282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
|
def suggest (self, sWord, nSuggLimit=10):
"returns a set of suggestions for <sWord>"
sWord = cp.spellingNormalization(sWord)
sPfx, sWord, sSfx = cp.cut(sWord)
nMaxSwitch = max(len(sWord) // 3, 1)
nMaxDel = len(sWord) // 5
nMaxHardRepl = max((len(sWord) - 5) // 4, 1)
oSuggResult = SuggResult(sWord)
self._suggest(oSuggResult, sWord, nMaxSwitch=nMaxSwitch, nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
if sWord.istitle():
self._suggest(oSuggResult, sWord.lower(), nMaxSwitch=nMaxSwitch, nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
elif sWord.islower():
self._suggest(oSuggResult, sWord.title(), nMaxSwitch=nMaxSwitch, nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
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 _suggest (self, oSuggResult, sRemain, nMaxSwitch=0, nMaxDel=0, nMaxHardRepl=0, nDeep=0, iAddr=0, sNewWord="", bAvoidLoop=False):
# recursive function
#logging.info((nDeep * " ") + sNewWord + ":" + sRemain)
if not sRemain:
if int.from_bytes(self.byDic[iAddr:iAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
oSuggResult.addSugg(sNewWord, nDeep)
for sTail in self._getTails(iAddr):
oSuggResult.addSugg(sNewWord+sTail, nDeep)
return
cCurrent = sRemain[0:1]
for cChar, jAddr in self._getCharArcs(iAddr):
if cChar in cp.d1to1.get(cCurrent, cCurrent):
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, jAddr, sNewWord+cChar)
elif not bAvoidLoop and nMaxHardRepl:
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl-1, nDeep+1, jAddr, sNewWord+cChar, True)
if not bAvoidLoop: # avoid infinite loop
if len(sRemain) > 1:
if cCurrent == sRemain[1:2]:
# same char, we remove 1 char without adding 1 to <sNewWord>
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord)
else:
# switching chars
if nMaxSwitch:
self._suggest(oSuggResult, sRemain[1:2]+sRemain[0:1]+sRemain[2:], nMaxSwitch-1, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
# delete char
if nMaxDel:
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel-1, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
# Phonetic replacements
for sRepl in cp.get1toXReplacement(sNewWord[-1:], cCurrent, sRemain[1:2]):
self._suggest(oSuggResult, sRepl + sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
for sRepl in cp.d2toX.get(sRemain[0:2], ()):
self._suggest(oSuggResult, sRepl + sRemain[2:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
# end of word
if len(sRemain) == 2:
for sRepl in cp.dFinal2.get(sRemain, ()):
self._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
elif len(sRemain) == 1:
self._suggest(oSuggResult, "", nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True) # remove last char and go on
for sRepl in cp.dFinal1.get(sRemain, ()):
self._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
#@timethis
def suggest2 (self, sWord, nMaxSugg=10):
"returns a set of suggestions for <sWord>"
sWord = cp.spellingNormalization(sWord)
sPfx, sWord, sSfx = cp.cut(sWord)
oSuggResult = SuggResult(sWord)
|
>
|
|
|
|
|
|
>
|
>
>
|
|
|
|
|
|
|
|
|
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
|
def suggest (self, sWord, nSuggLimit=10):
"returns a set of suggestions for <sWord>"
sWord = cp.spellingNormalization(sWord)
sPfx, sWord, sSfx = cp.cut(sWord)
nMaxSwitch = max(len(sWord) // 3, 1)
nMaxDel = len(sWord) // 5
nMaxHardRepl = max((len(sWord) - 5) // 4, 1)
nMaxJump = max(len(sWord) // 4, 1)
oSuggResult = SuggResult(sWord)
self._suggest(oSuggResult, sWord, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump)
if sWord.istitle():
self._suggest(oSuggResult, sWord.lower(), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump)
elif sWord.islower():
self._suggest(oSuggResult, sWord.title(), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump)
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 _suggest (self, oSuggResult, sRemain, nMaxSwitch=0, nMaxDel=0, nMaxHardRepl=0, nMaxJump=0, nDeep=0, iAddr=0, sNewWord="", bAvoidLoop=False):
# recursive function
#logging.info((nDeep * " ") + sNewWord + ":" + sRemain)
if not sRemain:
if int.from_bytes(self.byDic[iAddr:iAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
oSuggResult.addSugg(sNewWord, nDeep)
for sTail in self._getTails(iAddr):
oSuggResult.addSugg(sNewWord+sTail, nDeep)
return
cCurrent = sRemain[0:1]
for cChar, jAddr in self._getCharArcs(iAddr):
if cChar in cp.d1to1.get(cCurrent, cCurrent):
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, jAddr, sNewWord+cChar)
elif not bAvoidLoop:
if nMaxHardRepl:
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl-1, nMaxJump, nDeep+1, jAddr, sNewWord+cChar, True)
if nMaxJump:
self._suggest(oSuggResult, sRemain, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump-1, nDeep+1, jAddr, sNewWord+cChar, True)
if not bAvoidLoop: # avoid infinite loop
if len(sRemain) > 1:
if cCurrent == sRemain[1:2]:
# same char, we remove 1 char without adding 1 to <sNewWord>
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord)
else:
# switching chars
if nMaxSwitch:
self._suggest(oSuggResult, sRemain[1:2]+sRemain[0:1]+sRemain[2:], nMaxSwitch-1, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
# delete char
if nMaxDel:
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel-1, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
# Phonetic replacements
for sRepl in cp.get1toXReplacement(sNewWord[-1:], cCurrent, sRemain[1:2]):
self._suggest(oSuggResult, sRepl + sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
for sRepl in cp.d2toX.get(sRemain[0:2], ()):
self._suggest(oSuggResult, sRepl + sRemain[2:], nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
# end of word
if len(sRemain) == 2:
for sRepl in cp.dFinal2.get(sRemain, ()):
self._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
elif len(sRemain) == 1:
self._suggest(oSuggResult, "", nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True) # remove last char and go on
for sRepl in cp.dFinal1.get(sRemain, ()):
self._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
#@timethis
def suggest2 (self, sWord, nMaxSugg=10):
"returns a set of suggestions for <sWord>"
sWord = cp.spellingNormalization(sWord)
sPfx, sWord, sSfx = cp.cut(sWord)
oSuggResult = SuggResult(sWord)
|