Grammalecte  Check-in [e32c092585]

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

Overview
Comment:[graphspell] suggestions with trailing numbers: avoid to repeat splitting for each dictionary
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk | graphspell
Files: files | file ages | folders
SHA3-256:e32c092585e506508db1f8dbb0b25b4c96c62d503e394701c4ecaa25820115a7
User & Date: olr 2019-04-04 08:03:38
Context
2019-04-04
12:16
[fr] genfrdic: add flexion id to lexicon check-in: 0f24ce1e2c user: olr tags: fr, trunk
08:03
[graphspell] suggestions with trailing numbers: avoid to repeat splitting for each dictionary check-in: e32c092585 user: olr tags: graphspell, trunk
07:23
[graphspell][core][build][lo] remove extended dictionary check-in: 51a40c07e3 user: olr tags: build, core, graphspell, lo, trunk
Changes

Changes to graphspell-js/ibdawg.js.

317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335



336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352



353
354
355
356
357
358
359
            if (sWord.gl_isUpperCase() && sWord.length > 1) {
                l.push(...this.morph(sWord.gl_toCapitalize()));
            }
        }
        return l;
    }

    suggest (sWord, nSuggLimit=10) {
        // returns a array of suggestions for <sWord>
        //console.time("Suggestions for " + sWord);
        sWord = char_player.spellingNormalization(sWord);
        let sPfx = "";
        let sSfx = "";
        [sPfx, sWord, sSfx] = char_player.cut(sWord);
        let nMaxSwitch = Math.max(Math.floor(sWord.length / 3), 1);
        let nMaxDel = Math.floor(sWord.length / 5);
        let nMaxHardRepl = Math.max(Math.floor((sWord.length - 5) / 4), 1);
        let nMaxJump = Math.max(Math.floor(sWord.length / 4), 1);
        let oSuggResult = new SuggResult(sWord);



        this._splitSuggest(oSuggResult, sWord);
        this._suggest(oSuggResult, sWord, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump);
        let aSugg = oSuggResult.getSuggestions(nSuggLimit);
        if (sSfx || sPfx) {
            // we add what we removed
            return aSugg.map( (sSugg) => { return sPfx + sSugg + sSfx; } );
        }
        //console.timeEnd("Suggestions for " + sWord);
        return aSugg;
    }

    _splitSuggest (oSuggResult, sWord) {
        // split trailing numbers
        let m = /^([a-zA-Zà-öÀ-Ö_ø-ÿØ-ßĀ-ʯfi-st][a-zA-Zà-öÀ-Ö_ø-ÿØ-ßĀ-ʯfi-st-]+)([0-9]+)$/.exec(sWord);
        if (m) {
            oSuggResult.addSugg(m[1] + " " + char_player.numbersToExponent(m[2]));
        }



        // split at apostrophes
        for (let cSplitter of "'’") {
            if (sWord.includes(cSplitter)) {
                let [sWord1, sWord2] = sWord.split(cSplitter, 2);
                if (this.isValid(sWord1) && this.isValid(sWord2)) {
                    oSuggResult.addSugg(sWord1+" "+sWord2);
                }







|











>
>
>











|
<




>
>
>







317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350

351
352
353
354
355
356
357
358
359
360
361
362
363
364
            if (sWord.gl_isUpperCase() && sWord.length > 1) {
                l.push(...this.morph(sWord.gl_toCapitalize()));
            }
        }
        return l;
    }

    suggest (sWord, nSuggLimit=10, bSplitTrailingNumbers=false) {
        // returns a array of suggestions for <sWord>
        //console.time("Suggestions for " + sWord);
        sWord = char_player.spellingNormalization(sWord);
        let sPfx = "";
        let sSfx = "";
        [sPfx, sWord, sSfx] = char_player.cut(sWord);
        let nMaxSwitch = Math.max(Math.floor(sWord.length / 3), 1);
        let nMaxDel = Math.floor(sWord.length / 5);
        let nMaxHardRepl = Math.max(Math.floor((sWord.length - 5) / 4), 1);
        let nMaxJump = Math.max(Math.floor(sWord.length / 4), 1);
        let oSuggResult = new SuggResult(sWord);
        if (bSplitTrailingNumbers) {
            this._splitTrailingNumbers(oSuggResult, sWord);
        }
        this._splitSuggest(oSuggResult, sWord);
        this._suggest(oSuggResult, sWord, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump);
        let aSugg = oSuggResult.getSuggestions(nSuggLimit);
        if (sSfx || sPfx) {
            // we add what we removed
            return aSugg.map( (sSugg) => { return sPfx + sSugg + sSfx; } );
        }
        //console.timeEnd("Suggestions for " + sWord);
        return aSugg;
    }

    _splitTrailingNumbers (oSuggResult, sWord) {

        let m = /^([a-zA-Zà-öÀ-Ö_ø-ÿØ-ßĀ-ʯfi-st][a-zA-Zà-öÀ-Ö_ø-ÿØ-ßĀ-ʯfi-st-]+)([0-9]+)$/.exec(sWord);
        if (m) {
            oSuggResult.addSugg(m[1] + " " + char_player.numbersToExponent(m[2]));
        }
    }

    _splitSuggest (oSuggResult, sWord) {
        // split at apostrophes
        for (let cSplitter of "'’") {
            if (sWord.includes(cSplitter)) {
                let [sWord1, sWord2] = sWord.split(cSplitter, 2);
                if (this.isValid(sWord1) && this.isValid(sWord2)) {
                    oSuggResult.addSugg(sWord1+" "+sWord2);
                }

Changes to graphspell-js/spellchecker.js.

229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
            return this._dLemmas.get(sWord);
        }
        return Array.from(new Set(this.getMorph(sWord).map((sMorph) => { return sMorph.slice(1, sMorph.indexOf("/")); })));
    }

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

    * select (sFlexPattern="", sTagsPattern="") {
        // generator: returns all entries which flexion fits <sFlexPattern> and morphology fits <sTagsPattern>
        yield* this.oMainDic.select(sFlexPattern, sTagsPattern);
        if (this.bCommunityDic) {







|

|


|







229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
            return this._dLemmas.get(sWord);
        }
        return Array.from(new Set(this.getMorph(sWord).map((sMorph) => { return sMorph.slice(1, sMorph.indexOf("/")); })));
    }

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

    * select (sFlexPattern="", sTagsPattern="") {
        // generator: returns all entries which flexion fits <sFlexPattern> and morphology fits <sTagsPattern>
        yield* this.oMainDic.select(sFlexPattern, sTagsPattern);
        if (this.bCommunityDic) {

Changes to graphspell/ibdawg.py.

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")
            oSuggResult.addSugg(m.group(1) + " " + cp.numbersToExponent(m.group(2)))


        # 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)








|









>
>








|
<


<

>
>







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)

Changes to graphspell/spellchecker.py.

221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
        if self.dDefaultSugg:
            if sWord in self.dDefaultSugg:
                yield self.dDefaultSugg[sWord].split("|")
            elif sWord.istitle() and sWord.lower() in self.dDefaultSugg:
                lRes = self.dDefaultSugg[sWord.lower()].split("|")
                yield list(map(lambda sSugg: sSugg[0:1].upper()+sSugg[1:], lRes))
            else:
                yield self.oMainDic.suggest(sWord, nSuggLimit)
        else:
            yield self.oMainDic.suggest(sWord, nSuggLimit)
        if self.bCommunityDic:
            yield self.oCommunityDic.suggest(sWord, nSuggLimit)
        if self.bPersonalDic:
            yield self.oPersonalDic.suggest(sWord, nSuggLimit)

    def select (self, sFlexPattern="", sTagsPattern=""):
        "generator: returns all entries which flexion fits <sFlexPattern> and morphology fits <sTagsPattern>"
        yield from self.oMainDic.select(sFlexPattern, sTagsPattern)
        if self.bCommunityDic:
            yield from self.oCommunityDic.select(sFlexPattern, sTagsPattern)
        if self.bPersonalDic:







|

|

|

|







221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
        if self.dDefaultSugg:
            if sWord in self.dDefaultSugg:
                yield self.dDefaultSugg[sWord].split("|")
            elif sWord.istitle() and sWord.lower() in self.dDefaultSugg:
                lRes = self.dDefaultSugg[sWord.lower()].split("|")
                yield list(map(lambda sSugg: sSugg[0:1].upper()+sSugg[1:], lRes))
            else:
                yield self.oMainDic.suggest(sWord, nSuggLimit, True)
        else:
            yield self.oMainDic.suggest(sWord, nSuggLimit, True)
        if self.bCommunityDic:
            yield self.oCommunityDic.suggest(sWord, (nSuggLimit//2)+1)
        if self.bPersonalDic:
            yield self.oPersonalDic.suggest(sWord, (nSuggLimit//2)+1)

    def select (self, sFlexPattern="", sTagsPattern=""):
        "generator: returns all entries which flexion fits <sFlexPattern> and morphology fits <sTagsPattern>"
        yield from self.oMainDic.select(sFlexPattern, sTagsPattern)
        if self.bCommunityDic:
            yield from self.oCommunityDic.select(sFlexPattern, sTagsPattern)
        if self.bPersonalDic: