Grammalecte  Check-in [19dcb29939]

import traceback

import graphspell.ibdawg as ibdawg
from graphspell.echo import echo
from graphspell.str_transform import defineSuffixCode
import graphspell.tokenizer as tkz


oDict = None


class cd:
    """Context manager for changing the current working directory"""
    def __init__ (self, newPath):
        self.newPath = os.path.expanduser(newPath)

    def __enter__ (self):

                if sLine == "__END__":
                    break
                if sLine and not sLine.startswith("#"):
                    yield sLine
    else:
        raise OSError("# Error. File not found or not loadable: " + spf)


def loadDictionary ():
    global oDict
    if not oDict:
        try:
            oDict = ibdawg.IBDAWG("fr-allvars.bdic")
        except:
            traceback.print_exc()


def makeDictionaries (sp, sVersion):
    with cd(sp+"/dictionnaire"):
        os.system("genfrdic.py -s -gl -v "+sVersion)


def makeConj (sp, bJS=False):
    print("> Conjugaisons ", end="")
    print("(Python et JavaScript)"  if bJS  else "(Python seulement)")
    dVerb = {}
    lVinfo = []; dVinfo = {}; nVinfo = 0
    lTags = []; dTags = {}; nTags = 0
    dVerbNames = {}

    dPatternList = {
        ":PQ": [], ":Ip": [], ":Iq": [], ":Is": [], ":If": [], ":K": [], ":Sp": [], ":Sq": [], ":E": []
    }
    dTrad = {
        "infi": ":Y", "ppre": ":PQ", "ppas": ":PQ",
        "ipre": ":Ip", "iimp": ":Iq", "ipsi": ":Is", "ifut": ":If",
        "spre": ":Sp", "simp": ":Sq",
        "cond": ":K", "impe": ":E",
        "1sg": ":1s", "2sg": ":2s", "3sg": ":3s", "1pl": ":1p", "2pl": ":2p", "3pl": ":3p", "1isg": ":1ś",
        "mas sg": ":Q1", "mas pl": ":Q2", "mas inv": ":Q1", "fem sg": ":Q3", "fem pl": ":Q4", "epi inv": ":Q1"
    }

    loadDictionary()

    # read lexicon
    nStop = 0
    for n, sLine in enumerate(readFile(sp+"/data/dictConj.txt")):
        nTab = sLine.count("\t")
        if nTab == 1:
            # new entry
            sLemma, sVinfo = sLine.split("\t")
            dConj = {   ":PQ": { ":P": "", ":Q1": "", ":Q2": "", ":Q3": "", ":Q4": ""},
                        ":Ip": { ":1s": "", ":2s": "", ":3s": "", ":1p": "", ":2p": "", ":3p": "", ":1ś": "" },
                        ":Iq": { ":1s": "", ":2s": "", ":3s": "", ":1p": "", ":2p": "", ":3p": "" },
                        ":Is": { ":1s": "", ":2s": "", ":3s": "", ":1p": "", ":2p": "", ":3p": "" },
                        ":If": { ":1s": "", ":2s": "", ":3s": "", ":1p": "", ":2p": "", ":3p": "" },
                        ":K":  { ":1s": "", ":2s": "", ":3s": "", ":1p": "", ":2p": "", ":3p": "" },
                        ":Sp": { ":1s": "", ":2s": "", ":3s": "", ":1p": "", ":2p": "", ":3p": "", ":1ś": "" },
                        ":Sq": { ":1s": "", ":2s": "", ":3s": "", ":1p": "", ":2p": "", ":3p": "", ":1ś": "" },
                        ":E":  { ":2s": "", ":1p": "", ":2p": "" }
                    }
            if sVinfo not in lVinfo:
                dVinfo[sVinfo] = nVinfo
                lVinfo.append(sVinfo)
                nVinfo += 1
            # looking for names derivating from verb
            for sMorph in oDict.getMorph(sLemma):
                if ":N" in sMorph:
                    dVerbNames[sLemma] = { sLemma }
                    break
        elif nTab == 2:
            # flexion
            _, sTag, sFlex = sLine.split("\t")
            if sTag.count(" ") == 0:
                if sTag == "ppre":
                    dConj[":PQ"][":P"] = defineSuffixCode(sLemma, sFlex)
            else:
                try:
                    mode, g = sTag.split(maxsplit=1)
                    mode = dTrad[mode]
                    g = dTrad[g]
                    if dConj[mode][g] == "":
                        dConj[mode][g] = defineSuffixCode(sLemma, sFlex)
                    else:
                        # comment gérer les autres graphies ?
                        pass
                except:
                    echo(sLemma, " - ", sTag, " - non géré: ", mode, " / ", g)
            # looking for names derivating from verb
            for sMorph in oDict.getMorph(sFlex):
                if ":N" in sMorph:
                    if sLemma not in dVerbNames:
                        dVerbNames[sLemma] = { sFlex }
                    else:
                        dVerbNames[sLemma].add(sFlex)
                    break
        elif sLine == "$":
            # we store the dictionary of rules for this lemma
            if dConj[":Ip"][":1ś"] == "2è":
                dConj[":Ip"][":1ś"] = "2é"
            elif sLemma == "pouvoir":
                dConj[":Ip"][":1ś"] = "6uis"
            lConjTags = []
            for sTense in [":PQ", ":Ip", ":Iq", ":Is", ":If", ":K", ":Sp", ":Sq", ":E"]:
                bFound = False
                for i, d in enumerate(dPatternList[sTense]):
                    if dConj[sTense] == d:
                        bFound = True
                        lConjTags.append(i)
                        break
                if not bFound:
                    lConjTags.append(len(dPatternList[sTense]))
                    dPatternList[sTense].append(dConj[sTense])
            tConjTags = tuple(lConjTags)
            if tConjTags not in lTags:
                dTags[tConjTags] = nTags
                lTags.append(tConjTags)
                nTags += 1
            dVerb[sLemma] = (dVinfo[sVinfo], dTags[tConjTags])
        else:
            print("# Error - unknown line", n)













    for sLemma, aNames in dVerbNames.items():









        dVerbNames[sLemma] = tuple(aNames)  # convert set to tuple

    ## write file for Python
    sCode = "## generated data (do not edit)\n\n" + \
            "# Informations about verbs\n" + \
            "lVtyp = " + str(lVinfo) + "\n\n" + \
            "# indexes of tenses in _dPatternConj\n" + \
            "lTags = " + str(lTags) + "\n\n" + \
            "# lists of affix codes to generate inflected forms\n" + \
            "dPatternConj = " + str(dPatternList) + "\n\n" + \
            "# dictionary of verbs : (index of Vtyp, index of Tags)\n" + \
            "dVerb = " + str(dVerb) + "\n\n" + \
            "# names as derivatives from verbs\n" + \
            "dVerbNames = " + str(dVerbNames) + "\n"
    open(sp+"/modules/conj_data.py", "w", encoding="utf-8", newline="\n").write(sCode)

    if bJS:
        ## write file for JavaScript
        with open(sp+"/modules-js/conj_data.json", "w", encoding="utf-8", newline="\n") as hDst:
            hDst.write("{\n")
            hDst.write('    "lVtyp": ' + json.dumps(lVinfo, ensure_ascii=False) + ",\n")
            hDst.write('    "lTags": ' + json.dumps(lTags, ensure_ascii=False) + ",\n")
            hDst.write('    "dPatternConj": ' + json.dumps(dPatternList, ensure_ascii=False) + ",\n")
            hDst.write('    "dVerb": ' + json.dumps(dVerb, ensure_ascii=False) + "\n")
            hDst.write('    "dVerbNames": ' + json.dumps(dVerbNames, ensure_ascii=False) + "\n")
            hDst.write("}\n")


def makeMfsp (sp, bJS=False):
    print("> Pluriel/singulier/masculin/féminin ", end="")
    print("(Python et JavaScript)"  if bJS  else "(Python seulement)")
    aPlurS = set()

                '    "dMasForm": ' +  json.dumps(dMasForm, ensure_ascii=False) + "\n}"
        open(sp+"/modules-js/mfsp_data.json", "w", encoding="utf-8", newline="\n").write(sCode)


def makePhonetTable (sp, bJS=False):
    print("> Correspondances phonétiques ", end="")
    print("(Python et JavaScript)"  if bJS  else "(Python seulement)")

    import gc_lang.fr.modules.conj as conj

    loadDictionary()





    # set of homophonic words
    lSet = []
    for sLine in readFile(sp+"/data/phonet_simil.txt"):
        lWord = sLine.split()
        aMore = set()
        for sWord in lWord:

}

var conj = {
    _lVtyp: [],
    _lTags: [],
    _dPatternConj: {},
    _dVerb: {},
    _dVerbNames: {},

    bInit: false,
    init: function (sJSONData) {
        try {
            let oData = JSON.parse(sJSONData);
            this._lVtyp = oData.lVtyp;
            this._lTags = oData.lTags;
            this._dPatternConj = oData.dPatternConj;
            this._dVerb = oData.dVerb;
            this._dVerbNames = oData.dVerbNames;
            this.bInit = true;
        }
        catch (e) {
            console.error(e);
        }
    },

                    aSugg.add("eut");
                } else {
                    aSugg.add("étais");
                    aSugg.add("était");
                }
                aSugg.delete("");
            } else {
                if (this._dVerbNames.hasOwnProperty(sInfi)) {
                    // there are names derivated from the verb
                    aSugg.update(this._dVerbNames[sInfi]);
                } else {
                    // we suggest past participles
                    aSugg.add(this._getConjWithTags(sInfi, tTags, ":PQ", ":Q1"));
                    aSugg.add(this._getConjWithTags(sInfi, tTags, ":PQ", ":Q2"));
                    aSugg.add(this._getConjWithTags(sInfi, tTags, ":PQ", ":Q3"));
                    aSugg.add(this._getConjWithTags(sInfi, tTags, ":PQ", ":Q4"));
                    aSugg.delete("");
                    // if there is only one past participle (epi inv), unreliable.
                    if (aSugg.size === 1) {
                        aSugg.clear();
                    }
                }
            }
        }
        return aSugg;
    },

    _getTags: function (sVerb) {

"""
Grammalecte - Conjugueur
"""

# License: GPL 3

import re
import traceback

from .conj_data import lVtyp as _lVtyp
from .conj_data import lTags as _lTags
from .conj_data import dPatternConj as _dPatternConj
from .conj_data import dVerb as _dVerb
from .conj_data import dVerbNames as _dVerbNames


_zStartVoy = re.compile("^[aeéiouœê]")
_zNeedTeuph = re.compile("[tdc]$")
#_zNEEDACCENTWITHJE = re.compile("[^i]e$")

_dProSuj = { ":1s": "je", ":1ś": "je", ":2s": "tu", ":3s": "il", ":1p": "nous", ":2p": "vous", ":3p": "ils" }

                aSugg.add("eus")
                aSugg.add("eut")
            else:
                aSugg.add("étais")
                aSugg.add("était")
            aSugg.discard("")
        else:
            if sInfi in _dVerbNames:
                # there are names derivated from the verb
                aSugg.update(_dVerbNames[sInfi])
            else:
                # we suggest past participles
                aSugg.add(_getConjWithTags(sInfi, tTags, ":PQ", ":Q1"))
                aSugg.add(_getConjWithTags(sInfi, tTags, ":PQ", ":Q2"))
                aSugg.add(_getConjWithTags(sInfi, tTags, ":PQ", ":Q3"))
                aSugg.add(_getConjWithTags(sInfi, tTags, ":PQ", ":Q4"))
                aSugg.discard("")
                # if there is only one past participle (epi inv), unreliable.
                if len(aSugg) == 1:
                    aSugg.clear()
    return aSugg


def getConjSimilInfiV1 (sInfi):
    "returns verbal forms phonetically similar to infinitive form (for verb in group 1)"
    if sInfi not in _dVerb:
        return set()