247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
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
|
l.extend(self.morph(sWord.capitalize()))
return l
@timethis
def suggest (self, sWord, nMaxSugg=10):
"returns a set of suggestions for <sWord>"
sPfx, sWord, sSfx = cp.cut(sWord)
nMaxDel = len(sWord) // 5
nMaxHardRepl = max((len(sWord) - 5) // 4, 1)
oSuggResult = SuggResult(sWord)
self._suggest(oSuggResult, sWord, nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
if sWord.istitle():
self._suggest(oSuggResult, sWord.lower(), nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
elif sWord.islower():
self._suggest(oSuggResult, sWord.title(), nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
aSugg = oSuggResult.getSuggestions()
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, nMaxDel=0, nMaxHardRepl=0, nDeep=0, iAddr=0, sNewWord="", sAction="", bAvoidLoop=False):
# recursive function
#logging.info((nDeep * " ") + sNewWord + ":" + sRemain + " · " + sAction)
if not sRemain:
if int.from_bytes(self.byDic[iAddr:iAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
oSuggResult.addSugg(sNewWord)
for sTail in self._getTails(iAddr):
oSuggResult.addSugg(sNewWord+sTail)
return
cCurrent = sRemain[0:1]
for cChar, jAddr in self._getSimilarCharArcs(cCurrent, iAddr):
self._suggest(oSuggResult, sRemain[1:], nMaxDel, nMaxHardRepl, nDeep+1, jAddr, sNewWord+cChar, "*")
if not bAvoidLoop: # avoid infinite loop
if cCurrent == sRemain[1:2]:
# same char, we remove 1 char without adding 1 to <sNewWord>
self._suggest(oSuggResult, sRemain[1:], nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, cCurrent+"/2")
else:
# switching chars
self._suggest(oSuggResult, sRemain[1:2]+sRemain[0:1]+sRemain[2:], nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, "><",True)
# delete char
if nMaxDel > 0:
self._suggest(oSuggResult, sRemain[1:], nMaxDel-1, nMaxHardRepl, nDeep+1, iAddr, sNewWord, "-"+cCurrent, True)
# Phonetic replacements
for sRepl in cp.d1toX.get(cCurrent, ()):
self._suggest(oSuggResult, sRepl + sRemain[1:], nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, cCurrent+">"+sRepl, True)
for sRepl in cp.d2toX.get(sRemain[0:2], ()):
self._suggest(oSuggResult, sRepl + sRemain[2:], nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, sRemain[0:2]+">"+sRepl, True)
# Hard replacements
if nDeep > 3 and nMaxHardRepl and len(sRemain) >= 2:
for nVal, kAddr in self._getArcs1(iAddr):
if nVal in self.dCharVal:
cChar = self.dCharVal[nVal]
if cChar not in cp.d1to1.get(cCurrent, ""):
self._suggest(oSuggResult, sRemain[1:], nMaxDel, nMaxHardRepl-1, nDeep+1, kAddr, sNewWord+cChar, "[["+cChar+"]]", True)
# end of word
if len(sRemain) == 2:
for sRepl in cp.dFinal2.get(sRemain, ()):
self._suggest(oSuggResult, sRepl, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, sRemain + " >> " + sRepl, True)
elif len(sRemain) == 1:
self._suggest(oSuggResult, "", nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, sRemain + " [last char removed] ", True) # remove last char and go o
for sRepl in cp.dFinal1.get(sRemain, ()):
self._suggest(oSuggResult, sRepl, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, sRemain + " >> " + sRepl, True)
return
@timethis
def suggest2 (self, sWord, nMaxSugg=10):
"returns a set of suggestions for <sWord>"
sPfx, sWord, sSfx = cp.cut(sWord)
oSuggResult = SuggResult(sWord)
|
>
|
|
|
|
|
|
|
|
>
|
|
|
|
|
|
|
|
|
|
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
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
|
l.extend(self.morph(sWord.capitalize()))
return l
@timethis
def suggest (self, sWord, nMaxSugg=10):
"returns a set of suggestions for <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()
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="", sAction="", bAvoidLoop=False):
# recursive function
#logging.info((nDeep * " ") + sNewWord + ":" + sRemain + " · " + sAction)
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._getSimilarCharArcs(cCurrent, iAddr):
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, jAddr, sNewWord+cChar, "*")
if not bAvoidLoop: # avoid infinite loop
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, cCurrent+"/2")
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, "-"+cCurrent, True)
# Phonetic replacements
for sRepl in cp.d1toX.get(cCurrent, ()):
self._suggest(oSuggResult, sRepl + sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, cCurrent+">"+sRepl, True)
for sRepl in cp.d2toX.get(sRemain[0:2], ()):
self._suggest(oSuggResult, sRepl + sRemain[2:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, sRemain[0:2]+">"+sRepl, True)
# Hard replacements
if nDeep > 3 and nMaxHardRepl and len(sRemain) >= 2:
for nVal, kAddr in self._getArcs1(iAddr):
if nVal in self.dCharVal:
cChar = self.dCharVal[nVal]
if cChar not in cp.d1to1.get(cCurrent, ""):
self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl-1, nDeep+1, kAddr, sNewWord+cChar, "[["+cChar+"]]", 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, sRemain + " >> " + sRepl, True)
elif len(sRemain) == 1:
self._suggest(oSuggResult, "", nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, sRemain + " [last char removed] ", True) # remove last char and go o
for sRepl in cp.dFinal1.get(sRemain, ()):
self._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, sRemain + " >> " + sRepl, True)
return
@timethis
def suggest2 (self, sWord, nMaxSugg=10):
"returns a set of suggestions for <sWord>"
sPfx, sWord, sSfx = cp.cut(sWord)
oSuggResult = SuggResult(sWord)
|