372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
|
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
|
-
-
-
+
+
+
-
-
+
+
-
-
-
+
+
+
-
-
+
+
-
+
-
+
-
+
-
-
+
+
+
+
+
+
+
+
+
+
+
+
-
-
-
-
-
-
-
-
-
-
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
|
}
}
update (sSentence, bDebug=false) {
// update <sSentence> and retokenize
this.sSentence = sSentence;
let lNewToken = Array.from(_oTokenizer.genTokens(sSentence, true));
for (let dToken of lNewToken) {
if (this.dTokenPos.gl_get(dToken["nStart"], {}).hasOwnProperty("lMorph")) {
dToken["lMorph"] = this.dTokenPos.get(dToken["nStart"])["lMorph"];
for (let oToken of lNewToken) {
if (this.dTokenPos.gl_get(oToken["nStart"], {}).hasOwnProperty("lMorph")) {
oToken["lMorph"] = this.dTokenPos.get(oToken["nStart"])["lMorph"];
}
if (this.dTokenPos.gl_get(dToken["nStart"], {}).hasOwnProperty("aTags")) {
dToken["aTags"] = this.dTokenPos.get(dToken["nStart"])["aTags"];
if (this.dTokenPos.gl_get(oToken["nStart"], {}).hasOwnProperty("aTags")) {
oToken["aTags"] = this.dTokenPos.get(oToken["nStart"])["aTags"];
}
}
this.lToken = lNewToken;
this.dTokenPos.clear();
for (let dToken of this.lToken) {
if (dToken["sType"] != "INFO") {
this.dTokenPos.set(dToken["nStart"], dToken);
for (let oToken of this.lToken) {
if (oToken["sType"] != "INFO") {
this.dTokenPos.set(oToken["nStart"], oToken);
}
}
if (bDebug) {
console.log("UPDATE:");
console.log(this.asString());
}
}
* _getNextPointers (dToken, oGraph, oPointer, bDebug=false) {
// generator: return nodes where <dToken> “values” match <oNode> arcs
* _getNextPointers (oToken, oGraph, oPointer, bDebug=false) {
// generator: return nodes where <oToken> “values” match <oNode> arcs
try {
let oNode = oGraph[oPointer["iNode"]];
let iToken1 = oPointer["iToken1"];
let bTokenFound = false;
// token value
if (oNode.hasOwnProperty(dToken["sValue"])) {
if (oNode.hasOwnProperty(oToken["sValue"])) {
if (bDebug) {
console.log(" MATCH: " + dToken["sValue"]);
console.log(" MATCH: " + oToken["sValue"]);
}
yield { "iToken1": iToken1, "iNode": oNode[dToken["sValue"]] };
yield { "iToken1": iToken1, "iNode": oNode[oToken["sValue"]] };
bTokenFound = true;
}
if (dToken["sValue"].slice(0,2).gl_isTitle()) { // we test only 2 first chars, to make valid words such as "Laissez-les", "Passe-partout".
let sValue = dToken["sValue"].toLowerCase();
if (oToken["sValue"].slice(0,2).gl_isTitle()) { // we test only 2 first chars, to make valid words such as "Laissez-les", "Passe-partout".
let sValue = oToken["sValue"].toLowerCase();
if (oNode.hasOwnProperty(sValue)) {
if (bDebug) {
console.log(" MATCH: " + sValue);
}
yield { "iToken1": iToken1, "iNode": oNode[sValue] };
bTokenFound = true;
}
}
else if (oToken["sValue"].gl_isUpperCase()) {
let sValue = oToken["sValue"].toLowerCase();
if (oNode.hasOwnProperty(sValue)) {
if (bDebug) {
console.log(" MATCH: " + sValue);
}
yield { "iToken1": iToken1, "iNode": oNode[sValue] };
bTokenFound = true;
}
}
else if (dToken["sValue"].gl_isUpperCase()) {
let sValue = dToken["sValue"].toLowerCase();
if (oNode.hasOwnProperty(sValue)) {
if (bDebug) {
console.log(" MATCH: " + sValue);
}
yield { "iToken1": iToken1, "iNode": oNode[sValue] };
bTokenFound = true;
}
sValue = dToken["sValue"].gl_toCapitalize();
if (dToken["sType"] != "INFO" && dToken["sType"] != "PUNC" && dToken["sType"] != "SIGN") {
if (oToken["sType"] != "INFO" && oToken["sType"] != "PUNC" && oToken["sType"] != "SIGN") {
if (oNode.hasOwnProperty("<re_value>")) {
for (let sRegex in oNode["<re_value>"]) {
if (!sRegex.includes("¬")) {
// no anti-pattern
if (dToken["sValue"].search(sRegex) !== -1) {
if (oToken["sValue"].search(sRegex) !== -1) {
if (bDebug) {
console.log(" MATCH: ~" + sRegex);
}
yield { "iToken1": iToken1, "iNode": oNode["<re_value>"][sRegex] };
bTokenFound = true;
}
} else {
// there is an anti-pattern
let [sPattern, sNegPattern] = sRegex.split("¬", 2);
if (sNegPattern && dToken["sValue"].search(sNegPattern) !== -1) {
if (sNegPattern && oToken["sValue"].search(sNegPattern) !== -1) {
continue;
}
if (!sPattern || dToken["sValue"].search(sPattern) !== -1) {
if (!sPattern || oToken["sValue"].search(sPattern) !== -1) {
if (bDebug) {
console.log(" MATCH: ~" + sRegex);
}
yield { "iToken1": iToken1, "iNode": oNode["<re_value>"][sRegex] };
bTokenFound = true;
}
}
}
}
}
// analysable tokens
if (dToken["sType"].slice(0,4) == "WORD") {
if (oToken["sType"].slice(0,4) == "WORD") {
// token lemmas
if (oNode.hasOwnProperty("<lemmas>")) {
for (let sLemma of _oSpellChecker.getLemma(dToken["sValue"])) {
for (let sLemma of _oSpellChecker.getLemma(oToken["sValue"])) {
if (oNode["<lemmas>"].hasOwnProperty(sLemma)) {
if (bDebug) {
console.log(" MATCH: >" + sLemma);
}
yield { "iToken1": iToken1, "iNode": oNode["<lemmas>"][sLemma] };
bTokenFound = true;
}
}
}
// regex morph arcs
if (oNode.hasOwnProperty("<re_morph>")) {
let lMorph = (dToken.hasOwnProperty("lMorph")) ? dToken["lMorph"] : _oSpellChecker.getMorph(dToken["sValue"]);
let lMorph = (oToken.hasOwnProperty("lMorph")) ? oToken["lMorph"] : _oSpellChecker.getMorph(oToken["sValue"]);
if (lMorph.length > 0) {
for (let sRegex in oNode["<re_morph>"]) {
if (!sRegex.includes("¬")) {
// no anti-pattern
if (lMorph.some(sMorph => (sMorph.search(sRegex) !== -1))) {
if (bDebug) {
console.log(" MATCH: @" + sRegex);
|
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
|
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
|
-
-
+
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
|
}
}
}
}
}
}
// token tags
if (dToken.hasOwnProperty("aTags") && oNode.hasOwnProperty("<tags>")) {
for (let sTag of dToken["aTags"]) {
if (oToken.hasOwnProperty("aTags") && oNode.hasOwnProperty("<tags>")) {
for (let sTag of oToken["aTags"]) {
if (oNode["<tags>"].hasOwnProperty(sTag)) {
if (bDebug) {
console.log(" MATCH: /" + sTag);
}
yield { "iToken1": iToken1, "iNode": oNode["<tags>"][sTag] };
bTokenFound = true;
}
}
}
// meta arc (for token type)
if (oNode.hasOwnProperty("<meta>")) {
for (let sMeta in oNode["<meta>"]) {
// no regex here, we just search if <oNode["sType"]> exists within <sMeta>
if (sMeta == "*" || dToken["sType"] == sMeta) {
if (sMeta == "*" || oToken["sType"] == sMeta) {
if (bDebug) {
console.log(" MATCH: *" + sMeta);
}
yield { "iToken1": iToken1, "iNode": oNode["<meta>"][sMeta] };
bTokenFound = true;
}
else if (sMeta.includes("¬")) {
if (!sMeta.includes(dToken["sType"])) {
if (!sMeta.includes(oToken["sType"])) {
if (bDebug) {
console.log(" MATCH: *" + sMeta);
}
yield { "iToken1": iToken1, "iNode": oNode["<meta>"][sMeta] };
bTokenFound = true;
}
}
}
}
if (!bTokenFound && oPointer.hasOwnProperty("bKeep")) {
yield oPointer;
}
// JUMP
// Warning! Recurssion!
if (oNode.hasOwnProperty("<>")) {
let oPointer2 = { "iToken1": iToken1, "iNode": oNode["<>"], "bKeep": true };
yield* this._getNextPointers(dToken, oGraph, oPointer2, bDebug);
yield* this._getNextPointers(oToken, oGraph, oPointer2, bDebug);
}
}
catch (e) {
console.error(e);
}
}
parseGraph (oGraph, sCountry="${country_default}", dOptions=null, bShowRuleId=false, bDebug=false, bContext=false) {
// parse graph with tokens from the text and execute actions encountered
let lPointer = [];
let bTagAndRewrite = false;
try {
for (let [iToken, dToken] of this.lToken.entries()) {
for (let [iToken, oToken] of this.lToken.entries()) {
if (bDebug) {
console.log("TOKEN: " + dToken["sValue"]);
console.log("TOKEN: " + oToken["sValue"]);
}
// check arcs for each existing pointer
let lNextPointer = [];
for (let oPointer of lPointer) {
lNextPointer.push(...this._getNextPointers(dToken, oGraph, oPointer, bDebug));
lNextPointer.push(...this._getNextPointers(oToken, oGraph, oPointer, bDebug));
}
lPointer = lNextPointer;
// check arcs of first nodes
lPointer.push(...this._getNextPointers(dToken, oGraph, { "iToken1": iToken, "iNode": 0 }, bDebug));
lPointer.push(...this._getNextPointers(oToken, oGraph, { "iToken1": iToken, "iNode": 0 }, bDebug));
// check if there is rules to check for each pointer
for (let oPointer of lPointer) {
if (oGraph[oPointer["iNode"]].hasOwnProperty("<rules>")) {
let bChange = this._executeActions(oGraph, oGraph[oPointer["iNode"]]["<rules>"], oPointer["iToken1"]-1, iToken, dOptions, sCountry, bShowRuleId, bDebug, bContext);
if (bChange) {
bTagAndRewrite = true;
}
|