Grammalecte  Diff

    def __init__ (self, sText):
        self.sText = sText
        self.sText0 = sText
        self.sSentence = ""
        self.sSentence0 = ""
        self.nOffsetWithinParagraph = 0
        self.lTokens = []

        self.dTokenPos = {}         # {position: token}
        self.dTags = {}             # {position: tags}
        self.dError = {}            # {position: error}
        self.dSentenceError = {}    # {position: error} (for the current sentence only)
        self.dErrorPriority = {}    # {position: priority of the current error}

    def __str__ (self):

        dOpt = dOptions or gc_options.dOptions
        bShowRuleId = gc_options.dOptions.get('idrule', False)
        # parse paragraph
        try:
            self.parseText(self.sText, self.sText0, True, 0, sCountry, dOpt, bShowRuleId, bDebug, bContext)
        except:
            raise
        self.lTokens = None
        self.lTokens0 = None
        if bFullInfo:
            lParagraphErrors = list(self.dError.values())
            lSentences = []
            self.dSentenceError.clear()
        # parse sentences
        sText = self._getCleanText()
        for iStart, iEnd in text.getSentenceBoundaries(sText):

                        })
                        self.dSentenceError.clear()
                except:
                    raise
        if bFullInfo:
            # Grammar checking and sentence analysis
            return lParagraphErrors, lSentences
        else:
            # Grammar checking only
            return self.dError.values() # this is a view (iterable)

    def _getCleanText (self):
        sText = self.sText
        if " " in sText:
            sText = sText.replace(" ", ' ') # nbsp
        if " " in sText:
            sText = sText.replace(" ", ' ') # nnbsp

                if bUppercase:
                    sWhat = sWhat[0:1].upper() + sWhat[1:]
                self.lTokens[nTokenRewriteStart]["sNewValue"] = sWhat
            else:
                # several tokens
                lTokenValue = sWhat.split("|")
                if len(lTokenValue) != (nTokenRewriteEnd - nTokenRewriteStart + 1):
                    if (bDebug):
                        echo("Error. Text processor: number of replacements != number of tokens.")
                    return
                for i, sValue in zip(range(nTokenRewriteStart, nTokenRewriteEnd+1), lTokenValue):
                    if not sValue or sValue == "*":
                        self.lTokens[i]["bToRemove"] = True
                    else:
                        if bUppercase:
                            sValue = sValue[0:1].upper() + sValue[1:]
                        self.lTokens[i]["sNewValue"] = sValue

    def rewriteFromTags (self, bDebug=False):
        "rewrite the sentence, modify tokens, purge the token list"
        if bDebug:
            echo("REWRITE")
        lNewTokens = []
        lNewTokens0 = []
        nMergeUntil = 0
        dTokenMerger = {}
        for iToken, dToken in enumerate(self.lTokens):
            bKeepToken = True
            if dToken["sType"] != "INFO":
                if nMergeUntil and iToken <= nMergeUntil:
                    # token to merge