Grammalecte  Check-in [09254c7f4b]

Overview
Comment:[graphspell] spellchecker: activate/deactivate dictionaries
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | graphspell | multid
Files: files | file ages | folders
SHA3-256: 09254c7f4b314489e70f195fadcd9ffab4497a7b0c159003bf53d0708747bc12
User & Date: olr on 2018-03-20 18:06:39
Other Links: branch diff | manifest | tags
Context
2018-03-20
19:55
[graphspell] code clarification check-in: 3cb01f1bc2 user: olr tags: graphspell, multid
18:06
[graphspell] spellchecker: activate/deactivate dictionaries check-in: 09254c7f4b user: olr tags: graphspell, multid
13:57
[graphspell][js] dawg: use <Set> to prevent duplicate entries check-in: 52593d2f38 user: olr tags: graphspell, multid
Changes

Modified graphspell-js/spellchecker.js from [c98943d8b7] to [758b22b0da].

35
36
37
38
39
40
41



42
43
44
45
46
47
48
..
85
86
87
88
89
90
91
92

93
94
95
96
97
98

99
100
101
102
103
104

105































106
107
108
109
110
111
112
...
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
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
221
        if (!mainDic) {
            mainDic = dDefaultDictionaries.gl_get(sLangCode, "");
        }
        this.oMainDic = this._loadDictionary(mainDic, sPath, true);
        this.oExtendedDic = this._loadDictionary(extentedDic, sPath);
        this.oCommunityDic = this._loadDictionary(communityDic, sPath);
        this.oPersonalDic = this._loadDictionary(personalDic, sPath);



        this.oTokenizer = null;
    }

    _loadDictionary (dictionary, sPath, bNecessary=false) {
        // returns an IBDAWG object
        if (!dictionary) {
            return null;
................................................................................
        this.oMainDic = this._loadDictionary(dictionary);
        return Boolean(this.oMainDic);
    }

    setExtendedDictionary (dictionary) {
        // returns true if the dictionary is loaded
        this.oExtendedDic = this._loadDictionary(dictionary);
        return Boolean(this.oExtendedDic);

    }

    setCommunityDictionary (dictionary) {
        // returns true if the dictionary is loaded
        this.oCommunityDic = this._loadDictionary(dictionary);
        return Boolean(this.oCommunityDic);

    }

    setPersonalDictionary (dictionary) {
        // returns true if the dictionary is loaded
        this.oPersonalDic = this._loadDictionary(dictionary);
        return Boolean(this.oPersonalDic);

    }
































    // parse text functions

    parseParagraph (sText) {
        if (!this.oTokenizer) {
            this.loadTokenizer();
        }
................................................................................
    // IBDAWG functions

    isValidToken (sToken) {
        // checks if sToken is valid (if there is hyphens in sToken, sToken is split, each part is checked)
        if (this.oMainDic.isValidToken(sToken)) {
            return true;
        }
        if (this.oExtendedDic && this.oExtendedDic.isValidToken(sToken)) {
            return true;
        }
        if (this.oCommunityDic && this.oCommunityDic.isValidToken(sToken)) {
            return true;
        }
        if (this.oPersonalDic && this.oPersonalDic.isValidToken(sToken)) {
            return true;
        }
        return false;
    }

    isValid (sWord) {
        // checks if sWord is valid (different casing tested if the first letter is a capital)
        if (this.oMainDic.isValid(sWord)) {
            return true;
        }
        if (this.oExtendedDic && this.oExtendedDic.isValid(sWord)) {
            return true;
        }
        if (this.oCommunityDic && this.oCommunityDic.isValid(sToken)) {
            return true;
        }
        if (this.oPersonalDic && this.oPersonalDic.isValid(sWord)) {
            return true;
        }
        return false;
    }

    lookup (sWord) {
        // checks if sWord is in dictionary as is (strict verification)
        if (this.oMainDic.lookup(sWord)) {
            return true;
        }
        if (this.oExtendedDic && this.oExtendedDic.lookup(sWord)) {
            return true;
        }
        if (this.oCommunityDic && this.oCommunityDic.lookup(sToken)) {
            return true;
        }
        if (this.oPersonalDic && this.oPersonalDic.lookup(sWord)) {
            return true;
        }
        return false;
    }

    getMorph (sWord) {
        // retrieves morphologies list, different casing allowed
        let lResult = this.oMainDic.getMorph(sWord);
        if (this.oExtendedDic) {
            lResult.push(...this.oExtendedDic.getMorph(sWord));
        }
        if (this.oCommunityDic) {
            lResult.push(...this.oCommunityDic.getMorph(sWord));
        }
        if (this.oPersonalDic) {
            lResult.push(...this.oPersonalDic.getMorph(sWord));
        }
        return lResult;
    }

    * suggest (sWord, nSuggLimit=10) {
        // generator: returns 1, 2 or 3 lists of suggestions
        yield this.oMainDic.suggest(sWord, nSuggLimit);
        if (this.oExtendedDic) {
            yield this.oExtendedDic.suggest(sWord, nSuggLimit);
        }
        if (this.oCommunityDic) {
            yield this.oCommunityDic.suggest(sWord, nSuggLimit);
        }
        if (this.oPersonalDic) {
            yield this.oPersonalDic.suggest(sWord, nSuggLimit);
        }
    }

    * select (sPattern="") {
        // generator: returns all entries which morphology fits <sPattern>
        yield* this.oMainDic.select(sPattern)
        if (this.oExtendedDic) {
            yield* this.oExtendedDic.select(sPattern);
        }
        if (this.oCommunityDic) {
            yield* this.oCommunityDic.select(sPattern);
        }
        if (this.oPersonalDic) {
            yield* this.oPersonalDic.select(sPattern);
        }
    }
}

if (typeof(exports) !== 'undefined') {
    exports.SpellChecker = SpellChecker;
}







>
>
>







 







|
>





|
>





|
>

>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>







 







|


|


|










|


|


|










|


|


|








|


|


|








|


|


|







|


|


|








35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
..
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
...
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
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
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
        if (!mainDic) {
            mainDic = dDefaultDictionaries.gl_get(sLangCode, "");
        }
        this.oMainDic = this._loadDictionary(mainDic, sPath, true);
        this.oExtendedDic = this._loadDictionary(extentedDic, sPath);
        this.oCommunityDic = this._loadDictionary(communityDic, sPath);
        this.oPersonalDic = this._loadDictionary(personalDic, sPath);
        this.bExtendedDic = Boolean(this.oExtendedDic);
        this.bCommunityDic = Boolean(this.oCommunityDic);
        this.bPersonalDic = Boolean(this.oPersonalDic);
        this.oTokenizer = null;
    }

    _loadDictionary (dictionary, sPath, bNecessary=false) {
        // returns an IBDAWG object
        if (!dictionary) {
            return null;
................................................................................
        this.oMainDic = this._loadDictionary(dictionary);
        return Boolean(this.oMainDic);
    }

    setExtendedDictionary (dictionary) {
        // returns true if the dictionary is loaded
        this.oExtendedDic = this._loadDictionary(dictionary);
        this.bExtendedDic = Boolean(this.oExtendedDic);
        return this.bExtendedDic;
    }

    setCommunityDictionary (dictionary) {
        // returns true if the dictionary is loaded
        this.oCommunityDic = this._loadDictionary(dictionary);
        this.bCommunityDic = Boolean(this.oCommunityDic);
        return this.bCommunityDic;
    }

    setPersonalDictionary (dictionary) {
        // returns true if the dictionary is loaded
        this.oPersonalDic = this._loadDictionary(dictionary);
        this.bPersonalDic = Boolean(this.oPersonalDic);
        return this.bPersonalDic;
    }

    activateExtendedDictionary () {
        if (this.oExtendedDic) {
            this.bExtendedDic = true;
        }
    }

    activateCommunityDictionary () {
        if (this.oCommunityDic) {
            this.bCommunityDic = true;
        }
    }

    activatePersonalDictionary () {
        if (this.oPersonalDic) {
            this.bPersonalDic = true;
        }
    }

    deactivateExtendedDictionary () {
        this.bExtendedDic = false;
    }

    deactivateCommunityDictionary () {
        this.bCommunityDic = false;
    }

    deactivatePersonalDictionary () {
        this.bPersonalDic = false;
    }


    // parse text functions

    parseParagraph (sText) {
        if (!this.oTokenizer) {
            this.loadTokenizer();
        }
................................................................................
    // IBDAWG functions

    isValidToken (sToken) {
        // checks if sToken is valid (if there is hyphens in sToken, sToken is split, each part is checked)
        if (this.oMainDic.isValidToken(sToken)) {
            return true;
        }
        if (this.bExtendedDic && this.oExtendedDic.isValidToken(sToken)) {
            return true;
        }
        if (this.bCommunityDic && this.oCommunityDic.isValidToken(sToken)) {
            return true;
        }
        if (this.bPersonalDic && this.oPersonalDic.isValidToken(sToken)) {
            return true;
        }
        return false;
    }

    isValid (sWord) {
        // checks if sWord is valid (different casing tested if the first letter is a capital)
        if (this.oMainDic.isValid(sWord)) {
            return true;
        }
        if (this.bExtendedDic && this.oExtendedDic.isValid(sWord)) {
            return true;
        }
        if (this.bCommunityDic && this.oCommunityDic.isValid(sToken)) {
            return true;
        }
        if (this.bPersonalDic && this.oPersonalDic.isValid(sWord)) {
            return true;
        }
        return false;
    }

    lookup (sWord) {
        // checks if sWord is in dictionary as is (strict verification)
        if (this.oMainDic.lookup(sWord)) {
            return true;
        }
        if (this.bExtendedDic && this.oExtendedDic.lookup(sWord)) {
            return true;
        }
        if (this.bCommunityDic && this.oCommunityDic.lookup(sToken)) {
            return true;
        }
        if (this.bPersonalDic && this.oPersonalDic.lookup(sWord)) {
            return true;
        }
        return false;
    }

    getMorph (sWord) {
        // retrieves morphologies list, different casing allowed
        let lResult = this.oMainDic.getMorph(sWord);
        if (this.bExtendedDic) {
            lResult.push(...this.oExtendedDic.getMorph(sWord));
        }
        if (this.bCommunityDic) {
            lResult.push(...this.oCommunityDic.getMorph(sWord));
        }
        if (this.bPersonalDic) {
            lResult.push(...this.oPersonalDic.getMorph(sWord));
        }
        return lResult;
    }

    * suggest (sWord, nSuggLimit=10) {
        // generator: returns 1, 2 or 3 lists of suggestions
        yield this.oMainDic.suggest(sWord, nSuggLimit);
        if (this.bExtendedDic) {
            yield this.oExtendedDic.suggest(sWord, nSuggLimit);
        }
        if (this.bCommunityDic) {
            yield this.oCommunityDic.suggest(sWord, nSuggLimit);
        }
        if (this.bPersonalDic) {
            yield this.oPersonalDic.suggest(sWord, nSuggLimit);
        }
    }

    * select (sPattern="") {
        // generator: returns all entries which morphology fits <sPattern>
        yield* this.oMainDic.select(sPattern)
        if (this.bExtendedDic) {
            yield* this.oExtendedDic.select(sPattern);
        }
        if (this.bCommunityDic) {
            yield* this.oCommunityDic.select(sPattern);
        }
        if (this.bPersonalDic) {
            yield* this.oPersonalDic.select(sPattern);
        }
    }
}

if (typeof(exports) !== 'undefined') {
    exports.SpellChecker = SpellChecker;
}

Modified graphspell/spellchecker.py from [b7bf43a731] to [374c5d134a].

28
29
30
31
32
33
34



35
36
37
38
39
40
41
..
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
...
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
        self.sLangCode = sLangCode
        if not sfMainDic:
            sfMainDic = dDefaultDictionaries.get(sLangCode, "")
        self.oMainDic = self._loadDictionary(sfMainDic, True)
        self.oExtendedDic = self._loadDictionary(sfExtendedDic)
        self.oCommunityDic = self._loadDictionary(sfCommunityDic)
        self.oPersonalDic = self._loadDictionary(sfPersonalDic)



        self.oTokenizer = None

    def _loadDictionary (self, source, bNecessary=False):
        "returns an IBDAWG object"
        if not source:
            return None
        try:
................................................................................
        "returns True if the dictionary is loaded"
        self.oMainDic = self._loadDictionary(source)
        return bool(self.oMainDic)
            
    def setExtendedDictionary (self, source):
        "returns True if the dictionary is loaded"
        self.oExtendedDic = self._loadDictionary(source)
        return bool(self.oExtendedDic)


    def setCommunityDictionary (self, source):
        "returns True if the dictionary is loaded"
        self.oCommunityDic = self._loadDictionary(source)
        return bool(self.oPersonalDic)


    def setPersonalDictionary (self, source):
        "returns True if the dictionary is loaded"
        self.oPersonalDic = self._loadDictionary(source)












        return bool(self.oPersonalDic)












    # parse text functions

    def parseParagraph (self, sText, bSpellSugg=False):
        if not self.oTokenizer:
            self.loadTokenizer()
        aSpellErrs = []
................................................................................

    # IBDAWG functions

    def isValidToken (self, sToken):
        "checks if sToken is valid (if there is hyphens in sToken, sToken is split, each part is checked)"
        if self.oMainDic.isValidToken(sToken):
            return True
        if self.oExtendedDic and self.oExtendedDic.isValidToken(sToken):
            return True
        if self.oCommunityDic and self.oCommunityDic.isValidToken(sToken):
            return True
        if self.oPersonalDic and self.oPersonalDic.isValidToken(sToken):
            return True
        return False

    def isValid (self, sWord):
        "checks if sWord is valid (different casing tested if the first letter is a capital)"
        if self.oMainDic.isValid(sWord):
            return True
        if self.oExtendedDic and self.oExtendedDic.isValid(sWord):
            return True
        if self.oCommunityDic and self.oCommunityDic.isValid(sToken):
            return True
        if self.oPersonalDic and self.oPersonalDic.isValid(sWord):
            return True
        return False

    def lookup (self, sWord):
        "checks if sWord is in dictionary as is (strict verification)"
        if self.oMainDic.lookup(sWord):
            return True
        if self.oExtendedDic and self.oExtendedDic.lookup(sWord):
            return True
        if self.oCommunityDic and self.oCommunityDic.lookup(sToken):
            return True
        if self.oPersonalDic and self.oPersonalDic.lookup(sWord):
            return True
        return False

    def getMorph (self, sWord):
        "retrieves morphologies list, different casing allowed"
        lResult = self.oMainDic.getMorph(sWord)
        if self.oExtendedDic:
            lResult.extend(self.oExtendedDic.getMorph(sWord))
        if self.oCommunityDic:
            lResult.extend(self.oCommunityDic.getMorph(sWord))
        if self.oPersonalDic:
            lResult.extend(self.oPersonalDic.getMorph(sWord))
        return lResult

    def getLemma (self, sWord):
        return set([ s[1:s.find(" ")]  for s in self.getMorph(sWord) ])

    def suggest (self, sWord, nSuggLimit=10):
        "generator: returns 1, 2 or 3 lists of suggestions"
        yield self.oMainDic.suggest(sWord, nSuggLimit)
        if self.oExtendedDic:
            yield self.oExtendedDic.suggest(sWord, nSuggLimit)
        if self.oCommunityDic:
            yield self.oCommunityDic.suggest(sWord, nSuggLimit)
        if self.oPersonalDic:
            yield self.oPersonalDic.suggest(sWord, nSuggLimit)

    def select (self, sPattern=""):
        "generator: returns all entries which morphology fits <sPattern>"
        yield from self.oMainDic.select(sPattern)
        if self.oExtendedDic:
            yield from self.oExtendedDic.select(sPattern)
        if self.oCommunityDic:
            yield from self.oCommunityDic.select(sPattern)
        if self.oPersonalDic:
            yield from self.oPersonalDic.select(sPattern)

    def drawPath (self, sWord):
        self.oMainDic.drawPath(sWord)
        if self.oExtendedDic:
            print("-----")
            self.oExtendedDic.drawPath(sWord)
        if self.oCommunityDic:
            print("-----")
            self.oCommunityDic.drawPath(sWord)
        if self.oPersonalDic:
            print("-----")
            self.oPersonalDic.drawPath(sWord)







>
>
>







 







|
>




|
>




>
>
>
>
>
>
>
>
>
>
>
>
|
>
>
>
>
>
>
>
>
>
>
>







 







|

|

|







|

|

|







|

|

|






|

|

|









|

|

|





|

|

|




|


|


|


28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
..
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
105
106
107
108
109
110
111
...
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
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
        self.sLangCode = sLangCode
        if not sfMainDic:
            sfMainDic = dDefaultDictionaries.get(sLangCode, "")
        self.oMainDic = self._loadDictionary(sfMainDic, True)
        self.oExtendedDic = self._loadDictionary(sfExtendedDic)
        self.oCommunityDic = self._loadDictionary(sfCommunityDic)
        self.oPersonalDic = self._loadDictionary(sfPersonalDic)
        self.bExtendedDic = bool(self.oExtendedDic)
        self.bCommunityDic = bool(self.oCommunityDic)
        self.bPersonalDic = bool(self.oPersonalDic)
        self.oTokenizer = None

    def _loadDictionary (self, source, bNecessary=False):
        "returns an IBDAWG object"
        if not source:
            return None
        try:
................................................................................
        "returns True if the dictionary is loaded"
        self.oMainDic = self._loadDictionary(source)
        return bool(self.oMainDic)
            
    def setExtendedDictionary (self, source):
        "returns True if the dictionary is loaded"
        self.oExtendedDic = self._loadDictionary(source)
        self.bExtendedDic = bool(self.oExtendedDic)
        return self.bExtendedDic

    def setCommunityDictionary (self, source):
        "returns True if the dictionary is loaded"
        self.oCommunityDic = self._loadDictionary(source)
        self.bCommunityDic = bool(self.oCommunityDic)
        return self.bCommunityDic

    def setPersonalDictionary (self, source):
        "returns True if the dictionary is loaded"
        self.oPersonalDic = self._loadDictionary(source)
        self.bPersonalDic = bool(self.oPersonalDic)
        return self.bPersonalDic

    def activateExtendedDictionary (self):
        if self.oExtendedDic:
            self.bExtendedDic = True

    def activateCommunityDictionary (self):
        if self.oCommunityDic:
            self.bCommunityDic = True

    def activatePersonalDictionary (self):
        if self.oPersonalDic:
            self.bPersonalDic = True

    def deactivateExtendedDictionary (self):
        self.bExtendedDic = False

    def deactivateCommunityDictionary (self):
        self.bCommunityDic = False

    def deactivatePersonalDictionary (self):
        self.bPersonalDic = False


    # parse text functions

    def parseParagraph (self, sText, bSpellSugg=False):
        if not self.oTokenizer:
            self.loadTokenizer()
        aSpellErrs = []
................................................................................

    # IBDAWG functions

    def isValidToken (self, sToken):
        "checks if sToken is valid (if there is hyphens in sToken, sToken is split, each part is checked)"
        if self.oMainDic.isValidToken(sToken):
            return True
        if self.bExtendedDic and self.oExtendedDic.isValidToken(sToken):
            return True
        if self.bCommunityDic and self.oCommunityDic.isValidToken(sToken):
            return True
        if self.bPersonalDic and self.oPersonalDic.isValidToken(sToken):
            return True
        return False

    def isValid (self, sWord):
        "checks if sWord is valid (different casing tested if the first letter is a capital)"
        if self.oMainDic.isValid(sWord):
            return True
        if self.bExtendedDic and self.oExtendedDic.isValid(sWord):
            return True
        if self.bCommunityDic and self.oCommunityDic.isValid(sToken):
            return True
        if self.bPersonalDic and self.oPersonalDic.isValid(sWord):
            return True
        return False

    def lookup (self, sWord):
        "checks if sWord is in dictionary as is (strict verification)"
        if self.oMainDic.lookup(sWord):
            return True
        if self.bExtendedDic and self.oExtendedDic.lookup(sWord):
            return True
        if self.bCommunityDic and self.oCommunityDic.lookup(sToken):
            return True
        if self.bPersonalDic and self.oPersonalDic.lookup(sWord):
            return True
        return False

    def getMorph (self, sWord):
        "retrieves morphologies list, different casing allowed"
        lResult = self.oMainDic.getMorph(sWord)
        if self.bExtendedDic:
            lResult.extend(self.oExtendedDic.getMorph(sWord))
        if self.bCommunityDic:
            lResult.extend(self.oCommunityDic.getMorph(sWord))
        if self.bPersonalDic:
            lResult.extend(self.oPersonalDic.getMorph(sWord))
        return lResult

    def getLemma (self, sWord):
        return set([ s[1:s.find(" ")]  for s in self.getMorph(sWord) ])

    def suggest (self, sWord, nSuggLimit=10):
        "generator: returns 1, 2 or 3 lists of suggestions"
        yield self.oMainDic.suggest(sWord, nSuggLimit)
        if self.bExtendedDic:
            yield self.oExtendedDic.suggest(sWord, nSuggLimit)
        if self.bCommunityDic:
            yield self.oCommunityDic.suggest(sWord, nSuggLimit)
        if self.bPersonalDic:
            yield self.oPersonalDic.suggest(sWord, nSuggLimit)

    def select (self, sPattern=""):
        "generator: returns all entries which morphology fits <sPattern>"
        yield from self.oMainDic.select(sPattern)
        if self.bExtendedDic:
            yield from self.oExtendedDic.select(sPattern)
        if self.bCommunityDic:
            yield from self.oCommunityDic.select(sPattern)
        if self.bPersonalDic:
            yield from self.oPersonalDic.select(sPattern)

    def drawPath (self, sWord):
        self.oMainDic.drawPath(sWord)
        if self.bExtendedDic:
            print("-----")
            self.oExtendedDic.drawPath(sWord)
        if self.bCommunityDic:
            print("-----")
            self.oCommunityDic.drawPath(sWord)
        if self.bPersonalDic:
            print("-----")
            self.oPersonalDic.drawPath(sWord)