def wotan2universal(token, tag):
""" Converts a WOTAN tag to a universal tag.
For example: bokkenrijders/N(soort,mv,neut) => bokkenrijders/NOUN
"""
if tag.startswith("Adv"):
return (token, ADV)
return penntreebank2universal(*wotan2penntreebank(token, tag))
def find_tags(self, tokens, **kwargs):
if kwargs.get("tagset") in (PENN, None):
kwargs.setdefault("map", lambda token, tag: (token, tag))
if kwargs.get("tagset") == UNIVERSAL:
kwargs.setdefault(
"map", lambda token, tag: penntreebank2universal(token, tag))
return _Parser.find_tags(self, tokens, **kwargs)
def parole2universal(token, tag):
""" Converts a Parole tag to a universal tag.
For example: importantísimo/AQ => importantísimo/ADJ
"""
if tag == "CS":
return (token, CONJ)
if tag == "DP":
return (token, DET)
if tag in ("P0", "PD", "PI", "PP", "PR", "PT", "PX"):
return (token, PRON)
return penntreebank2universal(*parole2penntreebank(token, tag))
def stts2universal(token, tag):
""" Converts an STTS tag to a universal tag.
For example: ohne/APPR => ohne/PREP
"""
if tag in ("KON", "KOUI", "KOUS", "KOKOM"):
return (token, CONJ)
if tag in ("PTKZU", "PTKNEG", "PTKVZ", "PTKANT"):
return (token, PRT)
if tag in ("PDF", "PDAT", "PIS", "PIAT", "PIDAT", "PPER", "PPOS",
"PPOSAT"):
return (token, PRON)
if tag in ("PRELS", "PRELAT", "PRF", "PWS", "PWAT", "PWAV", "PAV"):
return (token, PRON)
return penntreebank2universal(*stts2penntreebank(token, tag))