self.static_dics.setdefault(ls[0].decode('utf-8'),[]).extend( ls[1].split(",") )
def __getitem__(self,tokens):
if len(tokens)==1:
return self.static_dics.get(tokens[0],[])
else:
return []
def __setitem__(self,tok,cats):
self.static_dics[tok] = cats
def has_key(self,tok):
return (tok in self.static_dics)
def get(self,toklist,defval):
ret = self.__getitem__(toklist)
return ret
parser = CCGParser()
parser.combinators = [LApp,RApp,LB,RB,Conj,RT("NP[sbj]"),LBx]
parser.terminators = ["ROOT","S","S[wq]","S[q]","S[imp]"]
parser.lexicon = Lexicon()
parser.concatenator = ""
def tokenize(s):
if len(s)==0:
return s
elif s[-1]==".":
tokens = s[:-1].split()
tokens.append( s[-1] )
return tokens
else:
return s.split()
return self.static_dics.get(tokens[0], [])
else:
return []
def __setitem__(self, tok, cats):
self.static_dics[tok] = cats
def has_key(self, tok):
return (tok in self.static_dics)
def get(self, toklist, defval):
ret = self.__getitem__(toklist)
return ret
parser = CCGParser()
parser.combinators = [LApp, RApp, LB, RB, Conj, RT("NP[sbj]"), LBx]
parser.terminators = ["ROOT", "S", "S[wq]", "S[q]", "S[imp]"]
parser.lexicon = Lexicon()
parser.concatenator = ""
def tokenize(s):
if len(s) == 0:
return s
elif s[-1] == ".":
tokens = s[:-1].split()
tokens.append(s[-1])
return tokens
else:
return s.split()
"(NP/NP)\\NP"
]
lexicon["don't"] = ["(S\\NP)/(S\\NP)"]
return lexicon
#-- special rule for English
def Rel(lt, rt):
if lt != Symbol("NP"):
return None
if rt == [BwdApp, Symbol("S[pss]"), Symbol("NP")]:
return lt
return None
parser = CCGParser()
parser.combinators = [
LApp, RApp, LB, RB,
LT("NP"),
LT("S\\NP"),
RT("NP"), Conj, SkipComma, Rel
]
parser.terminators = ["ROOT", "S", "S[q]", "S[wq]", "S[imp]"]
parser.lexicon = default_lexicon()
parser.concatenator = " "
def run(text, type=0):
for tokens in tokenize(text):
print(u"test run : tokens={0}".format(str(tokens)))
for t in parser.parse(tokens):
def default_lexicon():
ret = {}
ret[u"。"] = ["ROOT\\S", "ROOT\\S[imp]", "ROOT\\S[q]", "ROOT\\S[wq]"]
ret[u"?"] = ["ROOT\\S[q]", "ROOT\\S[wq]"]
for line in open(
os.path.join(os.path.dirname(os.path.abspath(__file__)),
"ccglex.ma")):
line = line.strip()
if len(line) == 0: continue
if line[0] == "#": continue
ls = line.split('\t')
ret.setdefault(ls[0].decode('utf-8'), []).extend(ls[2].split(","))
return ret
parser = CCGParser()
parser.combinators = [LApp, RApp, LB, RB, LBx, Conj, SkipComma, RT("NP")]
parser.terminators = ["ROOT", "S", "S[wq]", "S[q]", "S[imp]"]
parser.lexicon = default_lexicon()
parser.concatenator = ""
if __name__ == "__main__":
def __repr__(s):
if sys.stdout.encoding == 'UTF-8':
return s
else:
return repr(s)
for line in sys.stdin:
line = line.strip()
return lexicon
#-- special rule for English
def Rel(lt,rt):
if lt!=Symbol("NP"):
return None
if rt==[BwdApp , Symbol("S[pss]") , Symbol("NP")]:
return lt
return None
parser = CCGParser()
parser.combinators = [LApp,RApp,LB,RB,LT("NP"),LT("S\\NP"),RT("NP"),Conj,SkipComma,Rel]
parser.terminators = ["ROOT","S","S[q]","S[wq]","S[imp]"]
parser.lexicon = default_lexicon()
parser.concatenator = " "
def run(text,type=0):
for tokens in tokenize(text):
print(u"test run : tokens={0}".format(str(tokens)))
for t in parser.parse(tokens):
if type==0:
for r in t.leaves():
print(u"{0}\t{1}".format(r.token , r.catname))
break
def default_lexicon():
ret = {}
ret[u"。"] = ["ROOT\\S" , "ROOT\\S[imp]" , "ROOT\\S[q]" , "ROOT\\S[wq]"]
ret[u"?"] = ["ROOT\\S[q]" , "ROOT\\S[wq]"]
for line in open(os.path.join(os.path.dirname( os.path.abspath(__file__) ) ,"ccglex.ma")):
line = line.strip()
if len(line)==0:continue
if line[0]=="#":continue
ls = line.split('\t')
ret.setdefault(ls[0].decode('utf-8'),[]).extend( ls[2].split(",") )
return ret
parser = CCGParser()
parser.combinators = [LApp,RApp,LB,RB,LBx,Conj,SkipComma,RT("NP")]
parser.terminators = ["ROOT","S","S[wq]","S[q]","S[imp]"]
parser.lexicon = default_lexicon()
parser.concatenator = ""
if __name__=="__main__":
def __repr__(s):
if sys.stdout.encoding=='UTF-8':
return s
else:
return repr(s)
for line in sys.stdin:
line = line.strip()
line = line.decode('utf-8')
return True
elif term[0].value()=="forall":
return False
else:
assert(len(term)>=2),lt
return (check(term[1]) and check(term[2]))
if type(rt)==list or rt.value()!="COMMA":
return None
elif not check(lt):
return None
return lt
parser = CCGParser()
parser.combinators = [LApp,RApp,LB,RB,Conj,FwdRel,SkipCommaJP,RT("NP[sbj]"),RBx]
parser.terminators = ["ROOT","S","S[exc]","S[imp]","S[null]","S[q]","S[wq]","S[null-q]","S[nom]"]
parser.lexicon = default_lexicon()
parser.concatenator = ""
def run(text,type=0):
for sentence in sentencize(text):
print(u"test run : sentence={0}".format(sentence))
parser.lexicon.guess(sentence)
for t in parser.parse(sentence):
if type==0:
for r in t.leaves():
if r.token in parser.lexicon.guess_dics:
print(u"{0}\t{1}\t(guess)".format(r.token , r.catname))
else:
else:
return True
elif term[0].value() == "forall":
return False
else:
assert (len(term) >= 2), lt
return (check(term[1]) and check(term[2]))
if type(rt) == list or rt.value() != "COMMA":
return None
elif not check(lt):
return None
return lt
parser = CCGParser()
parser.combinators = [
LApp, RApp, LB, RB, Conj, FwdRel, SkipCommaJP,
RT("NP[sbj]"), RBx
]
parser.terminators = [
"ROOT", "S", "S[exc]", "S[imp]", "S[null]", "S[q]", "S[wq]", "S[null-q]",
"S[nom]"
]
parser.lexicon = default_lexicon()
parser.concatenator = ""
def run(text, type=0):
for sentence in sentencize(text):
print(u"test run : sentence={0}".format(sentence))