def normalize(form, unistr):
"""
Return the normal form 'form' for the Unicode string unistr. Valid
values for form are 'NFC', 'NFKC', 'NFD', and 'NFKD'.
"""
try:
normalizer_form = _forms[form]
except KeyError:
raise ValueError('invalid normalization form')
return Normalizer.normalize(unistr, normalizer_form)
def decomposition(unichr):
_validate_unichr(unichr)
d = Normalizer.decompose(unichr, True)
decomp_type = None
if len(d) == 1:
decomp_type = _get_decomp_type(unichr)
else:
for c in d:
decomp_type = _get_decomp_type(c)
# print "Got a decomp_type %r %r %r" % (c, d, decomp_type)
if decomp_type is not None:
break
hexed = " ".join(("{0:04X}".format(ord(c)) for c in d))
if decomp_type:
return "<{}> {}".format(decomp_type, hexed)
elif len(d) == 1:
return ""
else:
return hexed
