23
24
25
26
27
28
29
30
31
32
33
34
35
36
|
fStart = time.time()
result = func(*args, **kwargs)
fEnd = time.time()
print(func.__name__, fEnd - fStart)
return result
return wrapper
class IBDAWG:
"""INDEXABLE BINARY DIRECT ACYCLIC WORD GRAPH"""
def __init__ (self, sDicName):
self.by = pkgutil.get_data(__package__, "_dictionaries/" + sDicName)
if not self.by:
|
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
|
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
|
fStart = time.time()
result = func(*args, **kwargs)
fEnd = time.time()
print(func.__name__, fEnd - fStart)
return result
return wrapper
class SuggResult:
def __init__ (self, sWord, nDistLimit=-1):
self.sWord = sWord
self.sCleanWord = cp.cleanWord(sWord)
self.nDistLimit = nDistLimit if nDistLimit >= 0 else (len(sWord) // 3) + 1
self.nMinDist = 1000
self.nMaxDist = 0
self.aSugg = set()
self.dSugg = { 0: [], 1: [], 2: [] }
def addSugg (self, sSugg, nDeep=0):
"add a suggestion"
if sSugg not in self.aSugg:
nDist = st.distanceSift4(self.sCleanWord, cp.cleanWord(sSugg))
if nDist <= self.nDistLimit:
if nDist not in self.dSugg:
self.dSugg[nDist] = []
self.dSugg[nDist].append(sSugg)
logging.info((nDeep * " ") + "__" + sSugg + "__")
if nDist > self.nMaxDist:
self.nMaxDist = nDist
if nDist < self.nMinDist:
self.nMinDist = nDist
self.nDistLimit = min(self.nDistLimit, self.nMinDist+2)
def getSuggestions (self, nSuggLimit=10, nDistLimit=-1):
"return a list of suggestions"
lRes = []
for lSugg in self.dSugg.values():
lRes.extend(lSugg)
if len(lRes) > nSuggLimit:
break
lRes = list(cp.filterSugg(lRes))
if self.sWord.istitle():
lRes = list(map(lambda sSugg: sSugg.title(), lRes))
return lRes[:nSuggLimit]
class IBDAWG:
"""INDEXABLE BINARY DIRECT ACYCLIC WORD GRAPH"""
def __init__ (self, sDicName):
self.by = pkgutil.get_data(__package__, "_dictionaries/" + sDicName)
if not self.by:
|
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
|
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))
aSugg = cp.filterSugg(aSugg)
sCleanWord = cp.cleanWord(sWord)
aSugg = sorted(aSugg, key=lambda sSugg: cp.distanceDamerauLevenshtein(sCleanWord, cp.cleanWord(sSugg)))[:nMaxSugg]
if sSfx or sPfx:
# we add what we removed
return list(map(lambda sSug: sPfx + sSug + sSfx, aSugg))
return aSugg
def _suggest (self, sRemain, nMaxDel=0, nMaxHardRepl=0, nDeep=0, iAddr=0, sNewWord="", sAction="", bAvoidLoop=False):
"returns a set of suggestions"
|
|
|
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
|
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))
aSugg = cp.filterSugg(aSugg)
sCleanWord = cp.cleanWord(sWord)
aSugg = sorted(aSugg, key=lambda sSugg: st.distanceDamerauLevenshtein(sCleanWord, cp.cleanWord(sSugg)))[:nMaxSugg]
if sSfx or sPfx:
# we add what we removed
return list(map(lambda sSug: sPfx + sSug + sSfx, aSugg))
return aSugg
def _suggest (self, sRemain, nMaxDel=0, nMaxHardRepl=0, nDeep=0, iAddr=0, sNewWord="", sAction="", bAvoidLoop=False):
"returns a set of suggestions"
|
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
|
aSugg.update(self._suggest(sRepl, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, sRemain + " >> " + sRepl, True))
return aSugg
@timethis
def suggest2 (self, sWord, nMaxSugg=10):
"returns a set of suggestions for <sWord>"
sPfx, sWord, sSfx = cp.cut(sWord)
nMaxDist = (len(sWord) // 3) + 1
sCleanWord = cp.cleanWord(sWord)
aSugg = self._suggest2(sWord, sCleanWord, nMaxDist)
if sWord.istitle():
aSugg.update(self._suggest2(sWord.lower(), sCleanWord, nMaxDist))
aSugg = set(map(lambda sSugg: sSugg.title(), aSugg))
elif sWord.islower():
aSugg.update(self._suggest2(sWord.title(), sCleanWord, nMaxDist))
aSugg = cp.filterSugg(aSugg)
aSugg = sorted(aSugg, key=lambda sSugg: cp.distanceSift4(sCleanWord, cp.cleanWord(sSugg)))[:nMaxSugg]
if sSfx or sPfx:
# we add what we removed
return list(map(lambda sSug: sPfx + sSug + sSfx, aSugg))
return aSugg
def _suggest2 (self, sWord, sCleanWord, nMaxDist, nDeep=0, iAddr=0, sNewWord=""):
#logging.info((nDeep * " ") + sNewWord)
if nDeep >= nMaxDist:
sCleanNewWord = cp.cleanWord(sNewWord)
if cp.distanceSift4(sCleanWord[:len(sCleanNewWord)], sCleanNewWord) > nMaxDist:
return set()
aSugg = set()
if int.from_bytes(self.byDic[iAddr:iAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
#logging.info((nDeep * " ") + "__" + sNewWord + "__")
sCleanNewWord = cp.cleanWord(sNewWord)
if cp.distanceSift4(sCleanWord, sCleanNewWord) <= nMaxDist:
aSugg.add(sNewWord)
aSugg.update(self._suggest2(sWord, sCleanWord, nMaxDist, nDeep+1, jAddr, sNewWord+cChar))
return aSugg
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)
|
|
<
|
<
<
<
<
<
|
<
|
|
|
|
<
<
<
|
|
|
|
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
|
aSugg.update(self._suggest(sRepl, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, sRemain + " >> " + sRepl, True))
return aSugg
@timethis
def suggest2 (self, sWord, nMaxSugg=10):
"returns a set of suggestions for <sWord>"
sPfx, sWord, sSfx = cp.cut(sWord)
oSuggResult = SuggResult(sWord)
self._suggest2(oSuggResult)
aSugg = oSuggResult.getSuggestions()
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=""):
#logging.info((nDeep * " ") + sNewWord)
if nDeep >= oSuggResult.nDistLimit:
sCleanNewWord = cp.cleanWord(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:
#logging.info((nDeep * " ") + "__" + sNewWord + "__")
oSuggResult.addSugg(sNewWord, nDeep)
for cChar, jAddr in self._getCharArcs(iAddr):
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)
|