Grammalecte  Check-in [ca4c833876]

link = http://grammalecte.net
description = Correcteur grammatical pour le français.
extras = README_fr.txt
logo = logo.png

# main dictionary
lexicon_src = lexicons/French.lex
dic_filenames = fr-allvars,fr-classic,fr-reform
dic_name = Français,Français (Classique/Moderne),Français (Réforme 1990)
dic_filter = ,[*CMPX]$,[*RPX]$
dic_default_filename_py = fr-allvars
dic_default_filename_js = fr-allvars
# extended dictionary
lexicon_extended_src = lexicons/French.extended.lex
dic_extended_filename = fr.extended
dic_extended_name = Français - dictionnaire étendu
# community dictionary
lexicon_community_src = lexicons/French.community.lex
dic_community_filename = fr.community

               'shortname': '“Classique”',
               'asciiName': 'fr-classique',
               'mozAsciiName': 'fr-FR-classic',
               'subDicts': '*MCX',
               'mozId': 'fr-dicollecte-classique',
               'description': "Dictionnaire français “Classique”" }









dREFORME1990 = { 'name': 'DICTIONNAIRE ORTHOGRAPHIQUE FRANÇAIS “RÉFORME 1990”',
                 'shortname': '“Réforme 1990”',
                 'asciiName': 'fr-reforme1990',
                 'mozAsciiName': 'fr-FR-reform',
                 'subDicts': '*RX',
                 'mozId': 'fr-dicollecte-reforme1990',
                 'description': "Dictionnaire français “Réforme 1990”" }

    return s.replace("\u2019", "'").replace("\u2013", "–").replace("\u2014", "—")


class TestDictionary (unittest.TestCase):

    @classmethod
    def setUpClass (cls):
        cls.oDic = IBDAWG("${dic_main_filename_py}")

    def test_lookup (self):
        for sWord in ["branche", "Émilie"]:
            self.assertTrue(self.oDic.lookup(sWord), sWord)

    def test_lookup_failed (self):
        for sWord in ["Branche", "BRANCHE", "BranchE", "BRanche", "BRAnCHE", "émilie"]:

            if not oSpellChecker:
                xCurCtx = uno.getComponentContext()
                oGC = self.ctx.ServiceManager.createInstanceWithContext("org.openoffice.comp.pyuno.Lightproof.grammalecte", self.ctx)
                if hasattr(oGC, "getSpellChecker"):
                    # https://bugs.documentfoundation.org/show_bug.cgi?id=97790
                    oSpellChecker = oGC.getSpellChecker()
                else:
                    oSpellChecker = SpellChecker("${lang}", "fr-allvars.bdic")
            if not oLexicographe:
                oLexicographe = lxg.Lexicographe(oSpellChecker)
        except:
            traceback.print_exc()
        
    def execute (self, args):
        if not args:

                sPersonalDicJSON = self.xOptionNode.getPropertyValue("personal_dic")
                if sPersonalDicJSON:
                    try:
                        personal_dic = json.loads(sPersonalDicJSON)
                    except:
                        print("Graphspell: wrong personal_dic")
                        traceback.print_exc()
            self.oGraphspell = SpellChecker("fr", "fr-allvars.bdic", "", "", personal_dic)
            self.loadHunspell()
            # print("Graphspell: init done")
        except:
            print("Graphspell: init failed")
            traceback.print_exc()
    
    def loadHunspell (self):

    <<- morphex(\2, ":V", ":(?:G|2p|3p!|[ISK].*:2s)")
    -2>> =suggVerb(@, ":2s")                                 # Conjugaison erronée. Accord avec « \1 ». Le verbe devrait être à la 2ᵉ personne du singulier.

TEST: Tu ne {{ment}} jamais.
TEST: Tu {{a}} mal ?
TEST: Tu ne le lui {{prend}} pas.
TEST: Tu ne m’{{attendra}} pas.
TEST: toi qui n’y {{connaît}} rien, ne nous ennuie pas avec tes théories.


## 3sg
__[i]/conj(conj_il)__
    (?<!t’)(il) +({w_1})  @@0,$
    <<- morphex(\2, ":V", ":(?:3s|P|G)") and not (morph(\2, ":[PQ]", False) and morph(word(-1), ":V0.*:3s", False, False))
    -2>> =suggVerb(@, ":3s")                                 # Conjugaison erronée. Accord avec « \1 ». Le verbe devrait être à la 3ᵉ personne du singulier.

import sys
import os
import collections
import json
import time
import re
import traceback

from . import str_transform as st
from .progressbar import ProgressBar



def readFile (spf):

        lTag  = [];   dTag  = {}; nTag  = 0; dTagOccur = {}
        nErr = 0

        try:
            zFilter = re.compile(sSelectFilterRegex)  if sSelectFilterRegex  else None
        except:
            print(" # Error. Wrong filter regex. Filter ignored.")
            traceback.print_exc()
            zFilter = None

        # read lexicon
        if type(src) is str:
            iterable = readFile(src)
        else:
            iterable = src

                dTagOccur[sTag] = dTagOccur.get(sTag, 0) + 1
                aEntry.add((sFlex, dAff[sAff], dTag[sTag]))
        if not aEntry:
            raise ValueError("# Error. Empty lexicon")
        
        # Preparing DAWG
        print(" > Preparing list of words")
        print(" Filter: " + (sSelectFilterRegex or "[None]"))
        lVal = lChar + lAff + lTag
        lWord = [ [dChar[c] for c in sFlex] + [iAff+nChar] + [iTag+nChar+nAff]  for sFlex, iAff, iTag in aEntry ]
        aEntry = None
        
        # Dictionary of arc values occurrency, to sort arcs of each node
        dValOccur = dict( [ (dChar[c], dCharOccur[c])  for c in dChar ] \
                        + [ (dAff[aff]+nChar, dAffOccur[aff]) for aff in dAff ] \

#!python3

# Lexicon builder

import argparse
from distutils import dir_util

import graphspell.dawg as fsa
from graphspell.ibdawg import IBDAWG


def build (spfSrc, sLangCode, sLangName, sfDict, bJSON=False, sDicName="", sFilter="", cStemmingMethod="S", nCompressMethod=1):
    "transform a text lexicon as a binary indexable dictionary"
    oDAWG = fsa.DAWG(spfSrc, cStemmingMethod, sLangCode, sLangName, sDicName, sFilter)
    dir_util.mkpath("graphspell/_dictionaries")
    oDAWG.writeInfo("graphspell/_dictionaries/" + sfDict + ".info.txt")
    oDAWG.writeBinary("graphspell/_dictionaries/" + sfDict + ".bdic", int(nCompressMethod))
    if bJSON:
        dir_util.mkpath("graphspell-js/_dictionaries")
        oDic = IBDAWG(sfDict + ".bdic")
        oDic.writeAsJSObject("graphspell-js/_dictionaries/" + sfDict + ".json", bBinaryDictAsHexString=True)

    dVars["dic_main_filename_js"] = ""
    dVars["dic_extended_filename_py"] = ""
    dVars["dic_extended_filename_js"] = ""
    dVars["dic_community_filename_py"] = ""
    dVars["dic_community_filename_js"] = ""
    dVars["dic_personal_filename_py"] = ""
    dVars["dic_personal_filename_js"] = ""
    lDict = [ ("main", s)  for s in dVars['dic_filenames'].split(",") ]
    if bExtendedDict:
        lDict.append(("extended", dVars['dic_extended_filename']))
    if bCommunityDict:
        lDict.append(("community", dVars['dic_community_filename']))
    if bPersonalDict:
        lDict.append(("personal", dVars['dic_personal_filename']))
    for sType, sFileName in lDict:
        spfPyDic = "graphspell/_dictionaries/" + sFileName + ".bdic"
        spfJSDic = "graphspell-js/_dictionaries/" + sFileName + ".json"
        if not os.path.isfile(spfPyDic) or (bJavaScript and not os.path.isfile(spfJSDic)):
            buildDictionary(dVars, sType, bJavaScript)
        print(spfPyDic)
        file_util.copy_file(spfPyDic, "grammalecte/graphspell/_dictionaries")
        dVars['dic_'+sType+'_filename_py'] = sFileName + '.bdic'
        if bJavaScript:
            print(spfJSDic)
            file_util.copy_file(spfJSDic, "grammalecte-js/graphspell/_dictionaries")
            dVars['dic_'+sType+'_filename_js'] = sFileName + '.json'
    dVars['dic_main_filename_py'] = dVars['dic_default_filename_py'] + ".bdic"
    dVars['dic_main_filename_js'] = dVars['dic_default_filename_js'] + ".json"


def buildDictionary (dVars, sType, bJavaScript=False):
    if sType == "main":
        spfLexSrc = dVars['lexicon_src']
        l_sfDictDst = dVars['dic_filenames'].split(",")
        l_sDicName = dVars['dic_name'].split(",")
        l_sFilter = dVars['dic_filter'].split(",")
        for sfDictDst, sDicName, sFilter in zip(l_sfDictDst, l_sDicName, l_sFilter):
            lex_build.build(spfLexSrc, dVars['lang'], dVars['lang_name'], sfDictDst, bJavaScript, sDicName, sFilter, dVars['stemming_method'], int(dVars['fsa_method']))
    else:
        if sType == "extended":
            spfLexSrc = dVars['lexicon_extended_src']
            sfDictDst = dVars['dic_extended_filename']
            sDicName = dVars['dic_extended_name']
        elif sType == "community":
            spfLexSrc = dVars['lexicon_community_src']
            sfDictDst = dVars['dic_community_filename']
            sDicName = dVars['dic_community_name']
        elif sType == "personal":
            spfLexSrc = dVars['lexicon_personal_src']
            sfDictDst = dVars['dic_personal_filename']
            sDicName = dVars['dic_personal_name']
        lex_build.build(spfLexSrc, dVars['lang'], dVars['lang_name'], sfDictDst, bJavaScript, sDicName, "", dVars['stemming_method'], int(dVars['fsa_method']))



def main ():
    print("Python: " + sys.version)
    xParser = argparse.ArgumentParser()
    xParser.add_argument("lang", type=str, nargs='+', help="lang project to generate (name of folder in /lang)")