Grammalecte  Check-in [cca3887aad]

    for m in _zEndOfSentence.finditer(sText):
        yield (iStart, m.end())
        iStart = m.end()


def _proofread (oSentence, s, sx, nOffset, bParagraph, dDA, dPriority, sCountry, dOptions, bShowRuleId, bDebug, bContext):
    dErrs = {}
    bParagraphChange = False
    bSentenceChange = False
    for sOption, lRuleGroup in _getRules(bParagraph):
        if sOption == "@@@@":
            # graph rules
            if not bParagraph and bSentenceChange:
                oSentence.update(s)
                bSentenceChange = False
            for sGraphName, sLineId in lRuleGroup:
                if bDebug:
                    print(sGraphName, sLineId)
                bParagraphChange, errs = oSentence.parse(dAllGraph[sGraphName], dPriority, sCountry, dOptions, bShowRuleId, bDebug, bContext)
                dErrs.update(errs)
                if bParagraphChange:
                    s = oSentence.rewrite()
                    if bDebug:
                        print("~", oSentence.sSentence)
        elif not sOption or dOptions.get(sOption, False):
            # regex rules
            for zRegex, bUppercase, sLineId, sRuleId, nPriority, lActions in lRuleGroup:
                if sRuleId not in _aIgnoredRules:
                    for m in zRegex.finditer(s):

                                        nErrorStart = nOffset + m.start(eAct[0])
                                        if nErrorStart not in dErrs or nPriority > dPriority[nErrorStart]:
                                            dErrs[nErrorStart] = _createRegexError(s, sx, sWhat, nOffset, m, eAct[0], sLineId, sRuleId, bUppercase, eAct[1], eAct[2], bShowRuleId, sOption, bContext)
                                            dPriority[nErrorStart] = nPriority
                                    elif cActionType == "~":
                                        # text processor
                                        s = _rewrite(s, sWhat, eAct[0], m, bUppercase)
                                        bParagraphChange = True
                                        bSentenceChange = True
                                        if bDebug:
                                            echo("~ " + s + "  -- " + m.group(eAct[0]) + "  # " + sLineId)
                                    elif cActionType == "=":
                                        # disambiguation
                                        globals()[sWhat](s, m, dDA)
                                        if bDebug:
                                            echo("= " + m.group(0) + "  # " + sLineId + "\nDA: " + str(dDA))
                                    elif cActionType == ">":
                                        # we do nothing, this test is just a condition to apply all following actions
                                        pass
                                    else:
                                        echo("# error: unknown action at " + sLineId)
                                elif cActionType == ">":
                                    break
                            except Exception as e:
                                raise Exception(str(e), "# " + sLineId + " # " + sRuleId)
    if bParagraphChange:
        return (s, dErrs)
    return (False, dErrs)


def _createRegexWriterError (s, sx, sRepl, nOffset, m, iGroup, sLineId, sRuleId, bUppercase, sMsg, sURL, bShowRuleId, sOption, bContext):
    "error for Writer (LO/OO)"
    xErr = SingleProofreadingError()

    def __init__ (self, sSentence, sSentence0, nOffset):
        self.sSentence = sSentence
        self.sSentence0 = sSentence0
        self.nOffset = nOffset
        self.lToken = list(_oTokenizer.genTokens(sSentence, True))
        self.createError = self._createWriterError  if _bWriterError  else self._createDictError

    def update (self, sSentence):
        self.sSentence = sSentence
        self.lToken = list(_oTokenizer.genTokens(sSentence, True))

    def _getNextMatchingNodes (self, dToken, dGraph, dNode):
        "generator: return nodes where <dToken> “values” match <dNode> arcs"
        # token value
        if dToken["sValue"] in dNode:
            #print("value found: ", dToken["sValue"])
            yield dGraph[dNode[dToken["sValue"]]]
        # token lemmas

                    dToken["sValue"] = dToken["sNewValue"]
                    nDiffLen = len(dToken["sRealValue"]) - len(dToken["sNewValue"])
                    sNewRepl = (dToken["sNewValue"] + " " * nDiffLen)  if nDiffLen >= 0  else dToken["sNewValue"][:len(dToken["sRealValue"])]
                    self.sSentence = self.sSentence[:self.nOffset+dToken["nStart"]] + sNewRepl + self.sSentence[self.nOffset+dToken["nEnd"]:]
                    del dToken["sNewValue"]
        self.lToken.clear()
        self.lToken = lNewToken
        return self.sSentence



#### Analyse tokens

def g_morph (dToken, sPattern, sNegPattern=""):
    "analyse a token, return True if <sNegPattern> not in morphologies and <sPattern> in morphologies"

    def __init__ (self, sLang):
        self.sLang = sLang
        if sLang not in _PATTERNS:
            self.sLang = "default"
        self.zToken = re.compile( "(?i)" + '|'.join(sRegex for sRegex in _PATTERNS[sLang]) )

    def genTokens (self, sText, bStartEndToken=False):
        i = 0
        if bStartEndToken:
            yield { "i": 0, "sType": "INFO", "sValue": "<start>", "nStart": 0, "nEnd": 0 }
        for i, m in enumerate(self.zToken.finditer(sText), 1):
            yield { "i": i, "sType": m.lastgroup, "sValue": m.group(), "nStart": m.start(), "nEnd": m.end() }
        if bStartEndToken:
            iEnd = len(sText)
            yield { "i": i+1, "sType": "INFO", "sValue": "<end>", "nStart": iEnd, "nEnd": iEnd }