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
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
|
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
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
|
-
+
-
-
+
+
-
-
-
+
-
-
-
-
+
+
+
+
-
-
-
-
-
+
+
+
-
-
-
-
+
+
+
-
-
-
-
-
+
+
+
+
-
-
-
-
-
+
-
+
|
class SuggResult:
"""Structure for storing, classifying and filtering suggestions"""
def __init__ (self, sWord, nSuggLimit=10, nDistLimit=-1):
self.sWord = sWord
self.sSimplifiedWord = st.simplifyWord(sWord)
self.nDistLimit = nDistLimit if nDistLimit >= 0 else (len(sWord) // 3) + 1
self.nDistLimit = nDistLimit if nDistLimit >= 0 else (len(sWord) // 3) + 1 # used in suggest()
self.nMinDist = 1000
# Temporary sets
self.aAllSugg = set() # All suggestions, even the one rejected
self.dGoodSugg = {} # Acceptable suggestions
self.dBestSugg = {} # Best suggestions
self.nSuggLimitExt = nSuggLimit + 2 # we add few entries in case suggestions merge after casing modifications
self.nBestSuggLimit = floor(nSuggLimit * 2) # n times the requested limit
self.nGoodSuggLimit = nSuggLimit * 15 # n times the requested limit
nDistJaro = 1 - st.distanceJaroWinkler(self.sSimplifiedWord, st.simplifyWord(sSugg))
nDist = floor(nDistJaro * 10)
if nDist < self.nMinDist:
self.nMinDist = nDist
nSimDist = st.distanceSift4(self.sSimplifiedWord, st.simplifyWord(sSugg))
st.showDistance(self.sSimplifiedWord, st.simplifyWord(sSugg))
if nSimDist < self.nMinDist:
self.nMinDist = nSimDist
#logging.info((nDeep * " ") + "__" + sSugg + "__ " + str(round(nDistJaro*1000)))
if nDistJaro < .11: # Best suggestions
self.dBestSugg[sSugg] = round(nDistJaro*1000)
if len(self.dBestSugg) > self.nBestSuggLimit:
self.nDistLimit = -1 # make suggest() to end search
elif nDistJaro < .33: # Good suggestions
self.dGoodSugg[sSugg] = round(nDistJaro*1000)
if len(self.dGoodSugg) > self.nGoodSuggLimit:
self.nDistLimit = -1 # make suggest() to end search
if len(self.dBestSugg) > 0:
# sort only with simplified words
lResTmp = sorted(self.dBestSugg.items(), key=lambda x: x[1])
for i in range(min(self.nSuggLimitExt, len(lResTmp))):
lRes.append(lResTmp[i][0])
if len(lRes) < self.nSuggLimitExt:
# sort with simplified words and original word
lResTmp = sorted(self.dGoodSugg.items(), key=lambda x: ((1-st.distanceJaroWinkler(self.sWord, x[0]))*10, x[1]))
for i in range(min(self.nSuggLimitExt, len(lResTmp))):
lRes.append(lResTmp[i][0])
# dont’ use <.istitle>
# don’t use <.istitle>
lRes = list(OrderedDict.fromkeys(map(lambda sSugg: sSugg[0:1].upper()+sSugg[1:], lRes))) # use dict, when Python 3.6+
return lRes[:self.nSuggLimit]
class IBDAWG:
"""INDEXABLE BINARY DIRECT ACYCLIC WORD GRAPH"""
|