Grammalecte  Check-in [3e63ce4673]

// data
let _sAppContext = "";                                  // what software is running
let _dOptions = null;
let _aIgnoredRules = new Set();
let _oSpellChecker = null;
let _dAnalyses = new Map();                             // cache for data from dictionary


var gc_engine = {

    //// Informations

    lang: "${lang}",

                var spellchecker = require("resource://grammalecte/graphspell/spellchecker.js");
                _oSpellChecker = new spellchecker.SpellChecker("${lang}", "", "${dic_main_filename_js}", "${dic_extended_filename_js}", "${dic_community_filename_js}", "${dic_personal_filename_js}");
            } else {
                _oSpellChecker = new SpellChecker("${lang}", sPath, "${dic_main_filename_js}", "${dic_extended_filename_js}", "${dic_community_filename_js}", "${dic_personal_filename_js}");
            }
            _sAppContext = sContext;
            _dOptions = gc_options.getOptions(sContext).gl_shallowCopy();     // duplication necessary, to be able to reset to default

        }
        catch (e) {
            helpers.logerror(e);
        }
    },

    getSpellChecker: function () {

function displayInfo (dDA, aWord) {
    // for debugging: info of word
    if (!aWord) {
        helpers.echo("> nothing to find");
        return true;
    }
    if (!_dAnalyses.has(aWord[1]) && !_storeMorphFromFSA(aWord[1])) {

        helpers.echo("> not in FSA");
        return true;
    }
    if (dDA.has(aWord[0])) {
        helpers.echo("DA: " + dDA.get(aWord[0]));
    }
    helpers.echo("FSA: " + _dAnalyses.get(aWord[1]));
    return true;
}

function _storeMorphFromFSA (sWord) {
    // retrieves morphologies list from _oSpellChecker -> _dAnalyses
    //helpers.echo("register: "+sWord + " " + _oSpellChecker.getMorph(sWord).toString())
    _dAnalyses.set(sWord, _oSpellChecker.getMorph(sWord));
    return !!_dAnalyses.get(sWord);
}

function morph (dDA, aWord, sPattern, bStrict=true, bNoWord=false) {
    // analyse a tuple (position, word), return true if sPattern in morphologies (disambiguation on)
    if (!aWord) {
        //helpers.echo("morph: noword, returns " + bNoWord);
        return bNoWord;
    }
    //helpers.echo("aWord: "+aWord.toString());
    if (!_dAnalyses.has(aWord[1]) && !_storeMorphFromFSA(aWord[1])) {
        return false;
    }
    let lMorph = dDA.has(aWord[0]) ? dDA.get(aWord[0]) : _dAnalyses.get(aWord[1]);
    //helpers.echo("lMorph: "+lMorph.toString());
    if (lMorph.length === 0) {
        return false;
    }
    //helpers.echo("***");
    if (bStrict) {
        return lMorph.every(s  =>  (s.search(sPattern) !== -1));
    }
    return lMorph.some(s  =>  (s.search(sPattern) !== -1));
}

function morphex (dDA, aWord, sPattern, sNegPattern, bNoWord=false) {
    // analyse a tuple (position, word), returns true if not sNegPattern in word morphologies and sPattern in word morphologies (disambiguation on)
    if (!aWord) {
        //helpers.echo("morph: noword, returns " + bNoWord);
        return bNoWord;
    }
    //helpers.echo("aWord: "+aWord.toString());
    if (!_dAnalyses.has(aWord[1]) && !_storeMorphFromFSA(aWord[1])) {
        return false;
    }
    let lMorph = dDA.has(aWord[0]) ? dDA.get(aWord[0]) : _dAnalyses.get(aWord[1]);
    //helpers.echo("lMorph: "+lMorph.toString());
    if (lMorph.length === 0) {
        return false;
    }
    //helpers.echo("***");
    // check negative condition
    if (lMorph.some(s  =>  (s.search(sNegPattern) !== -1))) {
        return false;
    }
    // search sPattern
    return lMorph.some(s  =>  (s.search(sPattern) !== -1));
}

function analyse (sWord, sPattern, bStrict=true) {
    // analyse a word, return true if sPattern in morphologies (disambiguation off)
    if (!_dAnalyses.has(sWord) && !_storeMorphFromFSA(sWord)) {

        return false;
    }
    if (bStrict) {
        return _dAnalyses.get(sWord).every(s  =>  (s.search(sPattern) !== -1));
    }
    return _dAnalyses.get(sWord).some(s  =>  (s.search(sPattern) !== -1));
}

function analysex (sWord, sPattern, sNegPattern) {
    // analyse a word, returns True if not sNegPattern in word morphologies and sPattern in word morphologies (disambiguation off)
    if (!_dAnalyses.has(sWord) && !_storeMorphFromFSA(sWord)) {

        return false;
    }
    // check negative condition
    if (_dAnalyses.get(sWord).some(s  =>  (s.search(sNegPattern) !== -1))) {
        return false;
    }
    // search sPattern
    return _dAnalyses.get(sWord).some(s  =>  (s.search(sPattern) !== -1));
}

function stem (sWord) {
    // returns a list of sWord's stems
    if (!sWord) {
        return [];
    }
    if (!_dAnalyses.has(sWord) && !_storeMorphFromFSA(sWord)) {
        return [];
    }
    return _dAnalyses.get(sWord).map( s => s.slice(1, s.indexOf(" ")) );
}


//// functions to get text outside pattern scope

// warning: check compile_rules.py to understand how it works

function select (dDA, nPos, sWord, sPattern, lDefault=null) {
    if (!sWord) {
        return true;
    }
    if (dDA.has(nPos)) {
        return true;
    }
    if (!_dAnalyses.has(sWord) && !_storeMorphFromFSA(sWord)) {
        return true;
    }
    if (_dAnalyses.get(sWord).length === 1) {
        return true;
    }
    let lSelect = _dAnalyses.get(sWord).filter( sMorph => sMorph.search(sPattern) !== -1 );
    if (lSelect.length > 0) {
        if (lSelect.length != _dAnalyses.get(sWord).length) {
            dDA.set(nPos, lSelect);
        }
    } else if (lDefault) {
        dDA.set(nPos, lDefaul);
    }
    return true;
}

function exclude (dDA, nPos, sWord, sPattern, lDefault=null) {
    if (!sWord) {
        return true;
    }
    if (dDA.has(nPos)) {
        return true;
    }
    if (!_dAnalyses.has(sWord) && !_storeMorphFromFSA(sWord)) {
        return true;
    }
    if (_dAnalyses.get(sWord).length === 1) {
        return true;
    }
    let lSelect = _dAnalyses.get(sWord).filter( sMorph => sMorph.search(sPattern) === -1 );
    if (lSelect.length > 0) {
        if (lSelect.length != _dAnalyses.get(sWord).length) {
            dDA.set(nPos, lSelect);
        }
    } else if (lDefault) {
        dDA.set(nPos, lDefault);
    }
    return true;
}

    if (s2 == "vous") {
        return "vous";
    }
    if (s2 == "eux") {
        return "ils";
    }
    if (s2 == "elle" || s2 == "elles") {
        // We don’t check if word exists in _dAnalyses, for it is assumed it has been done before
        if (cregex.mbNprMasNotFem(_dAnalyses.gl_get(s1, ""))) {
            return "ils";
        }
        // si épicène, indéterminable, mais OSEF, le féminin l’emporte
        return "elles";
    }
    return s1 + " et " + s2;
}

function apposition (sWord1, sWord2) {
    // returns true if nom + nom (no agreement required)
    // We don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    return cregex.mbNomNotAdj(_dAnalyses.gl_get(sWord2, "")) && cregex.mbPpasNomNotAdj(_dAnalyses.gl_get(sWord1, ""));
}

function isAmbiguousNAV (sWord) {
    // words which are nom|adj and verb are ambiguous (except être and avoir)
    if (!_dAnalyses.has(sWord) && !_storeMorphFromFSA(sWord)) {

        return false;
    }
    if (!cregex.mbNomAdj(_dAnalyses.gl_get(sWord, "")) || sWord == "est") {
        return false;
    }
    if (cregex.mbVconj(_dAnalyses.gl_get(sWord, "")) && !cregex.mbMG(_dAnalyses.gl_get(sWord, ""))) {
        return true;
    }
    return false;
}

function isAmbiguousAndWrong (sWord1, sWord2, sReqMorphNA, sReqMorphConj) {
    //// use it if sWord1 won’t be a verb; word2 is assumed to be true via isAmbiguousNAV
    // We don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    let a2 = _dAnalyses.gl_get(sWord2, null);
    if (!a2 || a2.length === 0) {
        return false;
    }
    if (cregex.checkConjVerb(a2, sReqMorphConj)) {
        // verb word2 is ok
        return false;
    }
    let a1 = _dAnalyses.gl_get(sWord1, null);
    if (!a1 || a1.length === 0) {
        return false;
    }
    if (cregex.checkAgreement(a1, a2) && (cregex.mbAdj(a2) || cregex.mbAdj(a1))) {
        return false;
    }
    return true;
}

function isVeryAmbiguousAndWrong (sWord1, sWord2, sReqMorphNA, sReqMorphConj, bLastHopeCond) {
    //// use it if sWord1 can be also a verb; word2 is assumed to be true via isAmbiguousNAV
    // We don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    let a2 = _dAnalyses.gl_get(sWord2, null);
    if (!a2 || a2.length === 0) {
        return false;
    }
    if (cregex.checkConjVerb(a2, sReqMorphConj)) {
        // verb word2 is ok
        return false;
    }
    let a1 = _dAnalyses.gl_get(sWord1, null);
    if (!a1 || a1.length === 0) {
        return false;
    }
    if (cregex.checkAgreement(a1, a2) && (cregex.mbAdj(a2) || cregex.mbAdjNb(a1))) {
        return false;
    }
    // now, we know there no agreement, and conjugation is also wrong
    if (cregex.isNomAdj(a1)) {
        return true;
    }
    //if cregex.isNomAdjVerb(a1): # considered true
    if (bLastHopeCond) {
        return true;
    }
    return false;
}

function checkAgreement (sWord1, sWord2) {
    // We don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    let a2 = _dAnalyses.gl_get(sWord2, null);
    if (!a2 || a2.length === 0) {
        return true;
    }
    let a1 = _dAnalyses.gl_get(sWord1, null);
    if (!a1 || a1.length === 0) {
        return true;
    }
    return cregex.checkAgreement(a1, a2);
}

function mbUnit (s) {
    if (/[µ\/⁰¹²³⁴⁵⁶⁷⁸⁹Ωℓ·]/.test(s)) {

    var phonet = require("resource://grammalecte/fr/phonet.js");
}


//// verbs

function suggVerb (sFlex, sWho, funcSugg2=null) {
    // we don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    let aSugg = new Set();
    for (let sStem of stem(sFlex)) {
        let tTags = conj._getTags(sStem);
        if (tTags) {
            // we get the tense
            let aTense = new Set();
            for (let sMorph of _dAnalyses.gl_get(sFlex, [])) {
                let m;
                let zVerb = new RegExp (">"+sStem+" .*?(:(?:Y|I[pqsf]|S[pq]|K))", "g");
                while ((m = zVerb.exec(sMorph)) !== null) {
                    // stem must be used in regex to prevent confusion between different verbs (e.g. sauras has 2 stems: savoir and saurer)
                    if (m) {
                        if (m[1] === ":Y") {
                            aTense.add(":Ip");

        return Array.from(aSugg).join("|");
    }
    return "";
}

function suggVerbPpas (sFlex, sWhat=null) {
    let aSugg = new Set();
    for (let sStem of stem(sFlex)) {
        let tTags = conj._getTags(sStem);
        if (tTags) {
            if (!sWhat) {
                aSugg.add(conj._getConjWithTags(sStem, tTags, ":PQ", ":Q1"));
                aSugg.add(conj._getConjWithTags(sStem, tTags, ":PQ", ":Q2"));
                aSugg.add(conj._getConjWithTags(sStem, tTags, ":PQ", ":Q3"));
                aSugg.add(conj._getConjWithTags(sStem, tTags, ":PQ", ":Q4"));

        return Array.from(aSugg).join("|");
    }
    return "";
}

function suggVerbTense (sFlex, sTense, sWho) {
    let aSugg = new Set();
    for (let sStem of stem(sFlex)) {
        if (conj.hasConj(sStem, sTense, sWho)) {
            aSugg.add(conj.getConj(sStem, sTense, sWho));
        }
    }
    if (aSugg.size > 0) {
        return Array.from(aSugg).join("|");
    }
    return "";
}

function suggVerbImpe (sFlex) {
    let aSugg = new Set();
    for (let sStem of stem(sFlex)) {
        let tTags = conj._getTags(sStem);
        if (tTags) {
            if (conj._hasConjWithTags(tTags, ":E", ":2s")) {
                aSugg.add(conj._getConjWithTags(sStem, tTags, ":E", ":2s"));
            }
            if (conj._hasConjWithTags(tTags, ":E", ":1p")) {
                aSugg.add(conj._getConjWithTags(sStem, tTags, ":E", ":1p"));
            }
            if (conj._hasConjWithTags(tTags, ":E", ":2p")) {
                aSugg.add(conj._getConjWithTags(sStem, tTags, ":E", ":2p"));
            }
        }
    }
    if (aSugg.size > 0) {
        return Array.from(aSugg).join("|");
    }
    return "";
}

function suggVerbInfi (sFlex) {
    return stem(sFlex).filter(sStem => conj.isVerb(sStem)).join("|");
}


const _dQuiEst = new Map ([
    ["je", ":1s"], ["j’", ":1s"], ["j’en", ":1s"], ["j’y", ":1s"],
    ["tu", ":2s"], ["il", ":3s"], ["on", ":3s"], ["elle", ":3s"],
    ["nous", ":1p"], ["vous", ":2p"], ["ils", ":3p"], ["elles", ":3p"]

    if (!sWho) {
        if (sSuj[0].gl_isLowerCase()) { // pas un pronom, ni un nom propre
            return "";
        }
        sWho = ":3s";
    }
    let aSugg = new Set();
    for (let sStem of stem(sFlex)) {
        let tTags = conj._getTags(sStem);
        if (tTags) {
            for (let sTense of lMode) {
                if (conj._hasConjWithTags(tTags, sTense, sWho)) {
                    aSugg.add(conj._getConjWithTags(sStem, tTags, sTense, sWho));
                }
            }
        }
    }
    if (aSugg.size > 0) {
        return Array.from(aSugg).join("|");
    }
    return "";
}

//// Nouns and adjectives

function suggPlur (sFlex, sWordToAgree=null) {
    // returns plural forms assuming sFlex is singular
    if (sWordToAgree) {
        if (!_dAnalyses.has(sWordToAgree) && !_storeMorphFromFSA(sWordToAgree)) {

            return "";
        }
        let sGender = cregex.getGender(_dAnalyses.gl_get(sWordToAgree, []));
        if (sGender == ":m") {
            return suggMasPlur(sFlex);
        } else if (sGender == ":f") {
            return suggFemPlur(sFlex);
        }
    }
    let aSugg = new Set();

        return Array.from(aSugg).join("|");
    }
    return "";
}

function suggMasSing (sFlex, bSuggSimil=false) {
    // returns masculine singular forms
    // we don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    let aSugg = new Set();
    for (let sMorph of _dAnalyses.gl_get(sFlex, [])) {
        if (!sMorph.includes(":V")) {
            // not a verb
            if (sMorph.includes(":m") || sMorph.includes(":e")) {
                aSugg.add(suggSing(sFlex));
            } else {
                let sStem = cregex.getLemmaOfMorph(sMorph);
                if (mfsp.isFemForm(sStem)) {

        return Array.from(aSugg).join("|");
    }
    return "";
}

function suggMasPlur (sFlex, bSuggSimil=false) {
    // returns masculine plural forms
    // we don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    let aSugg = new Set();
    for (let sMorph of _dAnalyses.gl_get(sFlex, [])) {
        if (!sMorph.includes(":V")) {
            // not a verb
            if (sMorph.includes(":m") || sMorph.includes(":e")) {
                aSugg.add(suggPlur(sFlex));
            } else {
                let sStem = cregex.getLemmaOfMorph(sMorph);
                if (mfsp.isFemForm(sStem)) {

    }
    return "";
}


function suggFemSing (sFlex, bSuggSimil=false) {
    // returns feminine singular forms
    // we don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    let aSugg = new Set();
    for (let sMorph of _dAnalyses.gl_get(sFlex, [])) {
        if (!sMorph.includes(":V")) {
            // not a verb
            if (sMorph.includes(":f") || sMorph.includes(":e")) {
                aSugg.add(suggSing(sFlex));
            } else {
                let sStem = cregex.getLemmaOfMorph(sMorph);
                if (mfsp.isFemForm(sStem)) {

        return Array.from(aSugg).join("|");
    }
    return "";
}

function suggFemPlur (sFlex, bSuggSimil=false) {
    // returns feminine plural forms
    // we don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    let aSugg = new Set();
    for (let sMorph of _dAnalyses.gl_get(sFlex, [])) {
        if (!sMorph.includes(":V")) {
            // not a verb
            if (sMorph.includes(":f") || sMorph.includes(":e")) {
                aSugg.add(suggPlur(sFlex));
            } else {
                let sStem = cregex.getLemmaOfMorph(sMorph);
                if (mfsp.isFemForm(sStem)) {

    if (aSugg.size > 0) {
        return Array.from(aSugg).join("|");
    }
    return "";
}

function hasFemForm (sFlex) {
    for (let sStem of stem(sFlex)) {
        if (mfsp.isFemForm(sStem) || conj.hasConj(sStem, ":PQ", ":Q3")) {
            return true;
        }
    }
    if (phonet.hasSimil(sFlex, ":f")) {
        return true;
    }
    return false;
}

function hasMasForm (sFlex) {
    for (let sStem of stem(sFlex)) {
        if (mfsp.isFemForm(sStem) || conj.hasConj(sStem, ":PQ", ":Q1")) {
            // what has a feminine form also has a masculine form
            return true;
        }
    }
    if (phonet.hasSimil(sFlex, ":m")) {
        return true;
    }
    return false;
}

function switchGender (sFlex, bPlur=null) {
    // we don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    let aSugg = new Set();
    if (bPlur === null) {
        for (let sMorph of _dAnalyses.gl_get(sFlex, [])) {
            if (sMorph.includes(":f")) {
                if (sMorph.includes(":s")) {
                    aSugg.add(suggMasSing(sFlex));
                } else if (sMorph.includes(":p")) {
                    aSugg.add(suggMasPlur(sFlex));
                }
            } else if (sMorph.includes(":m")) {
                if (sMorph.includes(":s")) {
                    aSugg.add(suggFemSing(sFlex));
                } else if (sMorph.includes(":p")) {
                    aSugg.add(suggFemPlur(sFlex));
                } else {
                    aSugg.add(suggFemSing(sFlex));
                    aSugg.add(suggFemPlur(sFlex));
                }
            }
        }
    } else if (bPlur) {
        for (let sMorph of _dAnalyses.gl_get(sFlex, [])) {
            if (sMorph.includes(":f")) {
                aSugg.add(suggMasPlur(sFlex));
            } else if (sMorph.includes(":m")) {
                aSugg.add(suggFemPlur(sFlex));
            }
        }
    } else {
        for (let sMorph of _dAnalyses.gl_get(sFlex, [])) {
            if (sMorph.includes(":f")) {
                aSugg.add(suggMasSing(sFlex));
            } else if (sMorph.includes(":m")) {
                aSugg.add(suggFemSing(sFlex));
            }
        }
    }
    if (aSugg.size > 0) {
        return Array.from(aSugg).join("|");
    }
    return "";
}

function switchPlural (sFlex) {
    let aSugg = new Set();
    for (let sMorph of _dAnalyses.gl_get(sFlex, [])) { // we don’t check if word exists in _dAnalyses, for it is assumed it has been done before
        if (sMorph.includes(":s")) {
            aSugg.add(suggPlur(sFlex));
        } else if (sMorph.includes(":p")) {
            aSugg.add(suggSing(sFlex));
        }
    }
    if (aSugg.size > 0) {
        return Array.from(aSugg).join("|");
    }
    return "";
}

function hasSimil (sWord, sPattern=null) {
    return phonet.hasSimil(sWord, sPattern);
}

function suggSimil (sWord, sPattern=null, bSubst=false) {
    // return list of words phonetically similar to sWord and whom POS is matching sPattern
    let aSugg = phonet.selectSimil(sWord, sPattern);
    for (let sMorph of _dAnalyses.gl_get(sWord, [])) {
        for (let e of conj.getSimil(sWord, sMorph, bSubst)) {
            aSugg.add(e);
        }
    }
    if (aSugg.size > 0) {
        return Array.from(aSugg).join("|");
    }

    if (sWord[0] == "h" || sWord[0] == "H") {
        return "ce|cet";
    }
    return "ce";
}

function suggLesLa (sWord) {
    // we don’t check if word exists in _dAnalyses, for it is assumed it has been done before
    if (_dAnalyses.gl_get(sWord, []).some(s  =>  s.includes(":p"))) {
        return "les|la";
    }
    return "la";
}

function formatNumber (s) {
    let nLen = s.length;