210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
|
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
|
-
+
+
|
"nArc": this.nArc,
"lArcVal": this.lArcVal,
"nArcVal": this.nArcVal,
"nCompressionMethod": this.nCompressionMethod,
"nBytesArc": this.nBytesArc,
"nBytesNodeAddress": this.nBytesNodeAddress,
"nBytesOffset": this.nBytesOffset,
"sByDic": this.sByDic // binary word graph
"sByDic": this.sByDic, // binary word graph
"l2grams": this.l2grams
};
return oJSON;
}
isValidToken (sToken) {
// checks if sToken is valid (if there is hyphens in sToken, sToken is split, each part is checked)
sToken = char_player.spellingNormalization(sToken)
|
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
|
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
|
-
+
-
+
-
+
|
let cCurrent = sRemain.slice(0, 1);
for (let [cChar, jAddr] of this._getCharArcs(iAddr)) {
if (char_player.d1to1.gl_get(cCurrent, cCurrent).indexOf(cChar) != -1) {
this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDist, nDeep+1, jAddr, sNewWord+cChar);
}
else if (!bAvoidLoop) {
if (nMaxHardRepl) {
if (nMaxHardRepl && this.isNgramsOK(cChar+sRemain.slice(1,2))) {
this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl-1, nMaxJump, nDist+1, nDeep+1, jAddr, sNewWord+cChar, true);
}
if (nMaxJump) {
this._suggest(oSuggResult, sRemain, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump-1, nDist+1, nDeep+1, jAddr, sNewWord+cChar, true);
}
}
}
if (!bAvoidLoop) { // avoid infinite loop
if (sRemain.length > 1) {
if (cCurrent == sRemain.slice(1, 2)) {
// same char, we remove 1 char without adding 1 to <sNewWord>
this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDist, nDeep+1, iAddr, sNewWord);
}
else {
// switching chars
if (nMaxSwitch > 0) {
if (nMaxSwitch > 0 && this.isNgramsOK(sNewWord.slice(-1)+sRemain.slice(1,2)) && this.isNgramsOK(sRemain.slice(1,2)+sRemain.slice(0,1))) {
this._suggest(oSuggResult, sRemain.slice(1, 2)+sRemain.slice(0, 1)+sRemain.slice(2), nMaxSwitch-1, nMaxDel, nMaxHardRepl, nMaxJump, nDist+1, nDeep+1, iAddr, sNewWord, true);
}
// delete char
if (nMaxDel > 0) {
if (nMaxDel > 0 && this.isNgramsOK(sNewWord.slice(-1)+sRemain.slice(1,2))) {
this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel-1, nMaxHardRepl, nMaxJump, nDist+1, nDeep+1, iAddr, sNewWord, true);
}
}
// Phonetic replacements
for (let sRepl of char_player.get1toXReplacement(sNewWord.slice(-1), cCurrent, sRemain.slice(1,2))) {
this._suggest(oSuggResult, sRepl + sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDist, nDeep+1, iAddr, sNewWord, true);
}
|
393
394
395
396
397
398
399
400
401
402
403
404
405
406
|
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
|
+
+
+
+
+
+
+
|
this._suggest(oSuggResult, "", nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDist, nDeep+1, iAddr, sNewWord, true); // remove last char and go on
for (let sRepl of char_player.dFinal1.gl_get(sRemain, [])) {
this._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDist, nDeep+1, iAddr, sNewWord, true);
}
}
}
}
isNgramsOK (sChars) {
if (sChars.length != 2) {
return true;
}
return this.a2grams.has(sChars);
}
* _getCharArcs (iAddr) {
// generator: yield all chars and addresses from node at address <iAddr>
for (let [nVal, jAddr] of this._getArcs(iAddr)) {
if (nVal <= this.nChar) {
yield [this.dCharVal.get(nVal), jAddr];
}
|