|
1
|
1 |
|
import pickle |
|
2
|
1 |
|
import json |
|
3
|
1 |
|
import os |
|
4
|
|
|
|
|
5
|
1 |
|
class TextStream: |
|
6
|
|
|
""" |
|
7
|
|
|
A class to load and save files containing the nounification maps stored in txt format |
|
8
|
|
|
""" |
|
9
|
|
|
|
|
10
|
1 |
|
def __init__(self, directSymbol = ' -> ', inverseSymbol = ' <- ', separator = ', '): |
|
11
|
1 |
|
self.directSymbol = directSymbol |
|
12
|
1 |
|
self.inverseSymbol = inverseSymbol |
|
13
|
1 |
|
self.separator = separator |
|
14
|
|
|
|
|
15
|
1 |
|
def load(self, f): |
|
16
|
|
|
""" |
|
17
|
|
|
Load the content of f |
|
18
|
|
|
""" |
|
19
|
1 |
|
verbToNounsDirect = {} |
|
20
|
1 |
|
verbToNounsInverse = {} |
|
21
|
1 |
|
for line in f: |
|
22
|
1 |
|
line = line.replace('\n', '') |
|
23
|
1 |
|
if self.directSymbol in line: |
|
24
|
1 |
|
assert self.inverseSymbol not in line |
|
25
|
1 |
|
[key, nounList] = line.split(self.directSymbol, 1) |
|
26
|
1 |
|
target = verbToNounsDirect |
|
27
|
1 |
|
elif self.inverseSymbol in line: |
|
28
|
1 |
|
assert self.directSymbol not in line |
|
29
|
1 |
|
[key, nounList] = line.split(self.inverseSymbol, 1) |
|
30
|
1 |
|
target = verbToNounsInverse |
|
31
|
|
|
else: |
|
32
|
|
|
raise Exception("[nounDB] incorrect file format.") |
|
33
|
1 |
|
for noun in nounList.split(self.separator): |
|
34
|
1 |
|
if not key in target: |
|
35
|
1 |
|
target[key] = [noun] |
|
36
|
1 |
|
elif not noun in target[key]: |
|
37
|
1 |
|
target[key].append(noun) |
|
38
|
1 |
|
return [verbToNounsDirect, verbToNounsInverse] |
|
39
|
|
|
|
|
40
|
1 |
|
def dump(self, data, f): |
|
41
|
|
|
""" |
|
42
|
|
|
Save data into f |
|
43
|
|
|
""" |
|
44
|
1 |
|
[verbToNounsDirect, verbToNounsInverse] = data |
|
45
|
1 |
|
l = sorted([(x, 0) for x in verbToNounsDirect.keys()] + [(x, 1) for x in verbToNounsInverse.keys()]) |
|
46
|
1 |
|
for couple in l: |
|
47
|
1 |
|
if couple[1] == 0: |
|
48
|
1 |
|
f.write('%s%s%s\n' % (couple[0], self.directSymbol, self.separator.join(verbToNounsDirect[couple[0]]))) |
|
49
|
1 |
|
if couple[1] == 1: |
|
50
|
1 |
|
f.write('%s%s%s\n' % (couple[0], self.inverseSymbol, self.separator.join(verbToNounsInverse[couple[0]]))) |
|
51
|
|
|
|
|
52
|
1 |
|
class Nounificator: |
|
53
|
|
|
""" |
|
54
|
|
|
A class to handle the correspondances from the verbs to the nouns. |
|
55
|
|
|
""" |
|
56
|
1 |
|
pickleExtension = {'pickle', 'pkl', 'p'} |
|
57
|
1 |
|
jsonExtension = {'json'} |
|
58
|
1 |
|
txtExtension = {'txt'} |
|
59
|
1 |
|
def __init__(self): |
|
60
|
1 |
|
self.verbToNounsDirect = {} |
|
61
|
1 |
|
self.verbToNounsInverse = {} |
|
62
|
|
|
|
|
63
|
1 |
|
def select(self, x): |
|
64
|
1 |
|
if x[1] == 0: |
|
65
|
1 |
|
return ('%s:\t->%s' % (x[0], self.verbToNounsDirect[x[0]])) |
|
66
|
|
|
else: |
|
67
|
1 |
|
return ('%s:\t<-%s' % (x[0], self.verbToNounsInverse[x[0]])) |
|
68
|
|
|
|
|
69
|
1 |
|
def __str__(self): |
|
70
|
1 |
|
l = sorted([(x, 0) for x in self.verbToNounsDirect.keys()] + [(x, 1) for x in self.verbToNounsInverse.keys()]) |
|
71
|
1 |
|
return '\n'.join([self.select(x) for x in l]) |
|
72
|
|
|
|
|
73
|
1 |
|
def __eq__(self, other): |
|
74
|
1 |
|
return self.__dict__ == other.__dict__ |
|
75
|
|
|
|
|
76
|
1 |
|
def load(self, fileName): |
|
77
|
|
|
""" |
|
78
|
|
|
Load the database from the file of given name (pickle or json format). |
|
79
|
|
|
""" |
|
80
|
1 |
|
fileExtension = os.path.splitext(fileName)[1][1:] |
|
81
|
1 |
|
if fileExtension in self.pickleExtension: |
|
82
|
1 |
|
f = open(fileName, 'rb') |
|
83
|
1 |
|
module = pickle |
|
84
|
1 |
|
elif fileExtension in self.jsonExtension: |
|
85
|
1 |
|
f = open(fileName, 'r') |
|
86
|
1 |
|
module = json |
|
87
|
1 |
|
elif fileExtension in self.txtExtension: |
|
88
|
1 |
|
f = open(fileName, 'r') |
|
89
|
1 |
|
module = TextStream() |
|
90
|
1 |
|
[self.verbToNounsDirect, self.verbToNounsInverse] = module.load(f) |
|
91
|
1 |
|
f.close() |
|
92
|
|
|
|
|
93
|
1 |
|
def save(self, fileName): |
|
94
|
|
|
""" |
|
95
|
|
|
Save the database into the file of given name (pickle format). |
|
96
|
|
|
""" |
|
97
|
1 |
|
fileExtension = os.path.splitext(fileName)[1][1:] |
|
98
|
1 |
|
if fileExtension in self.pickleExtension: |
|
99
|
1 |
|
f = open(fileName, 'wb') |
|
100
|
1 |
|
pickle.dump([self.verbToNounsDirect, self.verbToNounsInverse], f) |
|
101
|
1 |
|
elif fileExtension in self.jsonExtension: |
|
102
|
1 |
|
f = open(fileName, 'w') |
|
103
|
1 |
|
json.dump([self.verbToNounsDirect, self.verbToNounsInverse], f, indent=4, sort_keys=True) |
|
104
|
1 |
|
elif fileExtension in self.txtExtension: |
|
105
|
1 |
|
f = open(fileName, 'w') |
|
106
|
1 |
|
TextStream().dump([self.verbToNounsDirect, self.verbToNounsInverse], f) |
|
107
|
1 |
|
f.close() |
|
108
|
|
|
|
|
109
|
1 |
|
def _add(self, verb, noun, target): |
|
110
|
1 |
|
try: |
|
111
|
1 |
|
if not noun in target[verb]: |
|
112
|
1 |
|
target[verb].append(noun) |
|
113
|
1 |
|
except: |
|
114
|
1 |
|
target[verb] = [noun] |
|
115
|
|
|
|
|
116
|
1 |
|
def addDirect(self, verb, noun): |
|
117
|
|
|
""" |
|
118
|
|
|
Add the given noun to the direct nounifications of the given verb. |
|
119
|
|
|
""" |
|
120
|
1 |
|
self._add(verb, noun, self.verbToNounsDirect) |
|
121
|
|
|
|
|
122
|
1 |
|
def addInverse(self, verb, noun): |
|
123
|
|
|
""" |
|
124
|
|
|
Add the given noun to the inverse nounifications of the given verb. |
|
125
|
|
|
""" |
|
126
|
1 |
|
self._add(verb, noun, self.verbToNounsInverse) |
|
127
|
|
|
|
|
128
|
1 |
|
def addListDirect(self, verb, nounList): |
|
129
|
|
|
""" |
|
130
|
|
|
Add the given list of nouns to the direct nounifications of the given verb. |
|
131
|
|
|
""" |
|
132
|
1 |
|
for noun in nounList: |
|
133
|
1 |
|
self.addDirect(verb, noun) |
|
134
|
|
|
|
|
135
|
1 |
|
def addListInverse(self, verb, nounList): |
|
136
|
|
|
""" |
|
137
|
|
|
Add the given list of nouns to the inverse nounifications of the given verb. |
|
138
|
|
|
""" |
|
139
|
1 |
|
for noun in nounList: |
|
140
|
1 |
|
self.addInverse(verb, noun) |
|
141
|
|
|
|
|
142
|
1 |
|
def _remove(self, verb, noun, target): |
|
143
|
1 |
|
target[verb].remove(noun) |
|
144
|
1 |
|
if target[verb] == []: |
|
145
|
1 |
|
target.pop(verb) |
|
146
|
|
|
|
|
147
|
1 |
|
def removeDirect(self, verb, noun): |
|
148
|
|
|
""" |
|
149
|
|
|
Remove the given noun from the direct nounifications of the given verb. |
|
150
|
|
|
""" |
|
151
|
1 |
|
self._remove(verb, noun, self.verbToNounsDirect) |
|
152
|
|
|
|
|
153
|
1 |
|
def removeInverse(self, verb, noun): |
|
154
|
|
|
""" |
|
155
|
|
|
Remove the given noun from the inverse nounifications of the given verb. |
|
156
|
|
|
""" |
|
157
|
1 |
|
self._remove(verb, noun, self.verbToNounsInverse) |
|
158
|
|
|
|
|
159
|
1 |
|
def _removeVerb(self, verb, target): |
|
160
|
1 |
|
target.pop(verb) |
|
161
|
|
|
|
|
162
|
1 |
|
def removeVerbDirect(self, verb): |
|
163
|
|
|
""" |
|
164
|
|
|
Remove all the direct nounifications of the given verb. |
|
165
|
|
|
""" |
|
166
|
1 |
|
self._removeVerb(verb, self.verbToNounsDirect) |
|
167
|
|
|
|
|
168
|
1 |
|
def removeVerbInverse(self, verb): |
|
169
|
|
|
""" |
|
170
|
|
|
Remove all the inverse nounifications of the given verb. |
|
171
|
|
|
""" |
|
172
|
1 |
|
self._removeVerb(verb, self.verbToNounsInverse) |
|
173
|
|
|
|
|
174
|
1 |
|
def _toNouns(self, verb, target): |
|
175
|
1 |
|
return target.get(verb, []) |
|
176
|
|
|
|
|
177
|
1 |
|
def directNouns(self, verb): |
|
178
|
|
|
""" |
|
179
|
|
|
Return the list of direct nounifications of the given noun. |
|
180
|
|
|
""" |
|
181
|
1 |
|
return self._toNouns(verb, self.verbToNounsDirect) |
|
182
|
|
|
|
|
183
|
1 |
|
def inverseNouns(self, verb): |
|
184
|
|
|
""" |
|
185
|
|
|
Return the list of inverse nounifications of the given noun. |
|
186
|
|
|
""" |
|
187
|
1 |
|
return self._toNouns(verb, self.verbToNounsInverse) |
|
188
|
|
|
|
|
189
|
1 |
|
def exists(self, verb): |
|
190
|
|
|
""" |
|
191
|
|
|
Return True if and only if there exists (direct or inverse) nounification(s) of the given verb. |
|
192
|
|
|
""" |
|
193
|
1 |
|
return verb in self.verbToNounsDirect or verb in self.verbToNounsInverse |
|
194
|
|
|
|
|
195
|
1 |
|
def merge(self, other): |
|
196
|
|
|
""" |
|
197
|
|
|
Merge with the given nounificator. |
|
198
|
|
|
""" |
|
199
|
1 |
|
for key, value in other.verbToNounsDirect.items(): |
|
200
|
1 |
|
self.addListDirect(key, value) |
|
201
|
1 |
|
for key, value in other.verbToNounsInverse.items(): |
|
202
|
|
|
self.addListInverse(key, value) |
|
203
|
|
|
|
ProjetPP /
PPP-QuestionParsing-Grammatical
