | ||
| 125 | # | ||
| 126 | # initialize language specific data hashs! | ||
| 127 | # | ||
| 128 | |||
| 129 | |||
| 130 | sub initLanguageData{ | ||
| 131 | my $self=shift; | ||
| 132 | |||
| 133 | #---------------------------------------------------------------------------- | ||
| 134 | # make skip token hashs (from skip token arrays) | ||
| 135 | #---------------------------------------------------------------------------- | ||
| 136 | |||
| 137 | my @skip=('skip phrase before', | ||
| 138 | 'skip phrase after', | ||
| 139 | 'skip phrase at', | ||
| 140 | 'skip token', | ||
| 141 | 'non-phrase-starter', | ||
| 142 | 'non-phrase-ender'); | ||
| 143 | |||
| 144 | my $data=$self->getLanguageData(); | ||
| 145 | |||
| 146 | if (defined $$data{'phrases'}){ | ||
| 147 | foreach (@skip){ | ||
| 148 | if (defined $$data{'phrases'}{$_}){ | ||
| 149 | if (ref($$data{'phrases'}{$_}) eq 'ARRAY'){ | ||
| 150 | my @SkipWords=@{$$data{'phrases'}{$_}}; | ||
| 151 | %{$$data{'phrases'}{"$_ hash"}}=(); | ||
| 152 | &ArrayToHash(\@SkipWords,$$data{'phrases'}{"$_ hash"}); | ||
| 153 | } | ||
| 154 | } | ||
| 155 | } | ||
| 156 | } | ||
| 157 | |||
| 158 | #---------------------------------------------------------------------------- | ||
| 159 | # make skip token string type arrays | ||
| 160 | #---------------------------------------------------------------------------- | ||
| 161 | |||
| 162 | if (defined $$data{'phrases'}){ | ||
| 163 | foreach (@skip){ | ||
| 164 | if (defined $$data{'phrases'}{"$_ string type"}){ | ||
| 165 | if (ref($$data{'phrases'}{"$_ string type"}) ne 'ARRAY'){ | ||
| 166 | my @arr=$$data{'phrases'}{"$_ string type"}; | ||
| 167 | delete $$data{'phrases'}{"$_ string type"}; | ||
| 168 | @{$$data{'phrases'}{"$_ string type"}}=@arr; | ||
| 169 | } | ||
| 170 | } | ||
| 171 | |||
| 172 | } | ||
| 173 | } | ||
| 174 | |||
| 175 | #---------------------------------------------------------------------------- | ||
| 176 | # make stop word hashs (from stop word arrays) | ||
| 177 | #---------------------------------------------------------------------------- | ||
| 178 | |||
| 179 | if (defined $$data{'stop words'}){ | ||
| 180 | if (defined $$data{'stop words'}{'wordform'}){ | ||
| 181 | my @words=@{$$data{'stop words'}{'wordform'}}; | ||
| 182 | &ArrayToHash(\@words,$$data{'stop word hash'},$data); | ||
| 183 | } | ||
| 184 | if (defined $$data{'stop words'}{'classes'}){ | ||
| 185 | my ($cat,$subcat); | ||
| 186 | foreach $cat (@{$$data{'stop words'}{'classes'}}){ | ||
| 187 | if (defined $$data{$cat}){ | ||
| 188 | foreach $subcat (keys %{$$data{$cat}}){ | ||
| 189 | foreach (@{$$data{$cat}{$subcat}}){ | ||
| 190 | $$data{'stop word hash'}{$_}=1; | ||
| 191 | $$data{'stop word class hash'}{$cat}{$_}=1; | ||
| 192 | $$data{'stop word subclass hash'}{$cat}{$subcat}{$_}=1; | ||
| 193 | } | ||
| 194 | } | ||
| 195 | } | ||
| 196 | } | ||
| 197 | } | ||
| 198 | } | ||
| 199 | |||
| 200 | #---------------------------------------------------------------------------- | ||
| 201 | # define some special character classes | ||
| 202 | #---------------------------------------------------------------------------- | ||
| 203 | |||
| 204 | my $cat='character specifications'; | ||
| 205 | |||
| 206 | $$data{$cat}{'alphabetic'}=$$data{$cat}{'letter'}.$$data{$cat}{'hyphen'}; | ||
| 207 | $$data{$cat}{'alphanumeric'}= | ||
| 208 | $$data{$cat}{'letter'}. | ||
| 209 | $$data{$cat}{'hyphen'}. | ||
| 210 | $$data{$cat}{'digit'}. | ||
| 211 | $$data{$cat}{'numeric symbol'}; | ||
| 212 | $$data{$cat}{'numeric'}= | ||
| 213 | $$data{$cat}{'digit'}.$$data{$cat}{'numeric symbol'}; | ||
| 214 | |||
| 215 | #------- compatibility with old spelling errors ....... ------------------- | ||
| 216 | |||
| 217 | $$data{$cat}{'alpha'}=$$data{$cat}{'alphabetic'}; | ||
| 218 | $$data{$cat}{'vowels'}=$$data{$cat}{'vowel'}; | ||
| 219 | $$data{$cat}{'consonants'}=$$data{$cat}{'consonant'}; | ||
| 220 | $$data{$cat}{'vocals'}=$$data{$cat}{'vowel'}; | ||
| 221 | } | ||
| 222 | |||
| 223 | |||
| 224 | ##################################################################### | ||
| 225 | # check stop word classes | ||
| 226 | ##################################################################### | ||
| 227 | |||
| 228 | |||
| 229 | |||
| 230 | sub isStopWord{ | ||
| 231 | my $self=shift; | ||
| 232 | my ($str)=@_; | ||
| 233 | my $cat='stop word hash'; | ||
| 234 | my $data=$self->getLanguageData($cat); | ||
| 235 | if (ref($data) eq 'HASH'){ | ||
| 236 | if (defined $$data{$str}){ | ||
| 237 | return 1; | ||
| 238 | } | ||
| 239 | } | ||
| 240 | return 0; | ||
| 241 | } | ||
| 242 | |||
| 243 | |||
| 244 | ##################################################################### | ||
| 245 | # string processing functions | ||
| 246 | ##################################################################### | ||
| 247 | |||
| 248 | |||
| 249 | # split into VC sequences .... | ||
| 250 | |||
| 251 | sub splitIntoVC{ | ||
| 252 | my $self=shift; | ||
| 253 | my $str=shift; | ||
| 254 | |||
| 255 | my $cat='character specifications'; | ||
| 256 | my $Vowel=$self->getLanguageData($cat,'vowel'); | ||
| 257 | my $Consonant=$self->getLanguageData($cat,'consonant'); | ||
| 258 | if (not $Vowel){return ($str);} | ||
| 259 | if (not $Consonant){return ($str);} | ||
| 260 | |||
| 261 | my @arr=(); | ||
| 262 | while ($str ne ''){ | ||
| 263 | if ($str=~/^([$Vowel]+)([^$Vowel].*)$/){ | ||
| 264 | push (@arr,$1); | ||
| 265 | $str=$2; | ||
| 266 | } | ||
| 267 | elsif ($str=~/^([$Consonant]+)([^$Consonant].*)$/){ | ||
| 268 | push (@arr,$1); | ||
| 269 | $str=$2; | ||
| 270 | } | ||
| 271 | elsif ($str=~/^([^$Vowel$Consonant]+)(.*)$/){ | ||
| 272 | push (@arr,$1); | ||
| 273 | $str=$2; | ||
| 274 | } | ||
| 275 | else{ | ||
| 276 | push @arr,$str; | ||
| 277 | $str=''; | ||
| 278 | } | ||
| 279 | } | ||
| 280 | return @arr; | ||
| 281 | } | ||
| 282 | |||
| 283 | |||
| 284 | |||
| 285 | #----------------------------------------------------------------------------- | ||
| 286 | # getTokens | ||
| 287 | # | ||
| 288 | # get all tokens which match the parameters in %{$param} | ||
| 289 | # | ||
| 290 | |||
| 291 | sub getTokens{ | ||
| 292 | my $self=shift; | ||
| 293 | my $param=shift; | ||
| 294 | my $accepted=shift; | ||
| 295 | |||
| 296 | if (ref($accepted) ne 'ARRAY'){ | ||
| 297 | $accepted=[]; | ||
| 298 | } | ||
| 299 | my @nodes; | ||
| 300 | my @tokens=(); | ||
| 301 | |||
| 302 | if (not defined $$param{'token label'}){ | ||
| 303 | @nodes=$self->contentElements(); | ||
| 304 | } | ||
| 305 | else{ | ||
| 306 | @nodes=$self->findNodes($$param{'token label'}); | ||
| 307 | } | ||
| 308 | my @accepted=(); | ||
| 309 | foreach my $n (@nodes){ | ||
| 310 | if (defined (my $t=$self->checkTokenParameter($n,$param))){ | ||
| 311 | push(@{$accepted},$n); | ||
| 312 | push(@tokens,$t); | ||
| 313 | } | ||
| 314 | } | ||
| 315 | return @tokens; | ||
| 316 | } | ||
| 317 | |||
| 318 | |||
| 319 | #--------------------------------------------------------------- | ||
| 320 | # GetChunks | ||
| 321 | # | ||
| 322 | # get all chunks from data which have been labeled with $label | ||
| 323 | # | ||
| 324 | |||
| 325 | |||
| 326 | sub getChunks{ | ||
| 327 | my $self=shift; | ||
| 328 | my $param=shift; | ||
| 329 | my ($phraseNodes,$tokenNodes,$tokens,$del)=@_; | ||
| 330 | |||
| 331 | my @phrases=(); | ||
| 332 | if (not defined $del){$del=$DEFAULTDELIMITER;} | ||
| 333 | if (not ref($tokenNodes)){ | ||
| 334 | $tokens=[]; | ||
| 335 | $tokenNodes=[]; | ||
| 336 | @{$tokens}=$self->getTokens($param,$tokenNodes); | ||
| 337 | } | ||
| 338 | |||
| 339 | #---------------------------------------------------------------------- | ||
| 340 | # check certain markup for additional phrases | ||
| 341 | # (chunks etc.) | ||
| 342 | |||
| 343 | if (defined $$param{'chunks'}){ | ||
| 344 | my @chunks=$self->findNodes($$param{'chunks'}); | ||
| 345 | foreach my $c (@chunks){ | ||
| 346 | if (not ref($c)){next;} | ||
| 347 | my @nodes; | ||
| 348 | if (defined $$param{'token label'}){ | ||
| 349 | # @nodes=$self->findNodes($$param{'token label'},undef,$c); | ||
| 350 | @nodes=$c->getElementsByTagName($$param{'token label'}); | ||
| 351 | } | ||
| 352 | else{@nodes=$c->getElementsByTagName('*');} | ||
| 353 | # else{@nodes=$self->findNodes('*',undef,$c);} | ||
| 354 | |||
| 355 | my @chunk=(); # find accepted tokens | ||
| 356 | my @chunkNodes=(); # in the chunk | ||
| 357 | foreach my $x (@nodes){ | ||
| 358 | foreach my $y (0..$#{$tokenNodes}){ | ||
| 359 | if ($x==$$tokenNodes[$y]){ | ||
| 360 | push(@chunkNodes,$$tokenNodes[$y]); | ||
| 361 | if (ref($tokens) eq 'ARRAY'){ | ||
| 362 | push(@chunk,$$tokens[$y]); | ||
| 363 | } | ||
| 364 | } | ||
| 365 | } | ||
| 366 | } | ||
| 367 | |||
| 368 | if (@chunkNodes){ | ||
| 369 | if (ref($phraseNodes)){ | ||
| 370 | my $idx=$#{$phraseNodes}+1; | ||
| 371 | @{$$phraseNodes[$idx]}=@chunkNodes; | ||
| 372 | } | ||
| 373 | if (@chunk){ | ||
| 374 | push (@phrases,join $del,@chunk); | ||
| 375 | } | ||
| 376 | else{ | ||
| 377 | my $idx=$#phrases+1; | ||
| 378 | push (@phrases,'#chunk'.$idx); | ||
| 379 | } | ||
| 380 | # $phrases[$idx]="chunk:".$phrases[$idx] | ||
| 381 | } | ||
| 382 | } | ||
| 383 | } | ||
| 384 | return @phrases; | ||
| 385 | } | ||
| 386 | |||
| 387 | |||
| 388 | #--------------------------------------------------------------- | ||
| 389 | # GetNgrams | ||
| 390 | # | ||
| 391 | # get all ngrams from data | ||
| 392 | # | ||
| 393 | |||
| 394 | |||
| 395 | sub getNgrams{ | ||
| 396 | my $self=shift; | ||
| 397 | my $param=shift; | ||
| 398 | my ($phraseNodes,$tokenNodes,$tokens,$del)=@_; | ||
| 399 | |||
| 400 | my $minTokenLength=1; | ||
| 401 | my $maxNgramLength=1; | ||
| 402 | if (defined $$param{'minimal length'}){ | ||
| 403 | $minTokenLength=$$param{'minimal length'}; | ||
| 404 | } | ||
| 405 | if (defined $$param{'maximal ngram length'}){ | ||
| 406 | $maxNgramLength=$$param{'maximal ngram length'}; | ||
| 407 | } | ||
| 408 | if ($maxNgramLength<2){return;} | ||
| 409 | |||
| 410 | #---------------------------------------------------------------------- | ||
| 411 | # 1) get tokens if necessary | ||
| 412 | #---------------------------------------------------------------------- | ||
| 413 | |||
| 414 | my @phrases=(); | ||
| 415 | if (not defined $del){$del=$DEFAULTDELIMITER;} | ||
| 416 | if (not ref($tokenNodes)){ | ||
| 417 | $tokens=[]; | ||
| 418 | $tokenNodes=[]; | ||
| 419 | @{$tokens}=$self->getTokens($param,$tokenNodes); | ||
| 420 | } | ||
| 421 | if (not ref($tokens)){$tokens=[];} | ||
| 422 | foreach (0..$#{$tokenNodes}){ | ||
| 423 | if (not defined $$tokens[$_]){ | ||
| 424 | $$tokens[$_]=$self->content($$tokenNodes[$_]); | ||
| 425 | } | ||
| 426 | } | ||
| 427 | |||
| 428 | #---------------------------------------------------------------------- | ||
| 429 | # 2) compute all N-grams (N>1) | ||
| 430 | #---------------------------------------------------------------------- | ||
| 431 | |||
| 432 | my @Ngram=(); | ||
| 433 | my @NgramNodes=(); | ||
| 434 | my @words=@{$tokens}; # save tokens in a new array | ||
| 435 | my @nodes=@{$tokenNodes}; # and corresponding nodes, too | ||
| 436 | push (@Ngram,shift(@words)); # (because we're shifting all | ||
| 437 | push (@NgramNodes,shift(@nodes)); # tokens from the array) | ||
| 438 | |||
| 439 | while (@words){ | ||
| 440 | |||
| 441 | my $t=shift(@words); | ||
| 442 | my $n=shift(@nodes); | ||
| 443 | |||
| 444 | # print STDERR "length: $minTokenLength\n"; | ||
| 445 | # if (length($t)<$minTokenLength){ | ||
| 446 | # print STDERR "skip $t\n"; | ||
| 447 | # next; | ||
| 448 | # } | ||
| 449 | # if ((length($t)<$minTokenLength) or ($self->skipToken($t))){ | ||
| 450 | |||
| 451 | if ($self->skipToken($t)){ | ||
| 452 | $self->addNgrams(\@Ngram,\@NgramNodes,\@phrases,$phraseNodes,$del); | ||
| 453 | # print STDERR "skip $t\n"; | ||
| 454 | @Ngram=(); | ||
| 455 | @NgramNodes=(); | ||
| 456 | next; | ||
| 457 | } | ||
| 458 | elsif ($self->skipPhraseAt($t)){ | ||
| 459 | $self->addNgrams(\@Ngram,\@NgramNodes,\@phrases,$phraseNodes,$del); | ||
| 460 | # print STDERR "skip at $t\n"; | ||
| 461 | @Ngram=(); | ||
| 462 | @NgramNodes=(); | ||
| 463 | next; | ||
| 464 | } | ||
| 465 | elsif ($self->skipPhraseBefore($t)){ | ||
| 466 | $self->addNgrams(\@Ngram,\@NgramNodes,\@phrases,$phraseNodes,$del); | ||
| 467 | @Ngram=(); | ||
| 468 | @NgramNodes=(); | ||
| 469 | # print STDERR "skip before $t\n"; | ||
| 470 | } | ||
| 471 | |||
| 472 | push (@Ngram,$t); | ||
| 473 | push (@NgramNodes,$n); | ||
| 474 | |||
| 475 | if (scalar @Ngram == $maxNgramLength){ | ||
| 476 | $self->addNgrams(\@Ngram,\@NgramNodes,\@phrases,$phraseNodes,$del); | ||
| 477 | shift (@Ngram); | ||
| 478 | shift (@NgramNodes); | ||
| 479 | # print STDERR "length = max\n"; | ||
| 480 | } | ||
| 481 | if ($self->skipPhraseAfter($t)){ | ||
| 482 | $self->addNgrams(\@Ngram,\@NgramNodes,\@phrases,$phraseNodes,$del); | ||
| 483 | @Ngram=(); | ||
| 484 | @NgramNodes=(); | ||
| 485 | # print STDERR "skip after $t\n"; | ||
| 486 | } | ||
| 487 | } | ||
| 488 | $self->addNgrams(\@Ngram,\@NgramNodes,\@phrases,$phraseNodes,$del); | ||
| 489 | return @phrases; | ||
| 490 | } | ||
| 491 | |||
| 492 | |||
| 493 | #----------------------------------------------------------- | ||
| 494 | # $OBJ->getPhrases($param,\@phraseNodes) | ||
| 495 | # | ||
| 496 | # get phrase candidates | ||
| 497 | # @phraseNodes: list of lists of data-nodes | ||
| 498 | # | ||
| 499 | # | ||
| 500 | |||
| 501 | |||
| 502 | sub getPhrases{ | ||
| 503 | my $self=shift; | ||
| 504 | my $param=shift; | ||
| 505 | my ($phraseNodes,$tokenNodes,$tokens,$del)=@_; | ||
| 506 | |||
| 507 | if (not defined $del){$del=$DEFAULTDELIMITER;} | ||
| 508 | if (not ref($tokenNodes)){ | ||
| 509 | $tokens=[]; | ||
| 510 | $tokenNodes=[]; | ||
| 511 | @{$tokens}=$self->getTokens($param,$tokenNodes); | ||
| 512 | } | ||
| 513 | |||
| 514 | my @phrases=$self->getChunks($param,$phraseNodes,$tokenNodes,$tokens,$del); | ||
| 515 | my @ngrams=$self->getNgrams($param,$phraseNodes,$tokenNodes,$tokens,$del); | ||
| 516 | push (@phrases,@ngrams); | ||
| 517 | |||
| 518 | #---------------------------------------------------------------------- | ||
| 519 | # add all single tokens that match the required string type | ||
| 520 | #---------------------------------------------------------------------- | ||
| 521 | |||
| 522 | # my $minTokenLength=1; | ||
| 523 | # if (defined $$param{'minimal length'}){ | ||
| 524 | # $minTokenLength=$$param{'minimal length'}; | ||
| 525 | # } | ||
| 526 | foreach (0..$#{$tokenNodes}){ | ||
| 527 | if (not ref($tokens) or $self->skipToken($$tokens[$_])){next;} | ||
| 528 | # if (length($$tokens[$_])<$minTokenLength){next;} | ||
| 529 | my $idx=$#phrases+1; | ||
| 530 | if (ref($phraseNodes)){ | ||
| 531 | @{$$phraseNodes[$idx]}=($$tokenNodes[$_]); | ||
| 532 | } | ||
| 533 | if (ref($tokens)){ | ||
| 534 | push (@phrases,$$tokens[$_]); | ||
| 535 | } | ||
| 536 | else{ | ||
| 537 | push (@phrases,'#token'.$idx); | ||
| 538 | } | ||
| 539 | } | ||
| 540 | |||
| 541 | @phrases=$self->removeIdenticalPhrases($phraseNodes,\@phrases); | ||
| 542 | return @phrases; | ||
| 543 | } | ||
| 544 | |||
| 545 | |||
| 546 | |||
| 547 | sub getPhrasePosition{ | ||
| 548 | my $self=shift; | ||
| 549 | my ($phr)=@_; | ||
| 550 | if (ref($phr) ne 'ARRAY'){return 0;} | ||
| 551 | my $tok=$phr->[0]; | ||
| 552 | if (not ref($tok)){return 0;} | ||
| 553 | my $tokID=$self->attribute($tok,'id'); | ||
| 554 | if ($tokID=~/(\A|[^0-9])([0-9]+)$/){ | ||
| 555 | return $2; | ||
| 556 | } | ||
| 557 | return undef; | ||
| 558 | } | ||
| 559 | |||
| 560 | |||
| 561 | sub getPhraseContent{ | ||
| 562 | my $self=shift; | ||
| 563 | my $nodes=shift; | ||
| 564 | my $param=shift; | ||
| 565 | my ($del)=@_; | ||
| 566 | |||
| 567 | if (not defined $del){$del=$DEFAULTDELIMITER;} | ||
| 568 | my @phrase=(); | ||
| 569 | foreach my $n (@{$nodes}){ | ||
| 570 | if (my $t=$self->checkTokenParameter($n,$param)){ | ||
| 571 | push(@phrase,$t); | ||
| 572 | } | ||
| 573 | } | ||
| 574 | return join $del,@phrase; | ||
| 575 | } | ||
| 576 | |||
| 577 | sub getPhraseFeature{ | ||
| 578 | my $self=shift; | ||
| 579 | my ($phraseNodes,$param)=@_; | ||
| 580 | |||
| 581 | if (ref($phraseNodes) ne 'ARRAY'){return undef;} | ||
| 582 | if (not @{$phraseNodes}){return '';} | ||
| 583 | |||
| 584 | |||
| 585 | if (not ref($param)){return $self->getPhraseContent($phraseNodes,$param);} | ||
| 586 | if (ref($param->{features}) ne 'HASH'){ | ||
| 587 | return $self->getPhraseContent($phraseNodes,$param); | ||
| 588 | } | ||
| 589 | |||
| 590 | my $FeatureString=''; | ||
| 591 | foreach my $f (sort keys %{$param->{features}}){ # for all features: | ||
| 592 | |||
| 593 | my %NodeHash=(); # feature node names | ||
| 594 | my @FeatNodes=(); # array of unique feature nodes | ||
| 595 | my @AllFeatNodes=(); # array of all feature nodes (for suffix/prefix) | ||
| 596 | my $attr=$f; # initialize feature-attribute | ||
| 597 | my $path=undef; # initialize feature-node path | ||
| 598 | # print STDERR "$f\n"; | ||
| 599 | if ($f=~/^(.*)\:([^:]+)$/){ # first part is the path | ||
| 600 | ($path,$attr)=($1,$2); # second is the attribute | ||
| 601 | } | ||
| 602 | |||
| 603 | foreach my $t (@{$phraseNodes}){ # for all token nodes: | ||
| 604 | my $node=$self->getFeatureNode($t,$path); # find the feature node | ||
| 605 | push(@AllFeatNodes,$node); # save feature nodes | ||
| 606 | if (not ref($node)){next;} | ||
| 607 | if (not defined $NodeHash{$node}){ | ||
| 608 | $NodeHash{$node}=1; # save unique | ||
| 609 | push(@FeatNodes,$node); # feature nodes | ||
| 610 | } | ||
| 611 | } | ||
| 612 | |||
| 613 | #--------------------------------------------------------------- | ||
| 614 | # check if the left neighbour of the first token has the same | ||
| 615 | # feature node as the first token itself ---> add prefix ')' | ||
| 616 | # check if the right neighbour of the last token has the same | ||
| 617 | # feature node as the last token itself ---> add suffix '(' | ||
| 618 | # | ||
| 619 | # e.g., if the feature node is a chunk-node | ||
| 620 | # and the neighbours of the current phrase belong to the same chunk | ||
| 621 | |||
| 622 | my $prefix; | ||
| 623 | my $suffix; | ||
| 624 | my $node=$self->getFeatureNode($$phraseNodes[0],'left:'.$path); | ||
| 625 | if ($node and ($node eq $AllFeatNodes[0])){ | ||
| 626 | $prefix=')'; | ||
| 627 | } | ||
| 628 | my $node=$self->getFeatureNode($$phraseNodes[-1],'right:'.$path); | ||
| 629 | if ($node and ($node eq $AllFeatNodes[-1])){ | ||
| 630 | $suffix='('; | ||
| 631 | } | ||
| 632 | |||
| 633 | #--------------------------------------------------------------- | ||
| 634 | |||
| 635 | |||
| 636 | my $feature=''; | ||
| 637 | my $pattern=$param->{features}->{$f}; # substitution pattern | ||
| 638 | my $re=undef; # regular expression | ||
| 639 | my $subst=undef; # substitution | ||
| 640 | if ($pattern=~/(.*)\/(.*)/){ # subst.-pattern found: | ||
| 641 | $re=$1; # set variables | ||
| 642 | $subst=$2; | ||
| 643 | } | ||
| 644 | |||
| 645 | foreach my $n (@FeatNodes){ # for all feature nodes: | ||
| 646 | my $value; # get feature string | ||
| 647 | if ($attr eq '#text'){ | ||
| 648 | $value=$self->content($n); | ||
| 649 | } | ||
| 650 | else{ | ||
| 651 | $value=$self->attribute($n,$attr); | ||
| 652 | } | ||
| 653 | if (defined $re){ | ||
| 654 | # eval { $value=~s/$re/$subst/; } | ||
| 655 | eval "\$value=~s/$re/$subst/;"; # change thos ?!? | ||
| 656 | } | ||
| 657 | $feature.=$value.' '; | ||
| 658 | } | ||
| 659 | chop $feature; | ||
| 660 | $FeatureString.=$prefix.$feature.$suffix.':'; # put everything together | ||
| 661 | } | ||
| 662 | chop $FeatureString; | ||
| 663 | return $FeatureString; | ||
| 664 | } | ||
| 665 | |||
| 666 | |||
| 667 | sub getFeatureNode{ | ||
| 668 | my $self=shift; | ||
| 669 | return $self->moveTo(@_); | ||
| 670 | } | ||
| 671 | |||
| 672 | |||
| 673 | #----------------------------------------------------------- | ||
| 674 | # | ||
| 675 | |||
| 676 | sub removeIdenticalPhrases{ | ||
| 677 | my $self=shift; | ||
| 678 | my ($nodes,$phrases)=@_; | ||
| 679 | |||
| 680 | if ((not ref($nodes)) and (ref($phrases))){ # easy! just check strings! | ||
| 681 | my %hash=(); | ||
| 682 | foreach (@{$phrases}){$hash{$_}=1;} | ||
| 683 | return keys %hash; | ||
| 684 | } | ||
| 685 | if (ref($nodes) ne 'ARRAY'){return undef;} | ||
| 686 | my @accNodes=(); | ||
| 687 | my @accPhrases=(); | ||
| 688 | my %hash; | ||
| 689 | foreach my $p (0..$#{$nodes}){ | ||
| 690 | my $key=join "\x00",@{$$nodes[$p]}; | ||
| 691 | if (not defined $hash{$key}){ | ||
| 692 | $hash{$key}=1; | ||
| 693 | push(@accNodes,$$nodes[$p]); | ||
| 694 | if (ref($phrases)){ | ||
| 695 | push(@accPhrases,$$phrases[$p]); | ||
| 696 | } | ||
| 697 | } | ||
| 698 | # else{print STDERR "remove phrase $$phrases[$p]\n";} | ||
| 699 | } | ||
| 700 | @{$nodes}=@accNodes; | ||
| 701 | return @accPhrases; | ||
| 702 | } | ||
| 703 | |||
| 704 | |||
| 705 | #--------------------------------- | ||
| 706 | # add all Ngrams and sub-Ngrams | ||
| 707 | # | ||
| 708 | |||
| 709 | sub addNgrams{ | ||
| 710 | my $self=shift; | ||
| 711 | my $tokens=shift; | ||
| 712 | my $nodes=shift; | ||
| 713 | |||
| 714 | if ($#{$tokens}<1){return;} # at least bigram! | ||
| 715 | $self->addNgram($tokens,$nodes,@_); # add this Ngram | ||
| 716 | |||
| 717 | my $t=shift @{$tokens}; # recursively add N-1_grams | ||
| 718 | my $n=shift @{$nodes}; # * without the initial token | ||
| 719 | $self->addNgrams($tokens,$nodes,@_); | ||
| 720 | unshift(@{$tokens},$t); | ||
| 721 | unshift(@{$nodes},$n); | ||
| 722 | $t=pop @{$tokens}; # * without the final token | ||
| 723 | $n=pop @{$nodes}; | ||
| 724 | $self->addNgrams($tokens,$nodes,@_); | ||
| 725 | push(@{$tokens},$t); # always restore | ||
| 726 | push(@{$nodes},$n); # the original arrays! | ||
| 727 | } | ||
| 728 | |||
| 729 | #--------------------------------- | ||
| 730 | # add Ngram | ||
| 731 | # | ||
| 732 | |||
| 733 | sub addNgram{ | ||
| 734 | my $self=shift; | ||
| 735 | my ($tokens,$nodes,$ngrams,$ngramNodes,$del)=@_; | ||
| 736 | |||
| 737 | if ($#{$tokens}<1){return;} # at least bigram | ||
| 738 | if (not $self->isNonStarter($$tokens[0])){ # no non-starter | ||
| 739 | if (not $self->isNonEnder($$tokens[-1])){ # no non-ender | ||
| 740 | if (not defined $del){$del=$DEFAULTDELIMITER;} | ||
| 741 | if (ref($ngramNodes) eq 'ARRAY'){ | ||
| 742 | my $idx=$#{$ngramNodes}+1; | ||
| 743 | @{$$ngramNodes[$idx]}=@{$nodes}; | ||
| 744 | } | ||
| 745 | if (ref($ngrams) eq 'ARRAY'){ | ||
| 746 | push(@{$ngrams},join $del,@{$tokens}); | ||
| 747 | } | ||
| 748 | } | ||
| 749 | } | ||
| 750 | } | ||
| 751 | |||
| 752 | |||
| 753 | |||
| 754 | |||
| 755 | |||
| 756 | sub checkTokenParameter{ | ||
| 757 | my $self=shift; | ||
| 758 | my $node=shift; | ||
| 759 | my $param=shift; | ||
| 760 | |||
| 761 | if (not ref($node)){return undef;} | ||
| 762 | my $token=$self->content($node); | ||
| 763 | if (ref($param) ne 'HASH'){return $token;} | ||
| 764 | |||
| 765 | if (defined $$param{'use attribute'}){ | ||
| 766 | my $attr=$self->attribute($node,$$param{'use attribute'}); | ||
| 767 | if (defined $attr){$token=$attr;} | ||
| 768 | # else{$token='_undef';} | ||
| 769 | } | ||
| 770 | if ($$param{'grep token'}){ | ||
| 771 | if (not $self->isStringType($token,$$param{'grep token'})){ | ||
| 772 | return undef; | ||
| 773 | } | ||
| 774 | } | ||
| 775 | if (defined $$param{'minimal length'}){ | ||
| 776 | if (length($token)<$$param{'minimal length'}){ | ||
| 777 | return undef; | ||
| 778 | } | ||
| 779 | } | ||
| 780 | if ($$param{'lower case'}){ | ||
| 781 | return $self->lowerCase($token); | ||
| 782 | } | ||
| 783 | return $token; | ||
| 784 | } | ||
| 785 | |||
| 786 | |||
| 787 | |||
| 788 | sub makeLowInitial{ | ||
| 789 | my $self=shift; | ||
| 790 | my $string=shift(@_); | ||
| 791 | my $LowerCaseLetter=$self->getLanguageData('character specifications', | ||
| 792 | 'lower case letter'); | ||
| 793 | if (not $LowerCaseLetter){return $string;} | ||
| 794 | if ($string=~/^(.)[$LowerCaseLetter]/){ | ||
| 795 | my $low=$self->lowerCase($1); | ||
| 796 | $string=~s/^./$low/; | ||
| 797 | } | ||
| 798 | elsif ($string=~/^.$/){ | ||
| 799 | $string=$self->LowerCase($string); | ||
| 800 | } | ||
| 801 | return $string; | ||
| 802 | } | ||
| 803 | |||
| 804 | # convert string to low case characters | ||
| 805 | |||
| 806 | sub lowerCase { | ||
| 807 | my $self=shift; | ||
| 808 | my $string=shift(@_); | ||
| 809 | my $LowerCaseLetter= | ||
| 810 | $self->getLanguageData('character specifications','lower case letter'); | ||
| 811 | my $UpperCaseLetter= | ||
| 812 | $self->getLanguageData('character specifications','upper case letter'); | ||
| 813 | |||
| 814 | if ((not $UpperCaseLetter) or (not $LowerCaseLetter)){ | ||
| 815 | return lc($string); | ||
| 816 | } | ||
| 817 | |||
| 818 | eval "\$string\=\~tr/$UpperCaseLetter/$LowerCaseLetter/;"; | ||
| 819 | return $string; | ||
| 820 | } | ||
| 821 | |||
| 822 | # get number of alphabetic characters in string | ||
| 823 | |||
| 824 | sub containsAlpha { | ||
| 825 | my $self=shift; | ||
| 826 | my $string=shift(@_); | ||
| 827 | my $Letter=$self->getLanguageData('character specifications','letter'); | ||
| 828 | my $result=eval("\$string\=\~tr/$Letter//"); | ||
| 829 | return $result; | ||
| 830 | } | ||
| 831 | |||
| 832 | sub containsNumeric { | ||
| 833 | my $self=shift; | ||
| 834 | my $string=shift(@_); | ||
| 835 | my $Numeric=$self->getLanguageData('character specifications','numeric'); | ||
| 836 | if (not $Numeric){return 0;} | ||
| 837 | my $pattern="\[$Numeric\]"; | ||
| 838 | return $string=~/$pattern/; | ||
| 839 | } | ||
| 840 | |||
| 841 | sub isVowel { | ||
| 842 | my $self=shift; | ||
| 843 | my $string=shift; | ||
| 844 | my $Vowel=$self->getLanguageData('character specifications','vowel'); | ||
| 845 | my $pattern="\[$Vowel\]\+"; | ||
| 846 | if (not $Vowel){return 0;} | ||
| 847 | return $string=~/^$pattern$/; | ||
| 848 | } | ||
| 849 | |||
| 850 | sub isConsonant { | ||
| 851 | my $self=shift; | ||
| 852 | my $string=shift; | ||
| 853 | my $Consonant= | ||
| 854 | $self->getLanguageData('character specifications','consonant'); | ||
| 855 | my $pattern="\[$Consonant\]\+"; | ||
| 856 | if (not $Consonant){return 0;} | ||
| 857 | return $string=~/^$pattern$/; | ||
| 858 | } | ||
| 859 | |||
| 860 | sub isLetter { | ||
| 861 | my $self=shift; | ||
| 862 | my $string=shift; | ||
| 863 | my $Letter=$self->getLanguageData('character specifications','letter'); | ||
| 864 | if (not $Letter){return 0;} | ||
| 865 | my $pattern="\[$Letter\]"; | ||
| 866 | return $string=~/^$pattern$/; | ||
| 867 | } | ||
| 868 | |||
| 869 | sub isLetterSeq { | ||
| 870 | my $self=shift; | ||
| 871 | my $string=shift; | ||
| 872 | my $Letter=$self->getLanguageData('character specifications','letter'); | ||
| 873 | if (not $Letter){return 0;} | ||
| 874 | my $pattern="\[$Letter\]\+"; | ||
| 875 | return $string=~/^$pattern$/; | ||
| 876 | } | ||
| 877 | |||
| 878 | sub isAlphabetic { | ||
| 879 | my $self=shift; | ||
| 880 | my $string=shift; | ||
| 881 | my $Alphabetic= | ||
| 882 | $self->getLanguageData('character specifications','alphabetic'); | ||
| 883 | if (not $Alphabetic){return 0;} | ||
| 884 | my $pattern="\[$Alphabetic\]\+"; | ||
| 885 | return $string=~/^$pattern$/; | ||
| 886 | } | ||
| 887 | |||
| 888 | sub isAlphanumeric { | ||
| 889 | my $self=shift; | ||
| 890 | my $string=shift; | ||
| 891 | my $Alphanumeric= | ||
| 892 | $self->getLanguageData('character specifications','alphanumeric'); | ||
| 893 | if (not $Alphanumeric){return 0;} | ||
| 894 | my $pattern="\[$Alphanumeric\]\+"; | ||
| 895 | return $string=~/^$pattern$/; | ||
| 896 | } | ||
| 897 | |||
| 898 | sub isNumeric { | ||
| 899 | my $self=shift; | ||
| 900 | my $string=shift; | ||
| 901 | my $Numeric= | ||
| 902 | $self->getLanguageData('character specifications','numeric'); | ||
| 903 | if (not $Numeric){return 0;} | ||
| 904 | my $pattern="\[$Numeric\]\+"; | ||
| 905 | return $string=~/^$pattern$/; | ||
| 906 | } | ||
| 907 | |||
| 908 | sub isNumber { | ||
| 909 | my $self=shift; | ||
| 910 | my $string=shift; | ||
| 911 | my $Digit= | ||
| 912 | $self->getLanguageData('character specifications','digit'); | ||
| 913 | if (not $Digit){return 0;} | ||
| 914 | my $pattern="\[$Digit\]\+"; | ||
| 915 | return $string=~/^$pattern$/; | ||
| 916 | } | ||
| 917 | |||
| 918 | sub isPunctuation { | ||
| 919 | my $self=shift; | ||
| 920 | my $string=shift; | ||
| 921 | my $Punctuation= | ||
| 922 | $self->getLanguageData('character specifications','punctuation'); | ||
| 923 | if (not $Punctuation){return 0;} | ||
| 924 | my $pattern="\[$Punctuation\]\+"; | ||
| 925 | return $string=~/^$pattern$/; | ||
| 926 | } | ||
| 927 | |||
| 928 | |||
| 929 | |||
| 930 | |||
| 931 | sub isStringType{ | ||
| 932 | my $self=shift; | ||
| 933 | my ($string,$type)=@_; | ||
| 934 | if ($type eq 'all'){ | ||
| 935 | return 1; | ||
| 936 | } | ||
| 937 | my $CharacterSpec=$self->getLanguageData('character specifications'); | ||
| 938 | |||
| 939 | my $pattern; | ||
| 940 | my $mod; | ||
| 941 | if ($type=~/^not[\_\s](.*)$/){ | ||
| 942 | $mod='not'; | ||
| 943 | $type=$1; | ||
| 944 | } | ||
| 945 | if ($type=~/^contains[\_\s](.*)$/){ | ||
| 946 | $mod.='contains'; | ||
| 947 | $type=$1; | ||
| 948 | } | ||
| 949 | if (defined $$CharacterSpec{$type}){ | ||
| 950 | $pattern=$$CharacterSpec{$type}; | ||
| 951 | if (not $pattern){return 1;} | ||
| 952 | if ($mod eq 'not'){ | ||
| 953 | return $string!~/^[$pattern]+$/; | ||
| 954 | } | ||
| 955 | elsif ($mod eq 'contains'){ | ||
| 956 | return $string=~/[$pattern]/; | ||
| 957 | } | ||
| 958 | elsif ($mod eq 'notcontains'){ | ||
| 959 | return $string!~/[$pattern]/; | ||
| 960 | } | ||
| 961 | else{ | ||
| 962 | return $string=~/^[$pattern]+$/; | ||
| 963 | } | ||
| 964 | } | ||
| 965 | return 1; | ||
| 966 | } | ||
| 967 | |||
| 968 | sub grepStringType{ | ||
| 969 | my $self=shift; | ||
| 970 | my ($list,$type)=@_; | ||
| 971 | return grep ($self->isStringType($_,$type),@{$list}); | ||
| 972 | } | ||
| 973 | |||
| 974 | sub grepStringTypeElements{ | ||
| 975 | my $self=shift; | ||
| 976 | my ($list,$type)=@_; | ||
| 977 | return grep ($self->IsStringType($$list[$_],$type),(0..$#{$list})); | ||
| 978 | } | ||
| 979 | |||
| 980 | |||
| 981 | |||
| 982 | sub isNonStarter{ | ||
| 983 | my $self=shift; | ||
| 984 | my $token=shift; | ||
| 985 | |||
| 986 | my $skip=$self->getLanguageData('phrases','non-phrase-starter hash'); | ||
| 987 | if (ref($skip) eq 'HASH'){ | ||
| 988 | if (exists $$skip{$token}){ | ||
| 989 | return 1; | ||
| 990 | } | ||
| 991 | } | ||
| 992 | my $skip= | ||
| 993 | $self->getLanguageData('phrases','non-phrase-starter string type'); | ||
| 994 | if (ref($skip) eq 'ARRAY'){ | ||
| 995 | foreach (@{$skip}){ | ||
| 996 | if (&IsStringType($token,$_)){ | ||
| 997 | return 1; | ||
| 998 | } | ||
| 999 | } | ||
| 1000 | } | ||
| 1001 | return 0; | ||
| 1002 | } | ||
| 1003 | |||
| 1004 | sub isNonEnder{ | ||
| 1005 | my $self=shift; | ||
| 1006 | my $token=shift; | ||
| 1007 | |||
| 1008 | my $skip=$self->getLanguageData('phrases','non-phrase-ender hash'); | ||
| 1009 | if (ref($skip) eq 'HASH'){ | ||
| 1010 | if (exists $$skip{$token}){ | ||
| 1011 | return 1; | ||
| 1012 | } | ||
| 1013 | } | ||
| 1014 | my $skip=$self->getLanguageData('phrases','non-phrase-ender string type'); | ||
| 1015 | if (ref($skip) eq 'ARRAY'){ | ||
| 1016 | foreach (@{$skip}){ | ||
| 1017 | if (&IsStringType($token,$_)){ | ||
| 1018 | return 1; | ||
| 1019 | } | ||
| 1020 | } | ||
| 1021 | } | ||
| 1022 | return 0; | ||
| 1023 | } | ||
| 1024 | |||
| 1025 | sub skipToken{ | ||
| 1026 | my $self=shift; | ||
| 1027 | my $token=shift; | ||
| 1028 | |||
| 1029 | my $skip=$self->getLanguageData('phrases','skip token hash'); | ||
| 1030 | if (ref($skip) eq 'HASH'){ | ||
| 1031 | if (exists $$skip{$token}){ | ||
| 1032 | return 1; | ||
| 1033 | } | ||
| 1034 | } | ||
| 1035 | my $skip=$self->getLanguageData('phrases','skip token string type'); | ||
| 1036 | if (ref($skip) eq 'ARRAY'){ | ||
| 1037 | foreach (@{$skip}){ | ||
| 1038 | if ($self->isStringType($token,$_)){ | ||
| 1039 | return 1; | ||
| 1040 | } | ||
| 1041 | } | ||
| 1042 | } | ||
| 1043 | return 0; | ||
| 1044 | } | ||
| 1045 | |||
| 1046 | sub skipPhraseAt{ | ||
| 1047 | my $self=shift; | ||
| 1048 | my $token=shift; | ||
| 1049 | |||
| 1050 | my $skip=$self->getLanguageData('phrases','skip phrase at hash'); | ||
| 1051 | if (ref($skip) eq 'HASH'){ | ||
| 1052 | if (exists $$skip{$token}){ | ||
| 1053 | return 1; | ||
| 1054 | } | ||
| 1055 | } | ||
| 1056 | my $skip=$self->getLanguageData('phrases','skip phrase at string type'); | ||
| 1057 | if (ref($skip) eq 'ARRAY'){ | ||
| 1058 | foreach (@{$skip}){ | ||
| 1059 | if ($self->isStringType($token,$_)){ | ||
| 1060 | # print STDERR "skip at $_ type\n"; | ||
| 1061 | return 1; | ||
| 1062 | } | ||
| 1063 | } | ||
| 1064 | } | ||
| 1065 | return 0; | ||
| 1066 | } | ||
| 1067 | |||
| 1068 | |||
| 1069 | sub skipPhraseBefore{ | ||
| 1070 | my $self=shift; | ||
| 1071 | my $token=shift; | ||
| 1072 | |||
| 1073 | my $skip=$self->getLanguageData('phrases','skip phrase before hash'); | ||
| 1074 | if (ref($skip) eq 'HASH'){ | ||
| 1075 | if (exists $$skip{$token}){ | ||
| 1076 | return 1; | ||
| 1077 | } | ||
| 1078 | } | ||
| 1079 | my $skip= | ||
| 1080 | $self->getLanguageData('phrases','skip phrase before string type'); | ||
| 1081 | if (ref($skip) eq 'ARRAY'){ | ||
| 1082 | foreach (@{$skip}){ | ||
| 1083 | if ($self->isStringType($token,$_)){ | ||
| 1084 | return 1; | ||
| 1085 | } | ||
| 1086 | } | ||
| 1087 | } | ||
| 1088 | return 0; | ||
| 1089 | } | ||
| 1090 | |||
| 1091 | sub skipPhraseAfter{ | ||
| 1092 | my $self=shift; | ||
| 1093 | my $token=shift; | ||
| 1094 | |||
| 1095 | my $skip=$self->getLanguageData('phrases','skip phrase after hash'); | ||
| 1096 | if (ref($skip) eq 'HASH'){ | ||
| 1097 | if (exists $$skip{$token}){ | ||
| 1098 | return 1; | ||
| 1099 | } | ||
| 1100 | } | ||
| 1101 | my $skip=$self->getLanguageData('phrases','skip phrase after string type'); | ||
| 1102 | if (ref($skip) eq 'ARRAY'){ | ||
| 1103 | foreach (@{$skip}){ | ||
| 1104 | if ($self->isStringType($token,$_)){ | ||
| 1105 | return 1; | ||
| 1106 | } | ||
| 1107 | } | ||
| 1108 | } | ||
| 1109 | return 0; | ||
| 1110 | } | ||
| 1111 | |||
| 1112 | |||
| 1113 | |||
| 1114 | |||
| 1115 | #----------------------------------------------------------------------------- | ||
| 1116 | # auxiliary functions .... | ||
| 1117 | |||
| 1118 | |||
| 1119 | sub ArrayToHash{ | ||
| 1120 | my ($Array,$Hash)=@_; | ||
| 1121 | if (ref($Array) eq 'ARRAY'){ | ||
| 1122 | foreach (@{$Array}){ | ||
| 1123 | $$Hash{$_}=1; | ||
| 1124 | } | ||
| 1125 | } | ||
| 1126 | } | ||
| 1127 | |||
| 1128 | |||
| 1129 | sub GetGeneralParam{ | ||
| 1130 | my $Param=shift; | ||
| 1131 | if (defined $$Param{general}){ | ||
| 1132 | if (ref($$Param{general}) eq 'HASH'){ | ||
| 1133 | return $$Param{general}; | ||
| 1134 | } | ||
| 1135 | } | ||
| 1136 | return $Param; | ||
| 1137 | } | ||
| 1138 | |||
| 1139 | sub GetNgramLengthParam{ | ||
| 1140 | my $Param=shift; | ||
| 1141 | my $length=1; | ||
| 1142 | |||
| 1143 | if (defined $$Param{'maximal ngram length'}){ | ||
| 1144 | $length=$$Param{'maximal ngram length'}; | ||
| 1145 | } | ||
| 1146 | foreach (keys %{$Param}){ | ||
| 1147 | if (ref($$Param{$_}) ne 'HASH'){next;} | ||
| 1148 | if ($$Param{$_}{'maximal ngram length'}>$length){ | ||
| 1149 | $length=$$Param{$_}{'maximal ngram length'}; | ||
| 1150 | } | ||
| 1151 | } | ||
| 1152 | return $length; | ||
| 1153 | } | ||
| 1154 | |||
| 1155 | 1; |