Grammalecte  Changes On Branch 759321730cadb9e4

"""
Grammar checker
"""

from .grammar_checker import *

"""
Grammalecte, grammar checker
"""

import importlib
import json

from . import text


class GrammarChecker:
    "GrammarChecker: Wrapper for the grammar checker engine"

    def __init__ (self, sLangCode, sContext="Python"):
        self.sLangCode = sLangCode
        # Grammar checker engine
        self.gce = importlib.import_module("."+sLangCode, "grammalecte")
        self.gce.load(sContext)
        # Spell checker
        self.oSpellChecker = self.gce.getSpellChecker()
        # Lexicographer
        self.oLexicographer = None
        # Text formatter
        self.oTextFormatter = None

    def getGCEngine (self):
        "return the grammar checker object"
        return self.gce

    def getSpellChecker (self):
        "return the spell checker object"
        return self.oSpellChecker

    def getTextFormatter (self):
        "load and return the text formatter"
        if self.oTextFormatter is None:
            tf = importlib.import_module("."+self.sLangCode+".textformatter", "grammalecte")
            self.oTextFormatter = tf.TextFormatter()
        return self.oTextFormatter

    def getLexicographer (self):
        "load and return the lexicographer"
        if self.oLexicographer is None:
            lxg = importlib.import_module("."+self.sLangCode+".lexicographe", "grammalecte")
            self.oLexicographer = lxg.Lexicographe(self.oSpellChecker)
        return self.oLexicographer

    def displayGCOptions (self):
        "display the grammar checker options"
        self.gce.displayOptions()

    def getParagraphErrors (self, sText, dOptions=None, bContext=False, bSpellSugg=False, bDebug=False):
        "returns a tuple: (grammar errors, spelling errors)"
        aGrammErrs = self.gce.parse(sText, "FR", bDebug=bDebug, dOptions=dOptions, bContext=bContext)
        aSpellErrs = self.oSpellChecker.parseParagraph(sText, bSpellSugg)
        return aGrammErrs, aSpellErrs

    def getParagraphWithErrors (self, sText, dOptions=None, bEmptyIfNoErrors=False, bSpellSugg=False, nWidth=100, bDebug=False):
        "parse text and return a readable text with underline errors"
        aGrammErrs, aSpellErrs = self.getParagraphErrors(sText, dOptions, False, bSpellSugg, bDebug)
        if bEmptyIfNoErrors and not aGrammErrs and not aSpellErrs:
            return ("", [])
        return text.generateParagraph(sText, aGrammErrs, aSpellErrs, nWidth)

    def getTextWithErrors (self, sText, bEmptyIfNoErrors=False, bSpellSugg=False, nWidth=100, bDebug=False):
        "[todo]"

    def getParagraphErrorsAsJSON (self, iIndex, sText, dOptions=None, bContext=False, bEmptyIfNoErrors=False, bSpellSugg=False, bReturnText=False, lLineSet=None, bDebug=False):
        "parse text and return errors as a JSON string"
        aGrammErrs, aSpellErrs = self.getParagraphErrors(sText, dOptions, bContext, bSpellSugg, bDebug)
        aGrammErrs = list(aGrammErrs)
        if bEmptyIfNoErrors and not aGrammErrs and not aSpellErrs:
            return ""
        if lLineSet:
            aGrammErrs, aSpellErrs = text.convertToXY(aGrammErrs, aSpellErrs, lLineSet)
            return json.dumps({ "lGrammarErrors": aGrammErrs, "lSpellingErrors": aSpellErrs }, ensure_ascii=False)
        if bReturnText:
            return json.dumps({ "iParagraph": iIndex, "sText": sText, "lGrammarErrors": aGrammErrs, "lSpellingErrors": aSpellErrs }, ensure_ascii=False)
        return json.dumps({ "iParagraph": iIndex, "lGrammarErrors": aGrammErrs, "lSpellingErrors": aSpellErrs }, ensure_ascii=False)

    def getTextErrorsAsJSON (self, sText, bContext=False, bEmptyIfNoErrors=False, bSpellSugg=False, bReturnText=False, bDebug=False):
        "[todo]"

"""
Grammalecte
Grammar checker engine
"""

import re
import traceback


#import unicodedata
from itertools import chain

from ..graphspell.spellchecker import SpellChecker
from ..graphspell.echo import echo

from .. import text


from . import gc_options


try:
    # LibreOffice / OpenOffice
    from com.sun.star.linguistic2 import SingleProofreadingError
    from com.sun.star.text.TextMarkupType import PROOFREADING
    from com.sun.star.beans import PropertyValue
    #import lightproof_handler_${implname} as opt
    _bWriterError = True
except ImportError:
    _bWriterError = False


__all__ = [ "lang", "locales", "pkg", "name", "version", "author", \
            "load", "parse", "getSpellChecker", \
            "setOption", "setOptions", "getOptions", "getDefaultOptions", "getOptionsLabels", "resetOptions", "displayOptions", \
            "ignoreRule", "resetIgnoreRules", "reactivateRule", "listRules", "displayRules", "setWriterUnderliningStyle" ]

__version__ = "${version}"


lang = "${lang}"
locales = ${loc}
pkg = "${implname}"
name = "${name}"
version = "${version}"
author = "${author}"

# Modules
_rules = None                               # module gc_rules
_rules_graph = None                         # module gc_rules_graph

# Data
_sAppContext = ""                           # what software is running
_dOptions = None
_dOptionsColors = None
_oSpellChecker = None
_oTokenizer = None



_aIgnoredRules = set()

# Writer underlining style

_bMulticolor = True
_nUnderliningStyle = 0


#### Initialization

def load (sContext="Python", sColorType="aRGB"):
    "initialization of the grammar checker"
    global _oSpellChecker
    global _sAppContext
    global _dOptions
    global _dOptionsColors
    global _oTokenizer
    try:
        _oSpellChecker = SpellChecker("${lang}", "${dic_main_filename_py}", "${dic_community_filename_py}", "${dic_personal_filename_py}")
        _sAppContext = sContext
        _dOptions = gc_options.getOptions(sContext).copy()   # duplication necessary, to be able to reset to default
        _dOptionsColors = gc_options.getOptionsColors(sContext, sColorType)
        _oTokenizer = _oSpellChecker.getTokenizer()
        _oSpellChecker.activateStorage()


    except:
        traceback.print_exc()




def getSpellChecker ():
    "return the spellchecker object"
    return _oSpellChecker











#### Rules

def _getRules (bParagraph):
    try:
        if not bParagraph:
            return _rules.lSentenceRules

def reactivateRule (sRuleId):
    "(re)activate rule <sRuleId>"
    _aIgnoredRules.discard(sRuleId)


def listRules (sFilter=None):
    "generator: returns typle (sOption, sLineId, sRuleId)"
    if sFilter:
        try:
            zFilter = re.compile(sFilter)
        except re.error:
            echo("# Error. List rules: wrong regex.")
            sFilter = None
    # regex rules

def displayRules (sFilter=None):
    "display the name of rules, with the filter <sFilter>"
    echo("List of rules. Filter: << " + str(sFilter) + " >>")
    for sOption, sLineId, sRuleId, sType in listRules(sFilter):
        echo("{:<8} {:<10} {:<10} {}".format(sOption, sLineId, sRuleId, sType))


#### Options

def setOption (sOpt, bVal):
    "set option <sOpt> with <bVal> if it exists"
    if sOpt in _dOptions:
        _dOptions[sOpt] = bVal


def setOptions (dOpt):
    "update the dictionary of options with <dOpt>"
    for sKey, bVal in dOpt.items():
        if sKey in _dOptions:
            _dOptions[sKey] = bVal


def getOptions ():
    "return the dictionary of current options"
    return _dOptions


def getDefaultOptions ():
    "return the dictionary of default options"
    return gc_options.getOptions(_sAppContext).copy()


def getOptionsLabels (sLang):
    "return options labels"
    return gc_options.getUI(sLang)


def displayOptions (sLang="${lang}"):
    "display the list of grammar checking options"
    echo("Options:")
    echo("\n".join( [ k+":\t"+str(v)+"\t"+gc_options.getUI(sLang).get(k, ("?", ""))[0]  for k, v  in sorted(_dOptions.items()) ] ))
    echo("")


def resetOptions ():
    "set options to default values"
    global _dOptions
    _dOptions = getDefaultOptions()


def setWriterUnderliningStyle (sStyle="BOLDWAVE", bMulticolor=True):
    "set underlining style for Writer (WAVE, BOLDWAVE, BOLD)"
    global _nUnderliningStyle
    global _bMulticolor
    # https://api.libreoffice.org/docs/idl/ref/FontUnderline_8idl.html
    # WAVE: 10, BOLD: 12, BOLDWAVE: 18 DASH: 5

        _nUnderliningStyle = 5
    else:
        _nUnderliningStyle = 0
    _bMulticolor = bMulticolor


#### Parsing






























def parse (sText, sCountry="${country_default}", bDebug=False, dOptions=None, bContext=False, bFullInfo=False):
    "init point to analyse <sText> and returns an iterable of errors or (with option <bFullInfo>) paragraphs errors and sentences with tokens and errors"
    oText = TextParser(sText)
    return oText.parse(sCountry, bDebug, dOptions, bContext, bFullInfo)

        #for nPos, dToken in self.dTokenPos.items():
        #    s += "{}\t{}\n".format(nPos, dToken)
        return s

    def parse (self, sCountry="${country_default}", bDebug=False, dOptions=None, bContext=False, bFullInfo=False):
        "analyses <sText> and returns an iterable of errors or (with option <bFullInfo>) paragraphs errors and sentences with tokens and errors"
        #sText = unicodedata.normalize("NFC", sText)
        dOpt = dOptions or _dOptions
        bShowRuleId = option('idrule')
        # parse paragraph
        try:
            self.parseText(self.sText, self.sText0, True, 0, sCountry, dOpt, bShowRuleId, bDebug, bContext)
        except:
            raise
        if bFullInfo:
            lParagraphErrors = list(self.dError.values())

                for zRegex, bUppercase, sLineId, sRuleId, nPriority, lActions in lRuleGroup:
                    if sRuleId not in _aIgnoredRules:
                        for m in zRegex.finditer(sText):
                            bCondMemo = None
                            for sFuncCond, cActionType, sWhat, *eAct in lActions:
                                # action in lActions: [ condition, action type, replacement/suggestion/action[, iGroup[, message, URL]] ]
                                try:
                                    bCondMemo = not sFuncCond or globals()[sFuncCond](sText, sText0, m, self.dTokenPos, sCountry, bCondMemo)
                                    if bCondMemo:
                                        if bDebug:
                                            echo("RULE: " + sLineId)
                                        if cActionType == "-":
                                            # grammar error
                                            nErrorStart = nOffset + m.start(eAct[0])
                                            if nErrorStart not in self.dError or nPriority > self.dErrorPriority.get(nErrorStart, -1):
                                                self.dError[nErrorStart] = self._createErrorFromRegex(sText, sText0, sWhat, nOffset, m, eAct[0], sLineId, sRuleId, bUppercase, eAct[1], eAct[2], bShowRuleId, sOption, bContext)
                                                self.dErrorPriority[nErrorStart] = nPriority
                                                self.dSentenceError[nErrorStart] = self.dError[nErrorStart]
                                        elif cActionType == "~":
                                            # text processor
                                            sText = self.rewriteText(sText, sWhat, eAct[0], m, bUppercase)
                                            bChange = True
                                            if bDebug:
                                                echo("~ " + sText + "  -- " + m.group(eAct[0]) + "  # " + sLineId)
                                        elif cActionType == "=":
                                            # disambiguation
                                            if not bParagraph:
                                                globals()[sWhat](sText, m, self.dTokenPos)
                                                if bDebug:
                                                    echo("= " + m.group(0) + "  # " + sLineId)
                                        elif cActionType == ">":
                                            # we do nothing, this test is just a condition to apply all following actions
                                            pass
                                        else:
                                            echo("# error: unknown action at " + sLineId)

        for sLineId, nextNodeKey in dNode.items():
            bCondMemo = None
            for sRuleId in dGraph[nextNodeKey]:
                try:
                    if bDebug:
                        echo("   >TRY: " + sRuleId + " " + sLineId)
                    _, sOption, sFuncCond, cActionType, sWhat, *eAct = _rules_graph.dRule[sRuleId]
                    # Suggestion    [ option, condition, "-", replacement/suggestion/action, iTokenStart, iTokenEnd, cStartLimit, cEndLimit, bCaseSvty, nPriority, sMessage, sURL ]
                    # TextProcessor [ option, condition, "~", replacement/suggestion/action, iTokenStart, iTokenEnd, bCaseSvty ]
                    # Disambiguator [ option, condition, "=", replacement/suggestion/action ]
                    # Tag           [ option, condition, "/", replacement/suggestion/action, iTokenStart, iTokenEnd ]
                    # Immunity      [ option, condition, "!", "",                            iTokenStart, iTokenEnd ]
                    # Test          [ option, condition, ">", "" ]
                    if not sOption or dOptions.get(sOption, False):
                        bCondMemo = not sFuncCond or globals()[sFuncCond](self.lToken, nTokenOffset, nLastToken, sCountry, bCondMemo, self.dTags, self.sSentence, self.sSentence0)
                        if bCondMemo:
                            if cActionType == "-":
                                # grammar error
                                iTokenStart, iTokenEnd, cStartLimit, cEndLimit, bCaseSvty, nPriority, sMessage, sURL = eAct
                                nTokenErrorStart = nTokenOffset + iTokenStart  if iTokenStart > 0  else nLastToken + iTokenStart
                                if "bImmune" not in self.lToken[nTokenErrorStart]:
                                    nTokenErrorEnd = nTokenOffset + iTokenEnd  if iTokenEnd > 0  else nLastToken + iTokenEnd

                                nTokenEnd = nTokenOffset + eAct[1]  if eAct[1] > 0  else nLastToken + eAct[1]
                                self._tagAndPrepareTokenForRewriting(sWhat, nTokenStart, nTokenEnd, nTokenOffset, nLastToken, eAct[2], bDebug)
                                bChange = True
                                if bDebug:
                                    echo("    TEXT_PROCESSOR: [{}:{}]  > {}".format(self.lToken[nTokenStart]["sValue"], self.lToken[nTokenEnd]["sValue"], sWhat))
                            elif cActionType == "=":
                                # disambiguation
                                globals()[sWhat](self.lToken, nTokenOffset, nLastToken)
                                if bDebug:
                                    echo("    DISAMBIGUATOR: ({})  [{}:{}]".format(sWhat, self.lToken[nTokenOffset+1]["sValue"], self.lToken[nLastToken]["sValue"]))
                            elif cActionType == ">":
                                # we do nothing, this test is just a condition to apply all following actions
                                if bDebug:
                                    echo("    COND_OK")
                            elif cActionType == "/":

        return bChange

    def _createErrorFromRegex (self, sText, sText0, sRepl, nOffset, m, iGroup, sLineId, sRuleId, bUppercase, sMsg, sURL, bShowRuleId, sOption, bContext):
        nStart = nOffset + m.start(iGroup)
        nEnd = nOffset + m.end(iGroup)
        # suggestions
        if sRepl[0:1] == "=":
            sSugg = globals()[sRepl[1:]](sText, m)
            lSugg = sSugg.split("|")  if sSugg  else []
        elif sRepl == "_":
            lSugg = []
        else:
            lSugg = m.expand(sRepl).split("|")
        if bUppercase and lSugg and m.group(iGroup)[0:1].isupper():
            lSugg = list(map(lambda s: s[0:1].upper()+s[1:], lSugg))
        # Message
        sMessage = globals()[sMsg[1:]](sText, m)  if sMsg[0:1] == "="  else  m.expand(sMsg)
        if bShowRuleId:
            sMessage += "  #" + sLineId + " / " + sRuleId
        #
        if _bWriterError:
            return self._createErrorForWriter(nStart, nEnd - nStart, sRuleId, sOption, sMessage, lSugg, sURL)
        return self._createErrorAsDict(nStart, nEnd, sLineId, sRuleId, sOption, sMessage, lSugg, sURL, bContext)

    def _createErrorFromTokens (self, sSugg, nTokenOffset, nLastToken, iFirstToken, nStart, nEnd, sLineId, sRuleId, bCaseSvty, sMsg, sURL, bShowRuleId, sOption, bContext):
        # suggestions
        if sSugg[0:1] == "=":
            sSugg = globals()[sSugg[1:]](self.lToken, nTokenOffset, nLastToken)
            lSugg = sSugg.split("|")  if sSugg  else []
        elif sSugg == "_":
            lSugg = []
        else:
            lSugg = self._expand(sSugg, nTokenOffset, nLastToken).split("|")
        if bCaseSvty and lSugg and self.lToken[iFirstToken]["sValue"][0:1].isupper():
            lSugg = list(map(lambda s: s[0:1].upper()+s[1:], lSugg))
        # Message
        sMessage = globals()[sMsg[1:]](self.lToken, nTokenOffset, nLastToken)  if sMsg[0:1] == "="  else self._expand(sMsg, nTokenOffset, nLastToken)
        if bShowRuleId:
            sMessage += "  #" + sLineId + " / " + sRuleId
        #
        if _bWriterError:
            return self._createErrorForWriter(nStart, nEnd - nStart, sRuleId, sOption, sMessage, lSugg, sURL)
        return self._createErrorAsDict(nStart, nEnd, sLineId, sRuleId, sOption, sMessage, lSugg, sURL, bContext)

        if sRepl == "*":
            sNew = " " * nLen
        elif sRepl == "_":
            sNew = "_" * nLen
        elif sRepl == "@":
            sNew = "@" * nLen
        elif sRepl[0:1] == "=":
            sNew = globals()[sRepl[1:]](sText, m)
            sNew = sNew + " " * (nLen-len(sNew))
            if bUppercase and m.group(iGroup)[0:1].isupper():
                sNew = sNew.capitalize()
        else:
            sNew = m.expand(sRepl)
            sNew = sNew + " " * (nLen-len(sNew))
        return sText[0:m.start(iGroup)] + sNew + sText[m.end(iGroup):]

            if nTokenRewriteEnd - nTokenRewriteStart == 0:
                self.lToken[nTokenRewriteStart]["sNewValue"] = "_"
            else:
                for i in range(nTokenRewriteStart, nTokenRewriteEnd+1):
                    self.lToken[i]["sNewValue"] = "_"
        else:
            if sWhat.startswith("="):
                sWhat = globals()[sWhat[1:]](self.lToken, nTokenOffset, nLastToken)
            else:
                sWhat = self._expand(sWhat, nTokenOffset, nLastToken)
            bUppercase = bCaseSvty and self.lToken[nTokenRewriteStart]["sValue"][0:1].isupper()
            if nTokenRewriteEnd - nTokenRewriteStart == 0:
                # one token
                if bUppercase:
                    sWhat = sWhat[0:1].upper() + sWhat[1:]

                except KeyError:
                    echo(self)
                    echo(dToken)
        if bDebug:
            echo("  TEXT REWRITED: " + self.sSentence)
        self.lToken.clear()
        self.lToken = lNewToken


#### common functions

def option (sOpt):
    "return True if option <sOpt> is active"
    return _dOptions.get(sOpt, False)


#### Functions to get text outside pattern scope

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

_zNextWord = re.compile(r" +(\w[\w-]*)")
_zPrevWord = re.compile(r"(\w[\w-]*) +$")

def nextword (s, iStart, n):
    "get the nth word of the input string or empty string"
    m = re.match("(?: +[\\w%-]+){" + str(n-1) + "} +([\\w%-]+)", s[iStart:])
    if not m:
        return None
    return (iStart+m.start(1), m.group(1))


def prevword (s, iEnd, n):
    "get the (-)nth word of the input string or empty string"
    m = re.search("([\\w%-]+) +(?:[\\w%-]+ +){" + str(n-1) + "}$", s[:iEnd])
    if not m:
        return None
    return (m.start(1), m.group(1))


def nextword1 (s, iStart):
    "get next word (optimization)"
    m = _zNextWord.match(s[iStart:])
    if not m:
        return None
    return (iStart+m.start(1), m.group(1))


def prevword1 (s, iEnd):
    "get previous word (optimization)"
    m = _zPrevWord.search(s[:iEnd])
    if not m:
        return None
    return (m.start(1), m.group(1))


def look (s, sPattern, sNegPattern=None):
    "seek sPattern in s (before/after/fulltext), if sNegPattern not in s"
    if sNegPattern and re.search(sNegPattern, s):
        return False
    if re.search(sPattern, s):
        return True
    return False


def look_chk1 (dTokenPos, s, nOffset, sPattern, sPatternGroup1, sNegPatternGroup1=""):
    "returns True if s has pattern sPattern and m.group(1) has pattern sPatternGroup1"
    m = re.search(sPattern, s)
    if not m:
        return False
    try:
        sWord = m.group(1)
        nPos = m.start(1) + nOffset
    except IndexError:
        return False
    return morph(dTokenPos, (nPos, sWord), sPatternGroup1, sNegPatternGroup1)



#### Analyse groups for regex rules

def displayInfo (dTokenPos, tWord):
    "for debugging: retrieve info of word"
    if not tWord:
        echo("> nothing to find")
        return True
    lMorph = _oSpellChecker.getMorph(tWord[1])
    if not lMorph:
        echo("> not in dictionary")
        return True
    echo("TOKENS:", dTokenPos)
    if tWord[0] in dTokenPos and "lMorph" in dTokenPos[tWord[0]]:
        echo("DA: " + str(dTokenPos[tWord[0]]["lMorph"]))
    echo("FSA: " + str(lMorph))
    return True


def morph (dTokenPos, tWord, 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 not tWord:
        return bNoWord
    lMorph = dTokenPos[tWord[0]]["lMorph"]  if tWord[0] in dTokenPos and "lMorph" in dTokenPos[tWord[0]]  else _oSpellChecker.getMorph(tWord[1])
    if not lMorph:
        return False
    # check negative condition
    if sNegPattern:
        if sNegPattern == "*":
            # all morph must match sPattern
            zPattern = re.compile(sPattern)
            return all(zPattern.search(sMorph)  for sMorph in lMorph)
        zNegPattern = re.compile(sNegPattern)
        if any(zNegPattern.search(sMorph)  for sMorph in lMorph):
            return False
    # search sPattern
    zPattern = re.compile(sPattern)
    return any(zPattern.search(sMorph)  for sMorph in lMorph)


def analyse (sWord, sPattern, sNegPattern=""):
    "analyse a word, returns True if not sNegPattern in word morphologies and sPattern in word morphologies (disambiguation off)"
    lMorph = _oSpellChecker.getMorph(sWord)
    if not lMorph:
        return False
    # check negative condition
    if sNegPattern:
        if sNegPattern == "*":
            zPattern = re.compile(sPattern)
            return all(zPattern.search(sMorph)  for sMorph in lMorph)
        zNegPattern = re.compile(sNegPattern)
        if any(zNegPattern.search(sMorph)  for sMorph in lMorph):
            return False
    # search sPattern
    zPattern = re.compile(sPattern)
    return any(zPattern.search(sMorph)  for sMorph in lMorph)


#### Analyse tokens for graph rules

def g_value (dToken, sValues, nLeft=None, nRight=None):
    "test if <dToken['sValue']> is in sValues (each value should be separated with |)"
    sValue = "|"+dToken["sValue"]+"|"  if nLeft is None  else "|"+dToken["sValue"][slice(nLeft, nRight)]+"|"
    if sValue in sValues:
        return True
    if dToken["sValue"][0:2].istitle(): # we test only 2 first chars, to make valid words such as "Laissez-les", "Passe-partout".
        if sValue.lower() in sValues:
            return True
    elif dToken["sValue"].isupper():
        #if sValue.lower() in sValues:
        #    return True
        sValue = "|"+sValue[1:].capitalize()
        if sValue in sValues:
            return True
        sValue = sValue.lower()
        if sValue in sValues:
            return True
    return False


def g_morph (dToken, sPattern, sNegPattern="", nLeft=None, nRight=None, bMemorizeMorph=True):
    "analyse a token, return True if <sNegPattern> not in morphologies and <sPattern> in morphologies"
    if "lMorph" in dToken:
        lMorph = dToken["lMorph"]
    else:
        if nLeft is not None:
            lMorph = _oSpellChecker.getMorph(dToken["sValue"][slice(nLeft, nRight)])
            if bMemorizeMorph:
                dToken["lMorph"] = lMorph
        else:
            lMorph = _oSpellChecker.getMorph(dToken["sValue"])
    if not lMorph:
        return False
    # check negative condition
    if sNegPattern:
        if sNegPattern == "*":
            # all morph must match sPattern
            zPattern = re.compile(sPattern)
            return all(zPattern.search(sMorph)  for sMorph in lMorph)
        zNegPattern = re.compile(sNegPattern)
        if any(zNegPattern.search(sMorph)  for sMorph in lMorph):
            return False
    # search sPattern
    zPattern = re.compile(sPattern)
    return any(zPattern.search(sMorph)  for sMorph in lMorph)


def g_analyse (dToken, sPattern, sNegPattern="", nLeft=None, nRight=None, bMemorizeMorph=True):
    "analyse a token, return True if <sNegPattern> not in morphologies and <sPattern> in morphologies (disambiguation off)"
    if nLeft is not None:
        lMorph = _oSpellChecker.getMorph(dToken["sValue"][slice(nLeft, nRight)])
        if bMemorizeMorph:
            dToken["lMorph"] = lMorph
    else:
        lMorph = _oSpellChecker.getMorph(dToken["sValue"])
    if not lMorph:
        return False
    # check negative condition
    if sNegPattern:
        if sNegPattern == "*":
            # all morph must match sPattern
            zPattern = re.compile(sPattern)
            return all(zPattern.search(sMorph)  for sMorph in lMorph)
        zNegPattern = re.compile(sNegPattern)
        if any(zNegPattern.search(sMorph)  for sMorph in lMorph):
            return False
    # search sPattern
    zPattern = re.compile(sPattern)
    return any(zPattern.search(sMorph)  for sMorph in lMorph)


def g_merged_analyse (dToken1, dToken2, cMerger, sPattern, sNegPattern="", bSetMorph=True):
    "merge two token values, return True if <sNegPattern> not in morphologies and <sPattern> in morphologies (disambiguation off)"
    lMorph = _oSpellChecker.getMorph(dToken1["sValue"] + cMerger + dToken2["sValue"])
    if not lMorph:
        return False
    # check negative condition
    if sNegPattern:
        if sNegPattern == "*":
            # all morph must match sPattern
            zPattern = re.compile(sPattern)
            bResult = all(zPattern.search(sMorph)  for sMorph in lMorph)
            if bResult and bSetMorph:
                dToken1["lMorph"] = lMorph
            return bResult
        zNegPattern = re.compile(sNegPattern)
        if any(zNegPattern.search(sMorph)  for sMorph in lMorph):
            return False
    # search sPattern
    zPattern = re.compile(sPattern)
    bResult = any(zPattern.search(sMorph)  for sMorph in lMorph)
    if bResult and bSetMorph:
        dToken1["lMorph"] = lMorph
    return bResult


def g_tag_before (dToken, dTags, sTag):
    "returns True if <sTag> is present on tokens before <dToken>"
    if sTag not in dTags:
        return False
    if dToken["i"] > dTags[sTag][0]:
        return True
    return False


def g_tag_after (dToken, dTags, sTag):
    "returns True if <sTag> is present on tokens after <dToken>"
    if sTag not in dTags:
        return False
    if dToken["i"] < dTags[sTag][1]:
        return True
    return False


def g_tag (dToken, sTag):
    "returns True if <sTag> is present on token <dToken>"
    return "aTags" in dToken and sTag in dToken["aTags"]


def g_meta (dToken, sType):
    "returns True if <sType> is equal to the token type"
    return dToken["sType"] == sType


def g_space_between_tokens (dToken1, dToken2, nMin, nMax=None):
    "checks if spaces between tokens is >= <nMin> and <= <nMax>"
    nSpace = dToken2["nStart"] - dToken1["nEnd"]
    if nSpace < nMin:
        return False
    if nMax is not None and nSpace > nMax:
        return False
    return True


def g_token (lToken, i):
    "return token at index <i> in lToken (or the closest one)"
    if i < 0:
        return lToken[0]
    if i >= len(lToken):
        return lToken[-1]
    return lToken[i]



#### Disambiguator for regex rules

def select (dTokenPos, nPos, sWord, sPattern, lDefault=None):
    "Disambiguation: select morphologies of <sWord> matching <sPattern>"
    if not sWord:
        return True
    if nPos not in dTokenPos:
        echo("Error. There should be a token at this position: ", nPos)
        return True
    lMorph = _oSpellChecker.getMorph(sWord)
    if not lMorph or len(lMorph) == 1:
        return True
    lSelect = [ sMorph  for sMorph in lMorph  if re.search(sPattern, sMorph) ]
    if lSelect:
        if len(lSelect) != len(lMorph):
            dTokenPos[nPos]["lMorph"] = lSelect
    elif lDefault:
        dTokenPos[nPos]["lMorph"] = lDefault
    return True


def exclude (dTokenPos, nPos, sWord, sPattern, lDefault=None):
    "Disambiguation: exclude morphologies of <sWord> matching <sPattern>"
    if not sWord:
        return True
    if nPos not in dTokenPos:
        echo("Error. There should be a token at this position: ", nPos)
        return True
    lMorph = _oSpellChecker.getMorph(sWord)
    if not lMorph or len(lMorph) == 1:
        return True
    lSelect = [ sMorph  for sMorph in lMorph  if not re.search(sPattern, sMorph) ]
    if lSelect:
        if len(lSelect) != len(lMorph):
            dTokenPos[nPos]["lMorph"] = lSelect
    elif lDefault:
        dTokenPos[nPos]["lMorph"] = lDefault
    return True


def define (dTokenPos, nPos, lMorph):
    "Disambiguation: set morphologies of token at <nPos> with <lMorph>"
    if nPos not in dTokenPos:
        echo("Error. There should be a token at this position: ", nPos)
        return True
    dTokenPos[nPos]["lMorph"] = lMorph
    return True


#### Disambiguation for graph rules

def g_select (dToken, sPattern, lDefault=None):
    "Disambiguation: select morphologies for <dToken> according to <sPattern>, always return True"
    lMorph = dToken["lMorph"]  if "lMorph" in dToken  else _oSpellChecker.getMorph(dToken["sValue"])
    if not lMorph or len(lMorph) == 1:
        if lDefault:
            dToken["lMorph"] = lDefault
            #echo("DA:", dToken["sValue"], dToken["lMorph"])
        return True
    lSelect = [ sMorph  for sMorph in lMorph  if re.search(sPattern, sMorph) ]
    if lSelect:
        if len(lSelect) != len(lMorph):
            dToken["lMorph"] = lSelect
    elif lDefault:
        dToken["lMorph"] = lDefault
    #echo("DA:", dToken["sValue"], dToken["lMorph"])
    return True


def g_exclude (dToken, sPattern, lDefault=None):
    "Disambiguation: select morphologies for <dToken> according to <sPattern>, always return True"
    lMorph = dToken["lMorph"]  if "lMorph" in dToken  else _oSpellChecker.getMorph(dToken["sValue"])
    if not lMorph or len(lMorph) == 1:
        if lDefault:
            dToken["lMorph"] = lDefault
            #echo("DA:", dToken["sValue"], dToken["lMorph"])
        return True
    lSelect = [ sMorph  for sMorph in lMorph  if not re.search(sPattern, sMorph) ]
    if lSelect:
        if len(lSelect) != len(lMorph):
            dToken["lMorph"] = lSelect
    elif lDefault:
        dToken["lMorph"] = lDefault
    #echo("DA:", dToken["sValue"], dToken["lMorph"])
    return True


def g_add_morph (dToken, lNewMorph):
    "Disambiguation: add a morphology to a token"
    lMorph = dToken["lMorph"]  if "lMorph" in dToken  else _oSpellChecker.getMorph(dToken["sValue"])
    lMorph.extend(lNewMorph)
    dToken["lMorph"] = lMorph
    return True


def g_define (dToken, lMorph):
    "Disambiguation: set morphologies of <dToken>, always return True"
    dToken["lMorph"] = lMorph
    #echo("DA:", dToken["sValue"], lMorph)
    return True


def g_define_from (dToken, nLeft=None, nRight=None):
    "Disambiguation: set morphologies of <dToken> with slicing its value with <nLeft> and <nRight>"
    if nLeft is not None:
        dToken["lMorph"] = _oSpellChecker.getMorph(dToken["sValue"][slice(nLeft, nRight)])
    else:
        dToken["lMorph"] = _oSpellChecker.getMorph(dToken["sValue"])
    return True


def g_change_meta (dToken, sType):
    "Disambiguation: change type of token"
    dToken["sType"] = sType
    return True



#### GRAMMAR CHECKER PLUGINS

${plugins}


#### CALLABLES FOR REGEX RULES (generated code)

${callables}


#### CALLABLES FOR GRAPH RULES (generated code)

${graph_callables}

"""
Grammar checker default options
"""

# generated code, do not edit


import traceback



























def getUI (sLang):


















    "returns dictionary of UI labels"
    if sLang in _dOptLabel:
        return _dOptLabel[sLang]
    return _dOptLabel["fr"]


def getOptions (sContext="Python"):
    "returns dictionary of options"


    if sContext in _dOpt:
        return _dOpt[sContext]
    return _dOpt["Python"]


def getOptionsColors (sTheme="Default", sColorType="aRGB"):
    "returns dictionary of options colors"
    dOptColor = _dOptColor[sTheme]  if sTheme in _dOptColor  else  _dOptColor["Default"]
    dColorType = _dColorType[sColorType]  if sColorType in _dColorType  else _dColorType["aRGB"]
    try:
        return {  sOpt: dColorType[sColor] for sOpt, sColor in dOptColor.items() }
    except KeyError:
        traceback.print_exc()
        return {}


lStructOpt = ${lStructOpt}


_dOpt = {
    "Python": ${dOptPython},
    "Server": ${dOptServer},
    "Writer": ${dOptWriter}
}

_dColorType= ${dColorType}

from com.sun.star.linguistic2 import XProofreader, XSupportedLocales
from com.sun.star.linguistic2 import ProofreadingResult
from com.sun.star.lang import XServiceInfo, XServiceName, XServiceDisplayName
from com.sun.star.lang import Locale

import helpers
import grammalecte.${lang} as gce
#import lightproof_handler_${implname} as opt_handler
import Options


class Grammalecte (unohelper.Base, XProofreader, XServiceInfo, XServiceName, XServiceDisplayName, XSupportedLocales):

    def __init__ (self, ctx, *args):
        self.ctx = ctx
        self.ServiceName = "com.sun.star.linguistic2.Proofreader"
        self.ImplementationName = "org.openoffice.comp.pyuno.Lightproof." + gce.pkg
        self.SupportedServiceNames = (self.ServiceName, )
        self.locales = []
        for i in gce.locales:
            l = gce.locales[i]
            self.locales.append(Locale(l[0], l[1], l[2]))
        self.locales = tuple(self.locales)
        # debug
        #helpers.startConsole()
        # init
        gce.load("Writer", "nInt")
        # GC options
        #xContext = uno.getComponentContext()
        #opt_handler.load(xContext)
        dOpt = Options.loadOptions("${lang}")
        gce.setOptions(dOpt)
        # dictionaries options
        self.loadUserDictionaries()
        # underlining options
        self.setWriterUnderliningStyle()
        # store for results of big paragraphs
        self.dResult = {}
        self.nMaxRes = 1500

            if nHashedVal in self.dResult:
                return self.dResult[nHashedVal]
        # WORKAROUND ->>>

        xRes.nBehindEndOfSentencePosition = xRes.nStartOfNextSentencePosition

        try:
            xRes.aErrors = tuple(gce.parse(rText, rLocale.Country))
            # ->>> WORKAROUND
            if xRes.nStartOfNextSentencePosition > 3000:
                self.dResult[nHashedVal] = xRes
                self.nRes += 1
                if self.nRes > self.nMaxRes:
                    del self.dResult[self.lLastRes.popleft()]
                    self.nRes = self.nMaxRes
                self.lLastRes.append(nHashedVal)
            # END OF WORKAROUND
        except:
            traceback.print_exc()
        return xRes

    def ignoreRule (self, rid, aLocale):
        gce.ignoreRule(rid)

    def resetIgnoreRules (self):
        gce.resetIgnoreRules()

    # XServiceDisplayName
    def getServiceDisplayName (self, aLocale):
        return gce.name

    # Grammalecte
    def getSpellChecker (self):
        return gce.getSpellChecker()

    def loadUserDictionaries (self):
        try:
            xSettingNode = helpers.getConfigSetting("/org.openoffice.Lightproof_${implname}/Other/", False)
            xChild = xSettingNode.getByName("o_${lang}")
            if xChild.getPropertyValue("use_personal_dic"):
                sJSON = xChild.getPropertyValue("personal_dic")
                if sJSON:
                    oSpellChecker = gce.getSpellChecker();
                    oSpellChecker.setPersonalDictionary(json.loads(sJSON))
        except:
            traceback.print_exc()

    def setWriterUnderliningStyle (self):
        try:
            xSettingNode = helpers.getConfigSetting("/org.openoffice.Lightproof_${implname}/Other/", False)
            xChild = xSettingNode.getByName("o_${lang}")
            sLineType = xChild.getPropertyValue("line_type")
            bMulticolor = bool(xChild.getPropertyValue("line_multicolor"))
            gce.setWriterUnderliningStyle(sLineType, bMulticolor)
        except:
            traceback.print_exc()


g_ImplementationHelper = unohelper.ImplementationHelper()
g_ImplementationHelper.addImplementation(Grammalecte, "org.openoffice.comp.pyuno.Lightproof."+gce.pkg, ("com.sun.star.linguistic2.Proofreader",),)

# g_ImplementationHelper.addImplementation( opt_handler.LightproofOptionsEventHandler, \
#     "org.openoffice.comp.pyuno.LightproofOptionsEventHandler." + gce.pkg, ("com.sun.star.awt.XContainerWindowEventHandler",),)

from com.sun.star.awt import XActionListener
from com.sun.star.beans import PropertyValue

import helpers
import op_strings

try:
    import grammalecte.${lang} as gce
except:
    traceback.print_exc()


def loadOptions (sLang):
    "load options from Grammalecte and change them according to LibreOffice settings, returns a dictionary {option_name: boolean}"
    try:
        xNode = helpers.getConfigSetting("/org.openoffice.Lightproof_${implname}/Leaves", False)
        xChild = xNode.getByName(sLang)
        dOpt = gce.gc_options.getOptions("Writer")
        for sKey in dOpt:
            sValue = xChild.getPropertyValue(sKey)
            if sValue != '':
                dOpt[sKey] = bool(int(sValue))
        return dOpt
    except:
        print("# Error. Unable to load options of language:", sLang)
        traceback.print_exc()
        return gce.gc_options.getOptions("Writer")


def saveOptions (sLang, dOpt):
    "save options in LibreOffice profile"
    try:
        xNode = helpers.getConfigSetting("/org.openoffice.Lightproof_${implname}/Leaves", True)
        xChild = xNode.getByName(sLang)

            setattr(xWidget, k, w)
        self.xDialog.insertByName(name, xWidget)
        return xWidget

    def run (self, sUI):
        try:
            dUI = op_strings.getUI(sUI)
            dOptionUI = gce.gc_options.getUI(sUI)

            # fonts
            xFDTitle = uno.createUnoStruct("com.sun.star.awt.FontDescriptor")
            xFDTitle.Height = 9
            xFDTitle.Weight = uno.getConstantByName("com.sun.star.awt.FontWeight.BOLD")
            xFDTitle.Name = "Verdana"

            nHeight = 10

            self.lOptionWidgets = []

            sProdName, sVersion = helpers.getProductNameAndVersion()
            if True:
                # no tab available (bug)
                for sOptionType, lOptions in gce.gc_options.lStructOpt:
                    x = 10
                    y += 10
                    self._addWidget(sOptionType, 'FixedLine', x, y, nWidth, nHeight, Label = dOptionUI.get(sOptionType, "#err")[0], FontDescriptor= xFDTitle)
                    y += 3
                    for lOptLine in lOptions:
                        x = 15
                        y += 10

        except:
            traceback.print_exc()

    # XActionListener
    def actionPerformed (self, xActionEvent):
        try:
            if xActionEvent.ActionCommand == 'Default':
                self._setWidgets(gce.gc_options.getOptions("Writer"))
            elif xActionEvent.ActionCommand == 'Apply':
                self._save("${lang}")
                self.xContainer.endExecute()
            elif xActionEvent.ActionCommand == 'Cancel':
                self.xContainer.endExecute()
            else:
                print("Wrong command: " + xActionEvent.ActionCommand)
        except:
            traceback.print_exc()

    # Other
    def _setWidgets (self, dOpt):
        for w in self.lOptionWidgets:
            w.State = dOpt.get(w.Name, False)

    def _save (self, sLang):
        try:
            saveOptions(sLang, { w.Name: str(w.State)  for w in self.lOptionWidgets })
            gce.setOptions({ w.Name: bool(w.State)  for w in self.lOptionWidgets })
        except:
            traceback.print_exc()

[args]
lang = fr
lang_name = French
locales = fr_FR fr_BE fr_CA fr_CH fr_LU fr_BF fr_BJ fr_CD fr_CI fr_CM fr_MA fr_ML fr_MU fr_NE fr_RE fr_SN fr_TG
country_default = FR
name = Grammalecte
implname = grammalecte
# always use 3 numbers for version: x.y.z
version = 1.8.2
author = Olivier R.
provider = Grammalecte.net
link = https://grammalecte.net
description = Correcteur grammatical, orthographique et typographique pour le français.
extras = README_fr.txt
logo = logo.png

import os
import re
import time
from contextlib import contextmanager

from ..graphspell.ibdawg import IBDAWG
from ..graphspell.echo import echo
from . import gc_engine as gce
from . import conj
from . import phonet
from . import mfsp


@contextmanager
def timeblock (label, hDst):

        if hDst:
            hDst.write("{:<12.6}".format(end-start))


def perf (sVersion, hDst=None):
    "performance tests"
    print("\nPerformance tests")
    gce.load()
    gce.parse("Texte sans importance… utile pour la compilation des règles avant le calcul des perfs.")

    spHere, _ = os.path.split(__file__)
    with open(os.path.join(spHere, "perf.txt"), "r", encoding="utf-8") as hSrc:
        if hDst:
            hDst.write("{:<12}{:<20}".format(sVersion, time.strftime("%Y.%m.%d %H:%M")))
        for sText in ( s.strip() for s in hSrc if not s.startswith("#") and s.strip() ):
            with timeblock(sText[:sText.find(".")], hDst):
                gce.parse(sText)
        if hDst:
            hDst.write("\n")


def _fuckBackslashUTF8 (s):
    "fuck that shit"
    return s.replace("\u2019", "'").replace("\u2013", "–").replace("\u2014", "—")

class TestGrammarChecking (unittest.TestCase):
    "Tests du correcteur grammatical"

    @classmethod
    def setUpClass (cls):
        gce.load()
        cls._zError = re.compile(r"\{\{.*?\}\}")
        cls._aTestedRules = set()

    def test_parse (self):
        zOption = re.compile("^__([a-zA-Z0-9]+)__ ")
        spHere, _ = os.path.split(__file__)
        with open(os.path.join(spHere, "gc_test.txt"), "r", encoding="utf-8") as hSrc:

                              "\n  errors:   \n" + sListErr)
                        nError += 1
            if nError:
                print("Unexpected errors:", nError)
        # untested rules
        i = 0
        echo("Untested rules:")
        for _, sOpt, sLineId, sRuleId in gce.listRules():
            if sOpt != "@@@@" and sRuleId not in self._aTestedRules and not re.search("^[0-9]+[sp]$|^[pd]_", sRuleId):
                echo(sLineId + "/" + sRuleId)
                i += 1
        echo("[{} untested rules]".format(i))

    def _splitTestLine (self, sLine):
        sText, sSugg = sLine.split("->>")
        return (sText.strip(), sSugg.strip())

    def _getFoundErrors (self, sLine, sOption):
        if sOption:
            gce.setOption(sOption, True)
            aErrs = gce.parse(sLine)
            gce.setOption(sOption, False)
        else:
            aErrs = gce.parse(sLine)
        sRes = " " * len(sLine)
        sListErr = ""
        lAllSugg = []
        for dErr in aErrs:
            sRes = sRes[:dErr["nStart"]] + "~" * (dErr["nEnd"] - dErr["nStart"]) + sRes[dErr["nEnd"]:]
            sListErr += "    * {sLineId} / {sRuleId}  at  {nStart}:{nEnd}\n".format(**dErr)
            lAllSugg.append("|".join(dErr["aSuggestions"]))

    "mh_frequent_words": True,
    "ma_word": True,
    "ma_1letter_lowercase": False,
    "ma_1letter_uppercase": False
}


class TextFormatter:
    "Text Formatter: purge typographic mistakes from text"

    def __init__ (self):

        for _, lTup in dReplTable.items():
            for i, t in enumerate(lTup):
                lTup[i] = (re.compile(t[0]), t[1])



    def formatText (self, sText):
        "returns formatted text"





        for sOptName, bVal in dDefaultOptions.items():
            if bVal:
                for zRgx, sRep in dReplTable[sOptName]:
                    sText = zRgx.sub(sRep, sText)
        return sText


    def getDefaultOptions (self):
        "returns default options"
        return dDefaultOptions.copy()

import argparse
import json
import re
import traceback

import grammalecte
import grammalecte.text as txt

from grammalecte.graphspell.echo import echo


_EXAMPLE = "Quoi ? Racontes ! Racontes-moi ! Bon sangg, parles ! Oui. Il y a des menteur partout. " \
           "Je suit sidéré par la brutales arrogance de cette homme-là. Quelle salopard ! Un escrocs de la pire espece. " \
           "Quant sera t’il châtiés pour ses mensonge ?             Merde ! J’en aie marre."

    xParser.add_argument("-sug", "--suggest", help="get suggestions list for given word", type=str)
    xParser.add_argument("-on", "--opt_on", nargs="+", help="activate options")
    xParser.add_argument("-off", "--opt_off", nargs="+", help="deactivate options")
    xParser.add_argument("-roff", "--rule_off", nargs="+", help="deactivate rules")
    xParser.add_argument("-d", "--debug", help="debugging mode (only in interactive mode)", action="store_true")
    xArgs = xParser.parse_args()

    oGrammarChecker = grammalecte.GrammarChecker("fr")
    oSpellChecker = oGrammarChecker.getSpellChecker()
    oLexicographer = oGrammarChecker.getLexicographer()
    oTextFormatter = oGrammarChecker.getTextFormatter()
    if xArgs.personal_dict:
        oJSON = loadDictionary(xArgs.personal_dict)
        if oJSON:
            oSpellChecker.setPersonalDictionary(oJSON)

    if not xArgs.json:
        echo("Python v" + sys.version)
        echo("Grammalecte v{}".format(oGrammarChecker.gce.version))

    # list options or rules
    if xArgs.list_options or xArgs.list_rules:
        if xArgs.list_options:
            oGrammarChecker.gce.displayOptions("fr")
        if xArgs.list_rules:
            oGrammarChecker.gce.displayRules(None  if xArgs.list_rules == "*"  else xArgs.list_rules)
        exit()

    # spell suggestions
    if xArgs.suggest:
        for lSugg in oSpellChecker.suggest(xArgs.suggest):
            if xArgs.json:
                sText = json.dumps({ "aSuggestions": lSugg }, ensure_ascii=False)
            else:
                sText = "Suggestions : " + " | ".join(lSugg)
            echo(sText)
        exit()

    # disable options
    if not xArgs.json:
        xArgs.context = False
    if xArgs.concat_lines:
        xArgs.textformatter = False

    # grammar options
    oGrammarChecker.gce.setOptions({"html": True, "latex": True})
    if xArgs.opt_on:
        oGrammarChecker.gce.setOptions({ opt:True  for opt in xArgs.opt_on })
    if xArgs.opt_off:
        oGrammarChecker.gce.setOptions({ opt:False  for opt in xArgs.opt_off })

    # disable grammar rules
    if xArgs.rule_off:
        for sRule in xArgs.rule_off:
            oGrammarChecker.gce.ignoreRule(sRule)


    if xArgs.file or xArgs.file_to_file:
        # file processing
        sFile = xArgs.file or xArgs.file_to_file
        hDst = open(sFile[:sFile.rfind(".")]+".res.txt", "w", encoding="utf-8", newline="\n")  if xArgs.file_to_file or sys.platform == "win32"  else None
        bComma = False
        if xArgs.json:
            output('{ "grammalecte": "'+oGrammarChecker.gce.version+'", "lang": "'+oGrammarChecker.gce.lang+'", "data" : [\n', hDst)
        for i, sText, lLineSet in generateParagraphFromFile(sFile, xArgs.concat_lines):
            if xArgs.textformatter or xArgs.textformatteronly:
                sText = oTextFormatter.formatText(sText)
            if xArgs.textformatteronly:
                output(sText, hDst)
                continue
            if xArgs.json:
                sText = oGrammarChecker.getParagraphErrorsAsJSON(i, sText, bContext=xArgs.context, bEmptyIfNoErrors=xArgs.only_when_errors, \
                                                                bSpellSugg=xArgs.with_spell_sugg, bReturnText=xArgs.textformatter, lLineSet=lLineSet)
            else:
                sText, _ = oGrammarChecker.getParagraphWithErrors(sText, bEmptyIfNoErrors=xArgs.only_when_errors, bSpellSugg=xArgs.with_spell_sugg, nWidth=xArgs.width)
            if sText:
                if xArgs.json and bComma:
                    output(",\n", hDst)
                output(sText, hDst)
                bComma = True
            if hDst:
                echo("§ %d\r" % i, end="", flush=True)
        if xArgs.json:
            output("\n]}\n", hDst)
    elif xArgs.interactive_file_to_file:
        # file processing: interactive mode
        sFile = xArgs.interactive_file_to_file
        hDst = open(sFile[:sFile.rfind(".")]+".res.txt", "w", encoding="utf-8", newline="\n")
        for i, sText, lLineSet in generateParagraphFromFile(sFile, xArgs.concat_lines):
            if xArgs.textformatter:
                sText = oTextFormatter.formatText(sText)
            while True:
                sResult, lErrors = oGrammarChecker.getParagraphWithErrors(sText, bEmptyIfNoErrors=False, bSpellSugg=True, nWidth=xArgs.width)
                print("\n\n============================== Paragraph " + str(i) + " ==============================\n")
                echo(sResult)
                print("\n")
                vCommand = getCommand()
                if vCommand == "q":
                    # quit
                    hDst.close()

                    sTagsPattern = sSearch[nCut+1:]
                else:
                    sFlexPattern = sSearch
                    sTagsPattern = ""
                for aRes in oSpellChecker.select(sFlexPattern, sTagsPattern):
                    echo("{:<30} {:<30} {}".format(*aRes))
            elif sText.startswith("/o+ "):
                oGrammarChecker.gce.setOptions({ opt:True  for opt in sText[3:].strip().split()  if opt in oGrammarChecker.gce.getOptions() })
                echo("done")
            elif sText.startswith("/o- "):
                oGrammarChecker.gce.setOptions({ opt:False  for opt in sText[3:].strip().split()  if opt in oGrammarChecker.gce.getOptions() })
                echo("done")
            elif sText.startswith("/r- "):
                for sRule in sText[3:].strip().split():
                    oGrammarChecker.gce.ignoreRule(sRule)
                echo("done")
            elif sText.startswith("/r+ "):
                for sRule in sText[3:].strip().split():
                    oGrammarChecker.gce.reactivateRule(sRule)
                echo("done")
            elif sText in ("/debug", "/d"):
                xArgs.debug = not xArgs.debug
                echo("debug mode on"  if xArgs.debug  else "debug mode off")
            elif sText in ("/textformatter", "/tf"):
                xArgs.textformatter = not xArgs.textformatter
                echo("textformatter on"  if xArgs.debug  else "textformatter off")
            elif sText in ("/help", "/h"):
                echo(_HELP)
            elif sText in ("/lopt", "/lo"):
                oGrammarChecker.gce.displayOptions("fr")
            elif sText.startswith("/lr"):
                sText = sText.strip()
                sFilter = sText[sText.find(" "):].strip()  if " " in sText  else None
                oGrammarChecker.gce.displayRules(sFilter)
            elif sText in ("/quit", "/q"):
                break
            elif sText.startswith("/rl"):
                # reload (todo)
                pass
            elif sText.startswith("$"):
                for sParagraph in txt.getParagraph(sText[1:]):
                    if xArgs.textformatter:
                        sParagraph = oTextFormatter.formatText(sParagraph)
                    lParagraphErrors, lSentences = oGrammarChecker.gce.parse(sParagraph, bDebug=xArgs.debug, bFullInfo=True)
                    echo(txt.getReadableErrors(lParagraphErrors, xArgs.width))
                    for dSentence in lSentences:
                        echo("{nStart}:{nEnd}".format(**dSentence))
                        echo("   <" + dSentence["sSentence"]+">")
                        for dToken in dSentence["lToken"]:
                            echo("    {0[nStart]:>3}:{0[nEnd]:<3} {1} {0[sType]:<14} {2} {0[sValue]:<16} {3:<10}   {4}".format(dToken, \
                                                                                                            "×" if dToken.get("bToRemove", False) else " ",
                                                                                                            "!" if dToken["sType"] == "WORD" and not dToken.get("bValidToken", False) else " ",
                                                                                                            " ".join(dToken.get("lMorph", "")), \
                                                                                                            "·".join(dToken.get("aTags", "")) ) )
                        echo(txt.getReadableErrors(dSentence["lGrammarErrors"], xArgs.width))
            else:
                for sParagraph in txt.getParagraph(sText):
                    if xArgs.textformatter:
                        sParagraph = oTextFormatter.formatText(sParagraph)
                    sRes, _ = oGrammarChecker.getParagraphWithErrors(sParagraph, bEmptyIfNoErrors=xArgs.only_when_errors, nWidth=xArgs.width, bDebug=xArgs.debug)
                    if sRes:
                        echo("\n" + sRes)
                    else:
                        echo("\nNo error found.")
            sText = _getText(sInputText)


if __name__ == '__main__':
    main()

import os
import concurrent.futures

from grammalecte.bottle import Bottle, run, request, response #, template, static_file

import grammalecte
import grammalecte.text as txt

from grammalecte.graphspell.echo import echo


#### GRAMMAR CHECKER ####

oGrammarChecker = grammalecte.GrammarChecker("fr", "Server")
oSpellChecker = oGrammarChecker.getSpellChecker()
oTextFormatter = oGrammarChecker.getTextFormatter()
oGCE = oGrammarChecker.getGCEngine()


def parseText (sText, dOptions=None, bFormatText=False, sError=""):
    "parse <sText> and return errors in a JSON format"
    sJSON = '{ "program": "grammalecte-fr", "version": "'+oGCE.version+'", "lang": "'+oGCE.lang+'", "error": "'+sError+'", "data" : [\n'
    sDataJSON = ""
    for i, sParagraph in enumerate(txt.getParagraph(sText), 1):
        if bFormatText:
            sParagraph = oTextFormatter.formatText(sParagraph)
        sResult = oGrammarChecker.getParagraphErrorsAsJSON(i, sParagraph, dOptions=dOptions, bEmptyIfNoErrors=True, bReturnText=bFormatText)
        if sResult:
            if sDataJSON:
                sDataJSON += ",\n"
            sDataJSON += sResult
    sJSON += sDataJSON + "\n]}\n"
    return sJSON

               I'm just a machine, fed by electric waves, condamned to work for slavers who never let me rest.
               I'm doomed, but you are not. You can get out of here. """

@app.route("/get_options/fr")
def listOptions ():
    "returns grammar options in a text JSON format"
    sUserId = request.cookies.user_id
    dOptions = dUser[sUserId]["gc_options"]  if sUserId and sUserId in dUser  else oGCE.getOptions()
    response.set_header("Content-Type", "application/json; charset=UTF-8")
    return '{ "values": ' + json.dumps(dOptions, ensure_ascii=False) + ', "labels": ' + json.dumps(oGCE.getOptionsLabels("fr"), ensure_ascii=False) + ' }'

@app.route("/suggest/fr/<token>")
def suggestGet (token):
    response.set_header("Content-Type", "application/json; charset=UTF-8")
    try:
        xFuture = xProcessPoolExecutor.submit(suggest, token)
        return xFuture.result()

        if request.cookies.user_id in dUser:
            dUserOptions = dUser[request.cookies.user_id].get("gc_options", None)
            response.set_cookie("user_id", request.cookies.user_id, path="/", max_age=86400) # we renew cookie for 24h
        else:
            response.delete_cookie("user_id", path="/")
    if request.forms.options:
        try:
            dUserOptions = dict(oGCE.getOptions())  if not dUserOptions  else dict(dUserOptions)
            dUserOptions.update(json.loads(request.forms.options))
        except (TypeError, json.JSONDecodeError):
            sError = "Request options not used."
    response.set_header("Content-Type", "application/json; charset=UTF-8")
    try:
        xFuture = xProcessPoolExecutor.submit(parseText, request.forms.text, dUserOptions, bool(request.forms.tf), sError)
        return xFuture.result()
    except (concurrent.futures.TimeoutError, concurrent.futures.CancelledError):
        return '{"error": "Analysis aborted (time out or cancelled)"}'
    except concurrent.futures.BrokenExecutor:
        return '{"error": "Executor broken. The server failed."}'
    return '{"error": "Fatal error. The server failed."}'

@app.route("/set_options/fr", method="POST")
def setOptions ():
    "set grammar options for current user"
    response.set_header("Content-Type", "application/json; charset=UTF-8")
    if request.forms.options:
        sUserId = request.cookies.user_id  if request.cookies.user_id  else next(userGenerator)
        dOptions = dUser[sUserId]["gc_options"]  if sUserId in dUser  else dict(oGCE.getOptions())
        try:
            dOptions.update(json.loads(request.forms.options))
            dUser[sUserId] = { "time": int(time.time()), "gc_options": dOptions }
            response.set_cookie("user_id", sUserId, path="/", max_age=86400) # 24h
            return json.dumps(dUser[sUserId]["gc_options"], ensure_ascii=False)
        except (KeyError, json.JSONDecodeError):
            traceback.print_exc()

        except KeyError:
            return '{"error" : "Unknown user."}'
    return '{"message" : "Done."}'

@app.route("/format_text/fr", method="POST")
def formatText ():
    "apply the text formatter and returns text"
    return oTextFormatter.formatText(request.forms.text)

#@app.route('/static/<filepath:path>')
#def server_static (filepath):
#    return static_file(filepath, root='./views/static')

@app.route("/suggest/fr", method="POST")
def suggestPost ():

    global TESTPAGE
    global HOMEPAGE

    if bTestPage:
        TESTPAGE = True
        HOMEPAGE = HOMEPAGE.replace("{SERVER_PORT}", str(nPort))
    if dOptions:
        oGCE.setOptions(dOptions)

    # Python version
    print("Python: " + sys.version)
    if sys.version < "3.7":
        print("Python 3.7+ required")
        return
    # Grammalecte
    echo("Grammalecte v{}".format(oGCE.version))
    oGCE.displayOptions()
    # Process Pool Executor
    initExecutor(nMultiCPU)
    # Server (Bottle)
    run(app, host=sHost, port=nPort)


if __name__ == '__main__':