def phonetic(s, method, concat=True, encoding='utf-8', decode_error='strict'):
"""
Convert names or strings into phonetic codes. The implemented
algorithms are `soundex <https://en.wikipedia.org/wiki/Soundex>`_,
`nysiis <https://en.wikipedia.org/wiki/New_York_State_Identification_and_
Intelligence_System>`_, `metaphone
<https://en.wikipedia.org/wiki/Metaphone>`_ or `match_rating
<https://en.wikipedia.org/wiki/Match_rating_approach>`_.
:param method: The algorithm that is used to phonetically encode the
values. The possible options are "soundex", "nysiis",
"metaphone" or "match rating".
:param encoding: If bytes are given, this encoding is used to
decode.
:param decode_error: Instruction on what to do if a byte Series is given
that contains characters not of the given `encoding`. By default,
it is 'strict', meaning that a UnicodeDecodeError will be raised.
Other values are 'ignore' and 'replace'. :type method: str
:return: A Series with phonetic encoded values.
:rtype: pandas.Series
:type encoding: string, default='utf-8'
:type decode_error: {'strict', 'ignore', 'replace'}
.. note::
The 'soundex' and 'nysiis' algorithms use the package 'jellyfish'.
It can be installed with pip (``pip install jellyfish``).
"""
# import jellyfish if present
_check_jellyfish()
# encoding
if sys.version_info[0] == 2:
s = s.apply(
lambda x: x.decode(encoding, decode_error) if type(x) == bytes else x
)
if concat:
s = s.str.replace('[\-\_\s]', '')
if method == 'soundex':
phonetic_callback = jellyfish.soundex
elif method == 'nysiis':
phonetic_callback = jellyfish.nysiis
elif method == 'metaphone':
phonetic_callback = jellyfish.metaphone
elif method in [
'match_rating', 'match rating', 'matchrating',
'match_rating_codex', 'matchratingcodex'
]:
phonetic_callback = jellyfish.match_rating_codex
else:
raise ValueError("The algorithm '{}' is not known.".format(method))
return s.str.upper().apply(
lambda x: phonetic_callback(x) if pandas.notnull(x) else np.nan)
def damerau_levenshtein_similarity(s1, s2):
# Check jellyfish
_check_jellyfish(raise_error=True)
conc = pandas.concat([s1, s2], axis=1, ignore_index=True)
def damerau_levenshtein_apply(x):
try:
return 1 - jellyfish.damerau_levenshtein_distance(x[0], x[1]) / np.max([len(x[0]), len(x[1])])
except Exception as err:
if pandas.isnull(x[0]) or pandas.isnull(x[1]):
return np.nan
else:
raise err
return conc.apply(damerau_levenshtein_apply, axis=1)
def jarowinkler_similarity(s1, s2):
# Check jellyfish
_check_jellyfish(raise_error=True)
conc = pandas.concat([s1, s2], axis=1, ignore_index=True)
def jaro_winkler_apply(x):
try:
return jellyfish.jaro_winkler(x[0], x[1])
except Exception as err:
if pandas.isnull(x[0]) or pandas.isnull(x[1]):
return np.nan
else:
raise err
return conc.apply(jaro_winkler_apply, axis=1)
