Grammalecte  Check-in [74cd9793a9]

Overview
Comment:[graphspell] suggestion mechanism: jumps
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk | graphspell
Files: files | file ages | folders
SHA3-256: 74cd9793a933b96bf8a5eaa03d7989e3ae8ccb30cc485aa624b0f42e02fb3b31
User & Date: olr on 2018-04-16 13:36:01
Other Links: manifest | tags
Context
2018-04-16
18:06
[fx] Write corrections in editable node also check-in: 6565051bbd user: olr tags: fx, trunk
13:36
[graphspell] suggestion mechanism: jumps check-in: 74cd9793a9 user: olr tags: graphspell, trunk
12:45
[graphspell] char_player update: ç -> c check-in: ccbcddb798 user: olr tags: graphspell, trunk
Changes

Modified graphspell-js/ibdawg.js from [d566558346] to [ee32a83c3f].

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
331
332
333
334
335
...
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
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
        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 oSuggResult = new SuggResult(sWord);
        this._suggest(oSuggResult, sWord, nMaxSwitch, nMaxDel, nMaxHardRepl);
        if (sWord.gl_isTitle()) {
            this._suggest(oSuggResult, sWord.toLowerCase(), nMaxSwitch, nMaxDel, nMaxHardRepl);
        }
        else if (sWord.gl_isLowerCase()) {
            this._suggest(oSuggResult, sWord.gl_toCapitalize(), nMaxSwitch, nMaxDel, nMaxHardRepl);
        }
        let aSugg = oSuggResult.getSuggestions(nSuggLimit);
        if (sSfx || sPfx) {
            // we add what we removed
            return aSugg.map( (sSugg) => { return sPfx + sSugg + sSfx } );
        }
        return aSugg;
    }

    _suggest (oSuggResult, sRemain, nMaxSwitch=0, nMaxDel=0, nMaxHardRepl=0, nDeep=0, iAddr=0, sNewWord="", bAvoidLoop=false) {
        // returns a set of suggestions
        // recursive function
        if (sRemain == "") {
            if (this._convBytesToInteger(this.byDic.slice(iAddr, iAddr+this.nBytesArc)) & this._finalNodeMask) {
                oSuggResult.addSugg(sNewWord);
            }
            for (let sTail of this._getTails(iAddr)) {
................................................................................
            }
            return;
        }

        let cCurrent = sRemain.slice(0, 1);
        for (let [cChar, jAddr] of this._getCharArcs(iAddr)) {
            if (char_player.d1to1.gl_get(cCurrent, cCurrent).indexOf(cChar) != -1) {
                this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, jAddr, sNewWord+cChar);
            }
            else if (!bAvoidLoop && nMaxHardRepl) {

                this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl-1, nDeep+1, jAddr, sNewWord+cChar, true);




            }
        }
        if (!bAvoidLoop) { // avoid infinite loop
            if (sRemain.length > 1) {
                if (cCurrent == sRemain.slice(1, 2)) {
                    // same char, we remove 1 char without adding 1 to <sNewWord>
                    this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord);
                }
                else {
                    // switching chars
                    if (nMaxSwitch > 0) {
                        this._suggest(oSuggResult, sRemain.slice(1, 2)+sRemain.slice(0, 1)+sRemain.slice(2), nMaxSwitch-1, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, true);
                    }
                    // delete char
                    if (nMaxDel > 0) {
                        this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel-1, nMaxHardRepl, nDeep+1, iAddr, sNewWord, true);
                    }
                }
                // Phonetic replacements
                for (let sRepl of char_player.get1toXReplacement(sNewWord.slice(-1), cCurrent, sRemain.slice(1,2))) {
                    this._suggest(oSuggResult, sRepl + sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, true);
                }
                for (let sRepl of char_player.d2toX.gl_get(sRemain.slice(0, 2), [])) {
                    this._suggest(oSuggResult, sRepl + sRemain.slice(2), nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, true);
                }
            }
            // end of word
            if (sRemain.length == 2) {
                for (let sRepl of char_player.dFinal2.gl_get(sRemain, [])) {
                    this._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, true);
                }
            }
            else if (sRemain.length == 1) {
                this._suggest(oSuggResult, "", nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, true); // remove last char and go on
                for (let sRepl of char_player.dFinal1.gl_get(sRemain, [])) {
                    this._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, true);
                }
            }
        }
    }

    * _getCharArcs (iAddr) {
        // generator: yield all chars and addresses from node at address <iAddr>







>

|

|


|









|







 







|

|
>
|
>
>
>
>






|




|



|




|


|





|



|

|







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
331
332
333
334
335
336
...
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
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
        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._suggest(oSuggResult, sWord, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump);
        if (sWord.gl_isTitle()) {
            this._suggest(oSuggResult, sWord.toLowerCase(), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump);
        }
        else if (sWord.gl_isLowerCase()) {
            this._suggest(oSuggResult, sWord.gl_toCapitalize(), 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 } );
        }
        return aSugg;
    }

    _suggest (oSuggResult, sRemain, nMaxSwitch=0, nMaxDel=0, nMaxHardRepl=0, nMaxJump=0, nDeep=0, iAddr=0, sNewWord="", bAvoidLoop=false) {
        // returns a set of suggestions
        // recursive function
        if (sRemain == "") {
            if (this._convBytesToInteger(this.byDic.slice(iAddr, iAddr+this.nBytesArc)) & this._finalNodeMask) {
                oSuggResult.addSugg(sNewWord);
            }
            for (let sTail of this._getTails(iAddr)) {
................................................................................
            }
            return;
        }

        let cCurrent = sRemain.slice(0, 1);
        for (let [cChar, jAddr] of this._getCharArcs(iAddr)) {
            if (char_player.d1to1.gl_get(cCurrent, cCurrent).indexOf(cChar) != -1) {
                this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, jAddr, sNewWord+cChar);
            }
            else if (!bAvoidLoop) {
                if (nMaxHardRepl) {
                    this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl-1, nMaxJump, nDeep+1, jAddr, sNewWord+cChar, true);
                }
                if (nMaxJump) {
                    this._suggest(oSuggResult, sRemain, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump-1, nDeep+1, jAddr, sNewWord+cChar, true);
                }
            }
        }
        if (!bAvoidLoop) { // avoid infinite loop
            if (sRemain.length > 1) {
                if (cCurrent == sRemain.slice(1, 2)) {
                    // same char, we remove 1 char without adding 1 to <sNewWord>
                    this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord);
                }
                else {
                    // switching chars
                    if (nMaxSwitch > 0) {
                        this._suggest(oSuggResult, sRemain.slice(1, 2)+sRemain.slice(0, 1)+sRemain.slice(2), nMaxSwitch-1, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, true);
                    }
                    // delete char
                    if (nMaxDel > 0) {
                        this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel-1, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, true);
                    }
                }
                // Phonetic replacements
                for (let sRepl of char_player.get1toXReplacement(sNewWord.slice(-1), cCurrent, sRemain.slice(1,2))) {
                    this._suggest(oSuggResult, sRepl + sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, true);
                }
                for (let sRepl of char_player.d2toX.gl_get(sRemain.slice(0, 2), [])) {
                    this._suggest(oSuggResult, sRepl + sRemain.slice(2), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, true);
                }
            }
            // end of word
            if (sRemain.length == 2) {
                for (let sRepl of char_player.dFinal2.gl_get(sRemain, [])) {
                    this._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, true);
                }
            }
            else if (sRemain.length == 1) {
                this._suggest(oSuggResult, "", nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, true); // remove last char and go on
                for (let sRepl of char_player.dFinal1.gl_get(sRemain, [])) {
                    this._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, true);
                }
            }
        }
    }

    * _getCharArcs (iAddr) {
        // generator: yield all chars and addresses from node at address <iAddr>

Modified graphspell/ibdawg.py from [e811d4895f] to [cffe6cd67c].

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
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
341
342
343
344
345
346
347
    def suggest (self, sWord, nSuggLimit=10):
        "returns a set of suggestions for <sWord>"
        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)

        oSuggResult = SuggResult(sWord)
        self._suggest(oSuggResult, sWord, nMaxSwitch=nMaxSwitch, nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
        if sWord.istitle():
            self._suggest(oSuggResult, sWord.lower(), nMaxSwitch=nMaxSwitch, nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
        elif sWord.islower():
            self._suggest(oSuggResult, sWord.title(), nMaxSwitch=nMaxSwitch, nMaxDel=nMaxDel, nMaxHardRepl=nMaxHardRepl)
        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 _suggest (self, oSuggResult, sRemain, nMaxSwitch=0, nMaxDel=0, nMaxHardRepl=0, nDeep=0, iAddr=0, sNewWord="", bAvoidLoop=False):
        # recursive function
        #logging.info((nDeep * "  ") + sNewWord + ":" + sRemain)
        if not sRemain:
            if int.from_bytes(self.byDic[iAddr:iAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
                oSuggResult.addSugg(sNewWord, nDeep)
            for sTail in self._getTails(iAddr):
                oSuggResult.addSugg(sNewWord+sTail, nDeep)
            return
        cCurrent = sRemain[0:1]
        for cChar, jAddr in self._getCharArcs(iAddr):
            if cChar in cp.d1to1.get(cCurrent, cCurrent):
                self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, jAddr, sNewWord+cChar)
            elif not bAvoidLoop and nMaxHardRepl:

                self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl-1, nDeep+1, jAddr, sNewWord+cChar, True)


        if not bAvoidLoop: # avoid infinite loop
            if len(sRemain) > 1:
                if cCurrent == sRemain[1:2]:
                    # same char, we remove 1 char without adding 1 to <sNewWord>
                    self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord)
                else:
                    # switching chars
                    if nMaxSwitch:
                        self._suggest(oSuggResult, sRemain[1:2]+sRemain[0:1]+sRemain[2:], nMaxSwitch-1, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
                    # delete char
                    if nMaxDel:
                        self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel-1, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
                # Phonetic replacements
                for sRepl in cp.get1toXReplacement(sNewWord[-1:], cCurrent, sRemain[1:2]):
                    self._suggest(oSuggResult, sRepl + sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
                for sRepl in cp.d2toX.get(sRemain[0:2], ()):
                    self._suggest(oSuggResult, sRepl + sRemain[2:], nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
            # end of word
            if len(sRemain) == 2:
                for sRepl in cp.dFinal2.get(sRemain, ()):
                    self._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)
            elif len(sRemain) == 1:
                self._suggest(oSuggResult, "", nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True) # remove last char and go on
                for sRepl in cp.dFinal1.get(sRemain, ()):
                    self._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nDeep+1, iAddr, sNewWord, True)

    #@timethis
    def suggest2 (self, sWord, nMaxSugg=10):
        "returns a set of suggestions for <sWord>"
        sWord = cp.spellingNormalization(sWord)
        sPfx, sWord, sSfx = cp.cut(sWord)
        oSuggResult = SuggResult(sWord)







>

|

|

|






|











|
|
>
|
>
>




|



|


|


|

|



|

|

|







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
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
341
342
343
344
345
346
347
348
349
350
351
    def suggest (self, sWord, nSuggLimit=10):
        "returns a set of suggestions for <sWord>"
        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._suggest(oSuggResult, sWord, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump)
        if sWord.istitle():
            self._suggest(oSuggResult, sWord.lower(), nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump)
        elif sWord.islower():
            self._suggest(oSuggResult, sWord.title(), 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 _suggest (self, oSuggResult, sRemain, nMaxSwitch=0, nMaxDel=0, nMaxHardRepl=0, nMaxJump=0, nDeep=0, iAddr=0, sNewWord="", bAvoidLoop=False):
        # recursive function
        #logging.info((nDeep * "  ") + sNewWord + ":" + sRemain)
        if not sRemain:
            if int.from_bytes(self.byDic[iAddr:iAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
                oSuggResult.addSugg(sNewWord, nDeep)
            for sTail in self._getTails(iAddr):
                oSuggResult.addSugg(sNewWord+sTail, nDeep)
            return
        cCurrent = sRemain[0:1]
        for cChar, jAddr in self._getCharArcs(iAddr):
            if cChar in cp.d1to1.get(cCurrent, cCurrent):
                self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, jAddr, sNewWord+cChar)
            elif not bAvoidLoop:
                if nMaxHardRepl:
                    self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl-1, nMaxJump, nDeep+1, jAddr, sNewWord+cChar, True)
                if nMaxJump:
                    self._suggest(oSuggResult, sRemain, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump-1, nDeep+1, jAddr, sNewWord+cChar, True)
        if not bAvoidLoop: # avoid infinite loop
            if len(sRemain) > 1:
                if cCurrent == sRemain[1:2]:
                    # same char, we remove 1 char without adding 1 to <sNewWord>
                    self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord)
                else:
                    # switching chars
                    if nMaxSwitch:
                        self._suggest(oSuggResult, sRemain[1:2]+sRemain[0:1]+sRemain[2:], nMaxSwitch-1, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
                    # delete char
                    if nMaxDel:
                        self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel-1, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
                # Phonetic replacements
                for sRepl in cp.get1toXReplacement(sNewWord[-1:], cCurrent, sRemain[1:2]):
                    self._suggest(oSuggResult, sRepl + sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
                for sRepl in cp.d2toX.get(sRemain[0:2], ()):
                    self._suggest(oSuggResult, sRepl + sRemain[2:], nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
            # end of word
            if len(sRemain) == 2:
                for sRepl in cp.dFinal2.get(sRemain, ()):
                    self._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)
            elif len(sRemain) == 1:
                self._suggest(oSuggResult, "", nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True) # remove last char and go on
                for sRepl in cp.dFinal1.get(sRemain, ()):
                    self._suggest(oSuggResult, sRepl, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump, nDeep+1, iAddr, sNewWord, True)

    #@timethis
    def suggest2 (self, sWord, nMaxSugg=10):
        "returns a set of suggestions for <sWord>"
        sWord = cp.spellingNormalization(sWord)
        sPfx, sWord, sSfx = cp.cut(sWord)
        oSuggResult = SuggResult(sWord)