597
598
599
600
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
|
597
598
599
600
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
|
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
-
+
|
self.lToken = list(_oTokenizer.genTokens(sSentence, True))
def _getNextMatchingNodes (self, dToken, dGraph, 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"])
print(" MATCH:", dToken["sValue"])
yield 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)
print(" MATCH:", sValue)
yield dGraph[dNode[sValue]]
elif dToken["sValue"].isupper():
sValue = dToken["sValue"].lower()
if sValue in dNode:
if bDebug:
print("MATCH:", sValue)
print(" MATCH:", sValue)
yield dGraph[dNode[sValue]]
sValue = dToken["sValue"].capitalize()
if sValue in dNode:
if bDebug:
print("MATCH:", sValue)
print(" MATCH:", sValue)
yield dGraph[dNode[sValue]]
# token lemmas
if "<lemmas>" in dNode:
for sLemma in _oSpellChecker.getLemma(dToken["sValue"]):
if sLemma in dNode["<lemmas>"]:
if bDebug:
print("MATCH: >" + sLemma)
print(" MATCH: >" + sLemma)
yield dGraph[dNode["<lemmas>"][sLemma]]
# universal arc
if "*" in dNode:
if bDebug:
print("MATCH: *")
print(" MATCH: *")
yield dGraph[dNode["*"]]
# regex value arcs
if "<re_value>" in dNode:
for sRegex in dNode["<re_value>"]:
if re.search(sRegex, dToken["sValue"]):
if bDebug:
print("MATCH: ~" + sRegex)
print(" MATCH: ~" + sRegex)
yield dGraph[dNode["<re_value>"][sRegex]]
# regex morph arcs
if "<re_morph>" in dNode:
for sRegex in dNode["<re_morph>"]:
if "¬" not in sRegex:
# no anti-pattern
if any(re.search(sRegex, sMorph) for sMorph in _oSpellChecker.getMorph(dToken["sValue"])):
if bDebug:
print("MATCH: @" + sRegex)
print(" MATCH: @" + sRegex)
yield 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 all(re.search(sPattern, sMorph) for sMorph in _oSpellChecker.getMorph(dToken["sValue"])):
if bDebug:
print("MATCH: @" + sRegex)
print(" MATCH: @" + sRegex)
yield dGraph[dNode["<re_morph>"][sRegex]]
else:
if sNegPattern and any(re.search(sNegPattern, sMorph) for sMorph in _oSpellChecker.getMorph(dToken["sValue"])):
continue
if any(re.search(sPattern, sMorph) for sMorph in _oSpellChecker.getMorph(dToken["sValue"])):
if bDebug:
print("MATCH: @" + sRegex)
print(" MATCH: @" + sRegex)
yield dGraph[dNode["<re_morph>"][sRegex]]
def parse (self, dGraph, dPriority, sCountry="${country_default}", dOptions=None, bShowRuleId=False, bDebug=False, bContext=False):
dErr = {}
dPriority = {} # Key = position; value = priority
dOpt = _dOptions if not dOptions else dOptions
lPointer = []
|