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
|
"returns True if sChars in known 2grams"
if len(sChars) != 2:
return True
if not self.a2grams:
return True
return sChars in self.a2grams
#@timethis
def suggest2 (self, sWord, nSuggLimit=10):
"returns a set of suggestions for <sWord>"
sWord = cp.spellingNormalization(sWord)
sPfx, sWord, sSfx = cp.cut(sWord)
oSuggResult = SuggResult(sWord)
self._suggest2(oSuggResult)
aSugg = oSuggResult.getSuggestions(nSuggLimit)
if sSfx or sPfx:
# we add what we removed
return list(map(lambda sSug: sPfx + sSug + sSfx, aSugg))
return aSugg
def _suggest2 (self, oSuggResult, nDeep=0, iAddr=0, sNewWord=""):
# recursive function
#logging.info((nDeep * " ") + sNewWord)
if nDeep >= oSuggResult.nDistLimit:
sCleanNewWord = cp.simplifyWord(sNewWord)
if st.distanceSift4(oSuggResult.sCleanWord[:len(sCleanNewWord)], sCleanNewWord) > oSuggResult.nDistLimit:
return
if int.from_bytes(self.byDic[iAddr:iAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
oSuggResult.addSugg(sNewWord, nDeep)
for cChar, jAddr in self._getCharArcs(iAddr, oSuggResult.sWord[nDeep:nDeep+1]):
self._suggest2(oSuggResult, nDeep+1, jAddr, sNewWord+cChar)
return
def _getCharArcs (self, iAddr):
"generator: yield all chars and addresses from node at address <iAddr>"
for nVal, jAddr in self._getArcs(iAddr):
if nVal <= self.nChar:
yield (self.dCharVal[nVal], jAddr)
def _getTails (self, iAddr, sTail="", n=2):
|
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
<
|
385
386
387
388
389
390
391
392
393
394
395
396
397
398
|
"returns True if sChars in known 2grams"
if len(sChars) != 2:
return True
if not self.a2grams:
return True
return sChars in self.a2grams
def _getCharArcs (self, iAddr):
"generator: yield all chars and addresses from node at address <iAddr>"
for nVal, jAddr in self._getArcs(iAddr):
if nVal <= self.nChar:
yield (self.dCharVal[nVal], jAddr)
def _getTails (self, iAddr, sTail="", n=2):
|