1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
|
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
|
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
|
zNegPattern = re.compile(sNegPattern)
if any(zNegPattern.search(sMorph) for sMorph in lMorph):
return False
# search sPattern
zPattern = re.compile(sPattern)
return any(zPattern.search(sMorph) for sMorph in lMorph)
def g_merged_analyse (dToken1, dToken2, cMerger, sPattern, sNegPattern="", bSetMorph=True):
"merge two token values, return True if <sNegPattern> not in morphologies and <sPattern> in morphologies (disambiguation off)"
lMorph = _oSpellChecker.getMorph(dToken1["sValue"] + cMerger + dToken2["sValue"])
if not lMorph:
return False
# check negative condition
if sNegPattern:
if sNegPattern == "*":
# all morph must match sPattern
if not lMorph:
return False
zPattern = re.compile(sPattern)
bResult = all(zPattern.search(sMorph) for sMorph in lMorph)
if bResult and bSetMorph:
dToken1["lMorph"] = lMorph
return bResult
else:
zNegPattern = re.compile(sNegPattern)
if any(zNegPattern.search(sMorph) for sMorph in lMorph):
return False
# search sPattern
zPattern = re.compile(sPattern)
bResult = any(zPattern.search(sMorph) for sMorph in lMorph)
if bResult and bSetMorph:
dToken1["lMorph"] = lMorph
return bResult
def g_tag_before (dToken, dTags, sTag):
if sTag not in dTags:
return False
if dToken["i"] > dTags[sTag][0]:
return True
return False
|