Grammalecte  Check-in [5762496ac7]

Overview
Comment:[graphspell] suggestions: test for improvement (+remove deprecated code)
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | trunk | graphspell
Files: files | file ages | folders
SHA3-256: 5762496ac7e460f05977b107e9f1cd64df5760750fead253e30de328d15a686a
User & Date: olr on 2020-05-16 16:24:35
Other Links: manifest | tags
Context
2020-05-16
16:39
[graphspell] comment about the test check-in: 825218be76 user: olr tags: trunk, graphspell
16:24
[graphspell] suggestions: test for improvement (+remove deprecated code) check-in: 5762496ac7 user: olr tags: trunk, graphspell
16:23
[fr] ajustements check-in: 9a5c04c9b4 user: olr tags: trunk, fr
Changes

Modified graphspell-js/ibdawg.js from [9b51ad972c] to [9e7f2fda83].

387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
                    oSuggResult.addSugg(sNewWord+" "+sRemain);
                }
            }
        }
        if (nDist > oSuggResult.nDistLimit) {
            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, nDist, nDeep+1, jAddr, sNewWord+cChar);
            }
            else if (!bAvoidLoop) {
                if (nMaxHardRepl  &&  this.isNgramsOK(cChar+sRemain.slice(1,2))) {
                    this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl-1, nMaxJump, nDist+1, nDeep+1, jAddr, sNewWord+cChar, true);
                }
                if (nMaxJump) {
                    this._suggest(oSuggResult, sRemain, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump-1, nDist+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>







<










|







387
388
389
390
391
392
393

394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
                    oSuggResult.addSugg(sNewWord+" "+sRemain);
                }
            }
        }
        if (nDist > oSuggResult.nDistLimit) {
            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, nDist, nDeep+1, jAddr, sNewWord+cChar);
            }
            else if (!bAvoidLoop) {
                if (nMaxHardRepl  &&  this.isNgramsOK(cChar+sRemain.slice(1,2))) {
                    this._suggest(oSuggResult, sRemain.slice(1), nMaxSwitch, nMaxDel, nMaxHardRepl-1, nMaxJump, nDist+1, nDeep+1, jAddr, sNewWord+cChar, true);
                }
                if (nMaxJump) {
                    this._suggest(oSuggResult, sRemain, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump-1, nDist+1, nDeep+1, jAddr, sNewWord+cChar); // True for avoiding loop?
                }
            }
        }
        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>

Modified graphspell/ibdawg.py from [3cc04ec719] to [3570c25f90].

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
                oSuggResult.addSugg(sNewWord, nDeep)
                for sTail in self._getTails(iAddr):
                    oSuggResult.addSugg(sNewWord+sTail, nDeep)
                return
            if (len(sNewWord) + len(sRemain) == len(oSuggResult.sWord)) and oSuggResult.sWord.lower().startswith(sNewWord.lower()) and self.isValid(sRemain):
                if self.sLangCode == "fr" and sNewWord.lower() in ("l", "d", "n", "m", "t", "s", "c", "j", "qu", "lorsqu", "puisqu", "quoiqu", "jusqu", "quelqu") and sRemain[0:1] in cp.aVowel:
                    oSuggResult.addSugg(sNewWord+"’"+sRemain)
                if (len(sNewWord) > 1 and len(sRemain) > 1) or sNewWord in ("a", "à", "y") or sRemain in ("a", "à", "y"):
                    oSuggResult.addSugg(sNewWord+" "+sRemain)
        if nDist > oSuggResult.nDistLimit:
            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, nDist, nDeep+1, jAddr, sNewWord+cChar)
            elif not bAvoidLoop:
                if nMaxHardRepl and self.isNgramsOK(cChar+sRemain[1:2]):
                    self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl-1, nMaxJump, nDist+1, nDeep+1, jAddr, sNewWord+cChar, True)
                if nMaxJump:
                    self._suggest(oSuggResult, sRemain, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump-1, nDist+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, nDist, nDeep+1, iAddr, sNewWord)
                else:
                    # switching chars







|











|







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
                oSuggResult.addSugg(sNewWord, nDeep)
                for sTail in self._getTails(iAddr):
                    oSuggResult.addSugg(sNewWord+sTail, nDeep)
                return
            if (len(sNewWord) + len(sRemain) == len(oSuggResult.sWord)) and oSuggResult.sWord.lower().startswith(sNewWord.lower()) and self.isValid(sRemain):
                if self.sLangCode == "fr" and sNewWord.lower() in ("l", "d", "n", "m", "t", "s", "c", "j", "qu", "lorsqu", "puisqu", "quoiqu", "jusqu", "quelqu") and sRemain[0:1] in cp.aVowel:
                    oSuggResult.addSugg(sNewWord+"’"+sRemain)
                if (len(sNewWord) > 1 and len(sRemain) > 1) or sNewWord in "aày" or sRemain in "aày":
                    oSuggResult.addSugg(sNewWord+" "+sRemain)
        if nDist > oSuggResult.nDistLimit:
            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, nDist, nDeep+1, jAddr, sNewWord+cChar)
            elif not bAvoidLoop:
                if nMaxHardRepl and self.isNgramsOK(cChar+sRemain[1:2]):
                    self._suggest(oSuggResult, sRemain[1:], nMaxSwitch, nMaxDel, nMaxHardRepl-1, nMaxJump, nDist+1, nDeep+1, jAddr, sNewWord+cChar, True)
                if nMaxJump:
                    self._suggest(oSuggResult, sRemain, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump-1, nDist+1, nDeep+1, jAddr, sNewWord+cChar)
        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, nDist, nDeep+1, iAddr, sNewWord)
                else:
                    # switching chars
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
        #logging.info((nDeep * "  ") + sNewWord)
        if nDeep >= oSuggResult.nDistLimit:
            sCleanNewWord = cp.simplifyWord(sNewWord)
            if st.distanceSift4(oSuggResult.sCleanWord[:len(sCleanNewWord)], sCleanNewWord) > oSuggResult.nDistLimit:
                return
        if int.from_bytes(self.byDic[iAddr:iAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
            oSuggResult.addSugg(sNewWord, nDeep)
        for cChar, jAddr in self._getCharArcsWithPriority(iAddr, oSuggResult.sWord[nDeep:nDeep+1]):
            self._suggest2(oSuggResult, nDeep+1, jAddr, sNewWord+cChar)
        return

    def _getCharArcs (self, iAddr):
        "generator: yield all chars and addresses from node at address <iAddr>"
        for nVal, jAddr in self._getArcs(iAddr):
            if nVal <= self.nChar:
                yield (self.dCharVal[nVal], jAddr)

    def _getSimilarCharArcs (self, cChar, iAddr):
        "generator: yield similar char of <cChar> and address of the following node"
        for c in cp.d1to1.get(cChar, [cChar]):
            if c in self.dChar:
                jAddr = self._lookupArcNode(self.dChar[c], iAddr)
                if jAddr:
                    yield (c, jAddr)

    def _getCharArcsWithPriority (self, iAddr, cChar):
        if not cChar:
            yield from self._getCharArcs(iAddr)
        lTuple = list(self._getCharArcs(iAddr))
        lTuple.sort(key=lambda t: 0  if t[0] in cp.d1to1.get(cChar, cChar)  else  1)
        yield from lTuple

    def _getTails (self, iAddr, sTail="", n=2):
        "return a list of suffixes ending at a distance of <n> from <iAddr>"
        aTails = set()
        for nVal, jAddr in self._getArcs(iAddr):
            if nVal <= self.nChar:
                if int.from_bytes(self.byDic[jAddr:jAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
                    aTails.add(sTail + self.dCharVal[nVal])







|









<
<
<
<
<
<
<
<
<
<
<
<
<
<
<







407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423















424
425
426
427
428
429
430
        #logging.info((nDeep * "  ") + sNewWord)
        if nDeep >= oSuggResult.nDistLimit:
            sCleanNewWord = cp.simplifyWord(sNewWord)
            if st.distanceSift4(oSuggResult.sCleanWord[:len(sCleanNewWord)], sCleanNewWord) > oSuggResult.nDistLimit:
                return
        if int.from_bytes(self.byDic[iAddr:iAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
            oSuggResult.addSugg(sNewWord, nDeep)
        for cChar, jAddr in self._getCharArcs(iAddr, oSuggResult.sWord[nDeep:nDeep+1]):
            self._suggest2(oSuggResult, nDeep+1, jAddr, sNewWord+cChar)
        return

    def _getCharArcs (self, iAddr):
        "generator: yield all chars and addresses from node at address <iAddr>"
        for nVal, jAddr in self._getArcs(iAddr):
            if nVal <= self.nChar:
                yield (self.dCharVal[nVal], jAddr)
















    def _getTails (self, iAddr, sTail="", n=2):
        "return a list of suffixes ending at a distance of <n> from <iAddr>"
        aTails = set()
        for nVal, jAddr in self._getArcs(iAddr):
            if nVal <= self.nChar:
                if int.from_bytes(self.byDic[jAddr:jAddr+self.nBytesArc], byteorder='big') & self._finalNodeMask:
                    aTails.add(sTail + self.dCharVal[nVal])