601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
|
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
|
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
-
+
-
-
+
|
dToken["lMorph"] = self.dTokenPos[dToken["nStart"]]["lMorph"]
self.lToken = lNewToken
self.dTokenPos = { dToken["nStart"]: dToken for dToken in self.lToken if dToken["sType"] != "INFO" }
if bDebug:
print("UPDATE:")
print(self)
def _getNextMatchingNodes (self, dToken, dGraph, dNode, bDebug=False):
def _getNextMatchingNodes (self, dToken, dGraph, iNode1, dNode, bDebug=False):
"generator: return nodes where <dToken> “values” match <dNode> arcs"
# token value
if dToken["sValue"] in dNode:
if bDebug:
print(" MATCH:", dToken["sValue"])
yield dGraph[dNode[dToken["sValue"]]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode[dToken["sValue"]]] }
if dToken["sValue"][0:2].istitle(): # we test only 2 first chars, to make valid words such as "Laissez-les", "Passe-partout".
sValue = dToken["sValue"].lower()
if sValue in dNode:
if bDebug:
print(" MATCH:", sValue)
yield dGraph[dNode[sValue]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode[sValue]] }
elif dToken["sValue"].isupper():
sValue = dToken["sValue"].lower()
if sValue in dNode:
if bDebug:
print(" MATCH:", sValue)
yield dGraph[dNode[sValue]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode[sValue]] }
sValue = dToken["sValue"].capitalize()
if sValue in dNode:
if bDebug:
print(" MATCH:", sValue)
yield dGraph[dNode[sValue]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode[sValue]] }
# regex value arcs
if "<re_value>" in dNode:
for sRegex in dNode["<re_value>"]:
if "¬" not in sRegex:
# no anti-pattern
if re.search(sRegex, dToken["sValue"]):
if bDebug:
print(" MATCH: ~" + sRegex)
yield dGraph[dNode["<re_value>"][sRegex]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<re_value>"][sRegex]] }
else:
# there is an anti-pattern
sPattern, sNegPattern = sRegex.split("¬", 1)
if sNegPattern and re.search(sNegPattern, dToken["sValue"]):
continue
if not sPattern or re.search(sPattern, dToken["sValue"]):
if bDebug:
print(" MATCH: ~" + sRegex)
yield dGraph[dNode["<re_value>"][sRegex]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<re_value>"][sRegex]] }
# analysable tokens
if dToken["sType"][0:4] == "WORD":
# token lemmas
if "<lemmas>" in dNode:
for sLemma in _oSpellChecker.getLemma(dToken["sValue"]):
if sLemma in dNode["<lemmas>"]:
if bDebug:
print(" MATCH: >" + sLemma)
yield dGraph[dNode["<lemmas>"][sLemma]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<lemmas>"][sLemma]] }
# regex morph arcs
if "<re_morph>" in dNode:
for sRegex in dNode["<re_morph>"]:
if "¬" not in sRegex:
# no anti-pattern
lMorph = dToken.get("lMorph", _oSpellChecker.getMorph(dToken["sValue"]))
if any(re.search(sRegex, sMorph) for sMorph in lMorph):
if bDebug:
print(" MATCH: @" + sRegex)
yield dGraph[dNode["<re_morph>"][sRegex]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<re_morph>"][sRegex]] }
else:
# there is an anti-pattern
sPattern, sNegPattern = sRegex.split("¬", 1)
if sNegPattern == "*":
# all morphologies must match with <sPattern>
if sPattern:
lMorph = dToken.get("lMorph", _oSpellChecker.getMorph(dToken["sValue"]))
if lMorph and all(re.search(sPattern, sMorph) for sMorph in lMorph):
if bDebug:
print(" MATCH: @" + sRegex)
yield dGraph[dNode["<re_morph>"][sRegex]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<re_morph>"][sRegex]] }
else:
lMorph = dToken.get("lMorph", _oSpellChecker.getMorph(dToken["sValue"]))
if sNegPattern and any(re.search(sNegPattern, sMorph) for sMorph in lMorph):
continue
if not sPattern or any(re.search(sPattern, sMorph) for sMorph in lMorph):
if bDebug:
print(" MATCH: @" + sRegex)
yield dGraph[dNode["<re_morph>"][sRegex]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<re_morph>"][sRegex]] }
# token tags
if "tags" in dToken and "<tags>" in dNode:
for sTag in dToken["tags"]:
if sTag in dNode["<tags>"]:
if bDebug:
print(" MATCH: /" + sTag)
yield dGraph[dNode["<tags>"][sTag]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<tags>"][sTag]] }
# meta arc (for token type)
if "<meta>" in dNode:
for sMeta in dNode["<meta>"]:
# not regex here, we just search if <dNode["sType"]> exists within <sMeta>
# no regex here, we just search if <dNode["sType"]> exists within <sMeta>
if sMeta == "*":
if bDebug:
print(" MATCH: *" + sMeta)
yield dGraph[dNode["<meta>"]["*"]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<meta>"]["*"]] }
elif "¬" in sMeta:
if dToken["sType"] not in sMeta:
if bDebug:
print(" MATCH: *" + sMeta)
yield dGraph[dNode["<meta>"][sMeta]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<meta>"][sMeta]] }
elif dToken["sType"] in sMeta:
if bDebug:
print(" MATCH: *" + sMeta)
yield dGraph[dNode["<meta>"][sMeta]]
yield { "iNode1": iNode1, "dNode": dGraph[dNode["<meta>"][sMeta]] }
def parse (self, dGraph, dPriority, sCountry="${country_default}", dOptions=None, bShowRuleId=False, bDebug=False, bContext=False):
"parse tokens from the text and execute actions encountered"
dOpt = _dOptions if not dOptions else dOptions
lPointer = []
bTagAndRewrite = False
for i, dToken in enumerate(self.lToken):
for iToken, dToken in enumerate(self.lToken):
if bDebug:
print("TOKEN:", dToken["sValue"])
# check arcs for each existing pointer
lNextPointer = []
for dPointer in lPointer:
for dNode in self._getNextMatchingNodes(dToken, dGraph, dPointer["dNode"], bDebug):
lNextPointer.extend(self._getNextPointers(dToken, dGraph, dPointer["iNode1"], dPointer["dNode"], bDebug))
lNextPointer.append({"iToken": dPointer["iToken"], "dNode": dNode})
for dNode in self._getNextMatchingNodes(dToken, dGraph, dGraph[0], bDebug):
lPointer.extend(self._getNextPointers(dToken, dGraph, iToken, dGraph[0], bDebug))
lPointer.append({"iToken": i, "dNode": dNode})
bChange = self._executeActions(dGraph, dPointer["dNode"]["<rules>"], dPointer["iToken"]-1, i, dPriority, dOpt, sCountry, bShowRuleId, bDebug, bContext)
bChange = self._executeActions(dGraph, dPointer["dNode"]["<rules>"], dPointer["iNode1"]-1, iToken, dPriority, dOpt, sCountry, bShowRuleId, bDebug, bContext)
if bChange:
bTagAndRewrite = True
if bTagAndRewrite:
self.rewrite(bDebug)
if bDebug:
print(self)
return (bTagAndRewrite, self.sSentence)
|