250
251
252
253
254
255
256
257
258
259
260
261
262
263
|
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
|
+
+
|
s += "now: " + self.sSentence + "\n"
for dToken in self.lTokens:
s += '#{i}\t{nStart}:{nEnd}\t{sValue}\t{sType}'.format(**dToken)
if "lMorph" in dToken:
s += "\t" + str(dToken["lMorph"])
if "aTags" in dToken:
s += "\t" + str(dToken["aTags"])
if "nMultiStartTo" in dToken:
s += "\t>>" + str(dToken["nMultiStartTo"])
s += "\n"
#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"
|
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
|
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
|
-
+
|
nMultiEnd = -1 if cNodeType != "&" else dToken["nMultiStartTo"]
lPointers.append({ "iToken1": iToken, "iNode": iNode, "nMultiEnd": nMultiEnd })
# check if there is rules to check for each pointer
for dPointer in lPointers:
if dPointer["nMultiEnd"] != -1:
if dToken["i"] < dPointer["nMultiEnd"]:
continue
if dToken["i"] == dPointer["nMultiEnd"]:
if dToken["i"] >= dPointer["nMultiEnd"]:
dPointer["nMultiEnd"] = -1
if "<rules>" in dGraph[dPointer["iNode"]]:
bChange = self._executeActions(dGraph, dGraph[dPointer["iNode"]]["<rules>"], dPointer["iToken1"]-1, iToken, dOptions, sCountry, bShowRuleId, bDebug, bContext)
if bChange:
bTagAndRewrite = True
if bTagAndRewrite:
self.rewriteFromTags(bDebug)
|
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
|
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
|
-
-
+
+
-
+
|
nTokenEnd = nTokenOffset + eAct[1] if eAct[1] > 0 else nLastToken + eAct[1]
dMultiToken = {
"nTokenStart": nTokenStart,
"nTokenEnd": nTokenEnd,
"lTokens": self.lTokens[nTokenStart:nTokenEnd+1],
"lMorph": sAction.split("|") if sAction else [":HM"]
}
self.lTokens[nTokenStart]["nMultiStartTo"] = nTokenEnd
self.lTokens[nTokenEnd]["nMultiEndFrom"] = nTokenStart
self.lTokens[nTokenStart]["nMultiStartTo"] = self.lTokens[nTokenEnd]["i"]
self.lTokens[nTokenEnd]["nMultiEndFrom"] = self.lTokens[nTokenStart]["i"]
self.lTokens[nTokenStart]["dMultiToken"] = dMultiToken
self.lTokens[nTokenEnd]["dMultiToken"] = dMultiToken
if bDebug:
echo(" MULTI-TOKEN: ({}) [{}:{}]".format(sAction, self.lTokens[nTokenOffset+1]["sValue"], self.lTokens[nLastToken]["sValue"]))
echo(" MULTI-TOKEN: ({}) [{}:{}]".format(sAction, self.lTokens[nTokenStart]["sValue"], self.lTokens[nTokenEnd]["sValue"]))
#print(dMultiToken)
else:
echo("# error: unknown action at " + sLineId)
elif cActionType == ">":
if bDebug:
echo(" COND_BREAK")
break
|