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
|
if sWord[0:1].isupper():
l.extend(self.morph(sWord.lower()))
if sWord.isupper() and len(sWord) > 1:
l.extend(self.morph(sWord.capitalize()))
return l
#@timethis
def suggest (self, sWord, nSuggLimit=10):
"returns a set of suggestions for <sWord>"
sWord = sWord.rstrip(".") # useful for LibreOffice
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._splitSuggest(oSuggResult, sWord)
self._suggest(oSuggResult, sWord, 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 _splitSuggest (self, oSuggResult, sWord):
# split trailing numbers
m = re.match(r"(\D+)([0-9]+)$", sWord)
if m:
print("ok")
|
|
>
>
|
<
<
>
>
|
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
|
if sWord[0:1].isupper():
l.extend(self.morph(sWord.lower()))
if sWord.isupper() and len(sWord) > 1:
l.extend(self.morph(sWord.capitalize()))
return l
#@timethis
def suggest (self, sWord, nSuggLimit=10, bSplitTrailingNumbers=False):
"returns a set of suggestions for <sWord>"
sWord = sWord.rstrip(".") # useful for LibreOffice
sWord = cp.spellingNormalization(sWord)
sPfx, sWord, sSfx = cp.cut(sWord)
nMaxSwitch = max(len(sWord) // 3, 1)
nMaxDel = len(sWord) // 5
nMaxHardRepl = max((len(sWord) - 5) // 4, 1)
nMaxJump = max(len(sWord) // 4, 1)
oSuggResult = SuggResult(sWord)
if bSplitTrailingNumbers:
self._splitTrailingNumbers(oSuggResult, sWord)
self._splitSuggest(oSuggResult, sWord)
self._suggest(oSuggResult, sWord, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump)
aSugg = oSuggResult.getSuggestions(nSuggLimit)
if sSfx or sPfx:
# we add what we removed
return list(map(lambda sSug: sPfx + sSug + sSfx, aSugg))
return aSugg
def _splitTrailingNumbers (self, oSuggResult, sWord):
m = re.match(r"(\D+)([0-9]+)$", sWord)
if m:
oSuggResult.addSugg(m.group(1) + " " + cp.numbersToExponent(m.group(2)))
def _splitSuggest (self, oSuggResult, sWord):
# split at apostrophes
for cSplitter in "'’":
if cSplitter in sWord:
sWord1, sWord2 = sWord.split(cSplitter, 1)
if self.isValid(sWord1) and self.isValid(sWord2):
oSuggResult.addSugg(sWord1+" "+sWord2)
|