Grammalecte  Check-in [6569849b49]

Overview
Comment:[graphspell][js] remove specific trick in cleanWord()
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | graphspell | bdic_opt
Files: files | file ages | folders
SHA3-256: 6569849b4931161ac91040f42809c42bc5ddad497e3c411b371fae37d7e22c84
User & Date: olr on 2020-09-15 14:01:46
Other Links: branch diff | manifest | tags
Context
2020-09-15
14:09
[graphspell][js] str_transform: fix cleanWord() check-in: 2bfe79e9aa user: olr tags: graphspell, bdic_opt
14:01
[graphspell][js] remove specific trick in cleanWord() check-in: 6569849b49 user: olr tags: graphspell, bdic_opt
13:50
[graphspell][js] suggest optimisation with Jaro-Winkler (thanks to IllusionPerdu) check-in: 3b3a02f4d3 user: olr tags: graphspell, bdic_opt
Changes

Modified graphspell-js/char_player.js from [8dac23cf9b] to [dab3d08473].

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
        ["UN", ["EIN",]],
    ]),

    // End of word
    dFinal1: new Map([
        ["a", ["as", "at", "ant", "ah"]],
        ["A", ["AS", "AT", "ANT", "AH"]],
        ["c", ["ch",]],
        ["C", ["CH",]],
        ["e", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait", "ent", "eh"]],
        ["E", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT", "ENT", "EH"]],
        ["é", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait"]],
        ["É", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT"]],
        ["è", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait"]],
        ["È", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT"]],
        ["ê", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait"]],
        ["Ê", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT"]],
        ["ë", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait"]],
        ["Ë", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT"]],
        ["g", ["gh",]],
        ["G", ["GH",]],
        ["i", ["is", "it", "ie", "in"]],
        ["I", ["IS", "IT", "IE", "IN"]],


        ["n", ["nt", "nd", "ns", "nh"]],
        ["N", ["NT", "ND", "NS", "NH"]],
        ["o", ["aut", "ot", "os"]],
        ["O", ["AUT", "OT", "OS"]],
        ["ô", ["aut", "ot", "os"]],
        ["Ô", ["AUT", "OT", "OS"]],
        ["ö", ["aut", "ot", "os"]],







|
|














>
>







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
        ["UN", ["EIN",]],
    ]),

    // End of word
    dFinal1: new Map([
        ["a", ["as", "at", "ant", "ah"]],
        ["A", ["AS", "AT", "ANT", "AH"]],
        ["c", ["ch", "que"]],
        ["C", ["CH", "QUE"]],
        ["e", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait", "ent", "eh"]],
        ["E", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT", "ENT", "EH"]],
        ["é", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait"]],
        ["É", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT"]],
        ["è", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait"]],
        ["È", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT"]],
        ["ê", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait"]],
        ["Ê", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT"]],
        ["ë", ["et", "er", "ets", "ée", "ez", "ai", "ais", "ait"]],
        ["Ë", ["ET", "ER", "ETS", "ÉE", "EZ", "AI", "AIS", "AIT"]],
        ["g", ["gh",]],
        ["G", ["GH",]],
        ["i", ["is", "it", "ie", "in"]],
        ["I", ["IS", "IT", "IE", "IN"]],
        ["k", ["que"]],
        ["K", ["QUE"]],
        ["n", ["nt", "nd", "ns", "nh"]],
        ["N", ["NT", "ND", "NS", "NH"]],
        ["o", ["aut", "ot", "os"]],
        ["O", ["AUT", "OT", "OS"]],
        ["ô", ["aut", "ot", "os"]],
        ["Ô", ["AUT", "OT", "OS"]],
        ["ö", ["aut", "ot", "os"]],

Modified graphspell-js/ibdawg.js from [2160aa77f7] to [fa91a09215].

353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
            [sPfx, sWord, sSfx] = this.lexicographer.split(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, nSuggLimit);
        let sWord = str_transform.cleanWord(sWord);
        if (bSplitTrailingNumbers) {
            this._splitTrailingNumbers(oSuggResult, sWord);
        }
        this._splitSuggest(oSuggResult, sWord);
        this._suggest(oSuggResult, sWord, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump);
        let aSugg = oSuggResult.getSuggestions();
        if (this.lexicographer) {







|







353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
            [sPfx, sWord, sSfx] = this.lexicographer.split(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, nSuggLimit);
        sWord = str_transform.cleanWord(sWord);
        if (bSplitTrailingNumbers) {
            this._splitTrailingNumbers(oSuggResult, sWord);
        }
        this._splitSuggest(oSuggResult, sWord);
        this._suggest(oSuggResult, sWord, nMaxSwitch, nMaxDel, nMaxHardRepl, nMaxJump);
        let aSugg = oSuggResult.getSuggestions();
        if (this.lexicographer) {

Modified graphspell-js/str_transform.js from [8ec0376c2c] to [f62ed68e13].

69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86

    cleanWord: function (sWord) {
        // word clean for the user who make commun and preditive error help suggest
        // remove letters repeated more than 2 times
        if (sWord.match(/(.)(\1){2,}/igm)){
            sWord = sWord.replace(/(.*)(.)(.\2)/igm,'$1$2').replace(/(.)(\1)+/igm,'$1$1');
        }
        // words ending with -ik -> replace with -ique
        if (sWord.match(/ik$/ig)){
            sWord = sWord.replace(/(.*)ik$/ig,'$1ique');
        }
        return sWord;
    },

    _xTransNumbersToExponent: new Map([
        ["0", "⁰"], ["1", "¹"], ["2", "²"], ["3", "³"], ["4", "⁴"], ["5", "⁵"], ["6", "⁶"], ["7", "⁷"], ["8", "⁸"], ["9", "⁹"]
    ]),








<
<
<
<







69
70
71
72
73
74
75




76
77
78
79
80
81
82

    cleanWord: function (sWord) {
        // word clean for the user who make commun and preditive error help suggest
        // remove letters repeated more than 2 times
        if (sWord.match(/(.)(\1){2,}/igm)){
            sWord = sWord.replace(/(.*)(.)(.\2)/igm,'$1$2').replace(/(.)(\1)+/igm,'$1$1');
        }




        return sWord;
    },

    _xTransNumbersToExponent: new Map([
        ["0", "⁰"], ["1", "¹"], ["2", "²"], ["3", "³"], ["4", "⁴"], ["5", "⁵"], ["6", "⁶"], ["7", "⁷"], ["8", "⁸"], ["9", "⁹"]
    ]),