185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
|
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
|
-
-
+
-
-
-
-
-
+
-
+
-
+
|
l.extend(self.morph(sWord.lower()))
if sWord.isupper() and len(sWord) > 1:
l.extend(self.morph(sWord.capitalize()))
return l
def suggest (self, sWord, nMaxSugg=10):
"returns a set of suggestions for <sWord>"
sAdd = ""
if "-" in sWord:
sPfx, sWord, sSfx = cp.cut(sWord)
nLastHyphenPos = sWord.rfind("-")
if sWord[nLastHyphenPos+1:] in cp.aExcludedSfx:
sAdd = sWord[nLastHyphenPos:]
sWord = sWord[:nLastHyphenPos]
nMaxDel = len(sWord) // 5
nMaxHardRepl = max((len(sWord) - 5) // 4, 1)
aSugg = self._suggest(sWord, nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
if sWord.istitle():
aSugg.update(self._suggest(sWord.lower(), nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl))
aSugg = set(map(lambda sSugg: sSugg.title(), aSugg))
elif sWord.islower():
aSugg.update(self._suggest(sWord.title(), nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl))
if not aSugg:
#print("crush useless chars")
aSugg.update(self._suggestWithCrushedUselessChars(cp.clearWord(sWord)))
aSugg = filter(lambda sSugg: not sSugg.endswith(("è", "È")), aSugg) # fr language
aSugg = cp.filterSugg(aSugg)
aSugg = sorted(aSugg, key=lambda sSugg: cp.distanceDamerauLevenshtein(sWord, sSugg))[:nMaxSugg]
if sAdd:
if sSfx or sPfx:
# we add what we removed
return list(map(lambda sSug: sSug+sAdd, aSugg))
return list(map(lambda sSug: sPfx + sSug + sSfx, aSugg))
return aSugg
def _suggest (self, sRemain, nMaxDel=0, nMaxHardRepl=0, nDeep=0, iAddr=0, sNewWord="", bAvoidLoop=False):
"returns a set of suggestions"
# recursive function
#show(nDeep, sNewWord + ":" + sRemain)
aSugg = set()
|