Grammalecte  Diff

#!python3

# FRENCH DATA BUILDER
#
# by Olivier R.
# License: MPL 2

import json
import os
import itertools
import traceback
import platform

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

            oDict = ibdawg.IBDAWG("fr-allvars.bdic")
        except:
            traceback.print_exc()


def makeDictionaries (sp, sVersion):
    with cd(sp+"/dictionnaire"):
        if platform.system() == "Windows":
            os.system("genfrdic.py -s -gl -v "+sVersion)
        else:
            os.system("python3 ./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

            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()      # pluriels en -s
    dTag = {}
    lTagFemForm = []
    lTagMiscPlur = []   # pluriels spéciaux
    dMiscPlur = {}
    dMasForm = {}
    lTag = []
    lTagFemPl = []
    for n, sLine in enumerate(readFile(sp+"/data/dictDecl.txt")):
        nTab = sLine.count("\t")
        if nTab == 1:
            # new entry
            lTag.clear()
            lTagFemPl.clear()
            sLemma, sFlags = sLine.split("\t")
            if sFlags.startswith("S"):
                cType = "sg"
            elif sFlags.startswith("X"):
                cType = "pl"
            elif sFlags.startswith("A"):
                cType = "pl"
            elif sFlags.startswith("I"):
                cType = "pl"
            elif sFlags.startswith("F"):
                cType = "mf"
            elif sFlags.startswith("W"):
                cType = "mf"
            else:
                cType = "?"
                print(" > inconnu : " + sFlags)
        elif nTab == 2:
            if cType == "sg":
                # nothing to do
                continue
            _, sFlexTags, sFlex = sLine.split("\t")
            if cType == "pl":
                if sFlexTags.endswith("pl"):
                    lTag.append(defineSuffixCode(sLemma, sFlex))
            elif cType == "mf":
                if sFlexTags.endswith("fem sg"):
                    lTag.append(defineSuffixCode(sLemma, sFlex))
                if sFlexTags.endswith("fem pl"):
                    lTagFemPl.append(defineSuffixCode(sLemma, sFlex))
            else:
                print("erreur: " + cType)
        elif sLine == "$":
            if cType == "sg":
                aPlurS.add(sLemma)
            elif cType == "pl":
                sTag = "|".join(lTag)
                if sTag not in dTag:
                    dTag[sTag] = len(lTagMiscPlur)
                    lTagMiscPlur.append(sTag)
                dMiscPlur[sLemma] = dTag[sTag]
            elif cType == "mf":
                sTag = "|".join(lTag) + "/" + "|".join(lTagFemPl)


                if sTag not in dTag:
                    dTag[sTag] = len(lTagFemForm)
                    lTagFemForm.append(sTag)
                dMasForm[sLemma] = dTag[sTag]
            else:
                print("unknown tag: " + ctype)
        else:
            print("# Error - unknown line #", n)

    ## write file for Python
    sCode = "# generated data (do not edit)\n\n" + \
            "# list of affix codes\n" + \
            "lTagMiscPlur = " + str(lTagMiscPlur) + "\n" + \
            "lTagFemForm = " + str(lTagFemForm) + "\n\n" + \
            "# dictionary of words with uncommon plurals (-x, -ux, english, latin and italian plurals) and tags to generate them\n" + \
            "dMiscPlur = " + str(dMiscPlur) + "\n\n" + \
            "# dictionary of feminine forms and tags to generate masculine forms (singular and plural)\n" + \
            "dMasForm = " + str(dMasForm) + "\n"
    open(sp+"/modules/mfsp_data.py", "w", encoding="utf-8", newline="\n").write(sCode)

    if bJS:
        ## write file for JavaScript
        sCode = '{\n' + \
                '    "lTagMiscPlur": ' +  json.dumps(lTagMiscPlur, ensure_ascii=False) + ",\n" + \
                '    "lTagFemForm": ' +  json.dumps(lTagFemForm, ensure_ascii=False) + ",\n" + \
                '    "dMiscPlur": ' +  json.dumps(dMiscPlur, ensure_ascii=False) + ",\n" + \
                '    "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="")