abydos.phonetic._metaphone.metaphone() - Code Metrics - Inspection of "Merge pull request #149 from chrislit/0.3.6" - chrislit/abydos - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( f43547...71985b )

by Chris

created 2018-11-17 08:52 UTC

abydos.phonetic._metaphone.metaphone() A

↳ Parent: abydos.phonetic._metaphone

Complexity

Conditions

Size

Total Lines	31
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	2
CRAP Score	1

Importance

Changes

Metric	Value
cc	1
eloc	2
nop	2
dl	0
loc	31
ccs	2
cts	2
cp	1
crap	1
rs	10
c	0
b	0
f	0

# -*- coding: utf-8 -*-

# Copyright 2014-2018 by Christopher C. Little.
# This file is part of Abydos.
#
# Abydos is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# Abydos is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Abydos. If not, see <http://www.gnu.org/licenses/>.

"""abydos.phonetic._metaphone.

Metaphone
"""

from __future__ import (
    absolute_import,
    division,
    print_function,
    unicode_literals,
)

from six.moves import range

from ._phonetic import _Phonetic

__all__ = ['Metaphone', 'metaphone']


class Metaphone(_Phonetic):

    """Metaphone.

    Based on Lawrence Philips' Pick BASIC code from 1990 :cite:`Philips:1990`,
    as described in :cite:`Philips:1990b`.
    This incorporates some corrections to the above code, particularly
    some of those suggested by Michael Kuhn in :cite:`Kuhn:1995`.
    """

    _frontv = {'E', 'I', 'Y'}
    _varson = {'C', 'G', 'P', 'S', 'T'}

    def encode(self, word, max_length=-1):

        """Return the Metaphone code for a word.

        Based on Lawrence Philips' Pick BASIC code from 1990
        :cite:`Philips:1990`, as described in :cite:`Philips:1990b`.
        This incorporates some corrections to the above code, particularly
        some of those suggested by Michael Kuhn in :cite:`Kuhn:1995`.

        Parameters
        ----------
        word : str
            The word to transform
        max_length : int
            The maximum length of the returned Metaphone code (defaults to 64,
            but in Philips' original implementation this was 4)

        Returns
        -------
        str
            The Metaphone value

        Examples
        --------
        >>> pe = Metaphone()
        >>> pe.encode('Christopher')
        'KRSTFR'
        >>> pe.encode('Niall')
        'NL'
        >>> pe.encode('Smith')
        'SM0'
        >>> pe.encode('Schmidt')
        'SKMTT'

        """
        # Require a max_length of at least 4
        if max_length != -1:
            max_length = max(4, max_length)
        else:
            max_length = 64

        # As in variable sound--those modified by adding an "h"
        ename = ''.join(c for c in word.upper() if c.isalnum())
        ename = ename.replace('ß', 'SS')

        # Delete non-alphanumeric characters and make all caps
        if not ename:
            return ''
        if ename[0:2] in {'PN', 'AE', 'KN', 'GN', 'WR'}:
            ename = ename[1:]
        elif ename[0] == 'X':
            ename = 'S' + ename[1:]
        elif ename[0:2] == 'WH':
            ename = 'W' + ename[2:]

        # Convert to metaphone
        elen = len(ename) - 1
        metaph = ''
        for i in range(len(ename)):

            if len(metaph) >= max_length:
                break
            if (
                ename[i] not in {'G', 'T'}

                and i > 0

                and ename[i - 1] == ename[i]

            ):
                continue

            if ename[i] in self._uc_v_set and i == 0:
                metaph = ename[i]

            elif ename[i] == 'B':
                if i != elen or ename[i - 1] != 'M':
                    metaph += ename[i]

            elif ename[i] == 'C':
                if not (
                    i > 0

                    and ename[i - 1] == 'S'

                    and ename[i + 1 : i + 2] in self._frontv

                ):
                    if ename[i + 1 : i + 3] == 'IA':
                        metaph += 'X'
                    elif ename[i + 1 : i + 2] in self._frontv:
                        metaph += 'S'
                    elif i > 0 and ename[i - 1 : i + 2] == 'SCH':
                        metaph += 'K'
                    elif ename[i + 1 : i + 2] == 'H':
                        if (
                            i == 0

                            and i + 1 < elen

                            and ename[i + 2 : i + 3] not in self._uc_v_set

                        ):
                            metaph += 'K'
                        else:
                            metaph += 'X'
                    else:
                        metaph += 'K'

            elif ename[i] == 'D':
                if (
                    ename[i + 1 : i + 2] == 'G'

                    and ename[i + 2 : i + 3] in self._frontv

                ):
                    metaph += 'J'
                else:
                    metaph += 'T'

            elif ename[i] == 'G':
                if ename[i + 1 : i + 2] == 'H' and not (
                    i + 1 == elen or ename[i + 2 : i + 3] not in self._uc_v_set

                ):
                    continue
                elif i > 0 and (
                    (i + 1 == elen and ename[i + 1] == 'N')

                    or (i + 3 == elen and ename[i + 1 : i + 4] == 'NED')

                ):
                    continue
                elif (
                    i - 1 > 0

                    and i + 1 <= elen

                    and ename[i - 1] == 'D'

                    and ename[i + 1] in self._frontv

                ):
                    continue
                elif ename[i + 1 : i + 2] == 'G':
                    continue
                elif ename[i + 1 : i + 2] in self._frontv:
                    if i == 0 or ename[i - 1] != 'G':
                        metaph += 'J'
                    else:
                        metaph += 'K'
                else:
                    metaph += 'K'

            elif ename[i] == 'H':
                if (
                    i > 0

                    and ename[i - 1] in self._uc_v_set

                    and ename[i + 1 : i + 2] not in self._uc_v_set

                ):
                    continue
                elif i > 0 and ename[i - 1] in self._varson:
                    continue
                else:
                    metaph += 'H'

            elif ename[i] in {'F', 'J', 'L', 'M', 'N', 'R'}:
                metaph += ename[i]

            elif ename[i] == 'K':
                if i > 0 and ename[i - 1] == 'C':
                    continue
                else:
                    metaph += 'K'

            elif ename[i] == 'P':
                if ename[i + 1 : i + 2] == 'H':
                    metaph += 'F'
                else:
                    metaph += 'P'

            elif ename[i] == 'Q':
                metaph += 'K'

            elif ename[i] == 'S':
                if (
                    i > 0

                    and i + 2 <= elen

                    and ename[i + 1] == 'I'

                    and ename[i + 2] in 'OA'

                ):
                    metaph += 'X'
                elif ename[i + 1 : i + 2] == 'H':
                    metaph += 'X'
                else:
                    metaph += 'S'

            elif ename[i] == 'T':
                if (
                    i > 0

                    and i + 2 <= elen

                    and ename[i + 1] == 'I'

                    and ename[i + 2] in {'A', 'O'}

                ):
                    metaph += 'X'
                elif ename[i + 1 : i + 2] == 'H':
                    metaph += '0'
                elif ename[i + 1 : i + 3] != 'CH':
                    if ename[i - 1 : i] != 'T':
                        metaph += 'T'

            elif ename[i] == 'V':
                metaph += 'F'

            elif ename[i] in 'WY':
                if ename[i + 1 : i + 2] in self._uc_v_set:
                    metaph += ename[i]

            elif ename[i] == 'X':
                metaph += 'KS'

            elif ename[i] == 'Z':
                metaph += 'S'

        return metaph


def metaphone(word, max_length=-1):
    """Return the Metaphone code for a word.

    This is a wrapper for :py:meth:`Metaphone.encode`.

    Parameters
    ----------
    word : str
        The word to transform
    max_length : int
        The maximum length of the returned Metaphone code (defaults to 64, but
        in Philips' original implementation this was 4)

    Returns
    -------
    str
        The Metaphone value

    Examples
    --------
    >>> metaphone('Christopher')
    'KRSTFR'
    >>> metaphone('Niall')
    'NL'
    >>> metaphone('Smith')
    'SM0'
    >>> metaphone('Schmidt')
    'SKMTT'

    """
    return Metaphone().encode(word, max_length)


if __name__ == '__main__':
    import doctest

    doctest.testmod()


1		# -- coding: utf-8 --
2
3		# Copyright 2014-2018 by Christopher C. Little.
4		# This file is part of Abydos.
5		#
6		# Abydos is free software: you can redistribute it and/or modify
7		# it under the terms of the GNU General Public License as published by
8		# the Free Software Foundation, either version 3 of the License, or
9		# (at your option) any later version.
10		#
11		# Abydos is distributed in the hope that it will be useful,
12		# but WITHOUT ANY WARRANTY; without even the implied warranty of
13		# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14		# GNU General Public License for more details.
15		#
16		# You should have received a copy of the GNU General Public License
17		# along with Abydos. If not, see <http://www.gnu.org/licenses/>.
18
19	1	"""abydos.phonetic._metaphone.
20
21		Metaphone
22		"""
23
24	1	from __future__ import (
25		absolute_import,
26		division,
27		print_function,
28		unicode_literals,
29		)
30
31	1	from six.moves import range
32
33	1	from ._phonetic import _Phonetic
34
35	1	__all__ = ['Metaphone', 'metaphone']
36
37
38	1	class Metaphone(_Phonetic):
		0 ignored issues – show Unused Code introduced 2018-11-10 01:42 UTC by Report Bug Copy Issue Report The variable `__class__` seems to be unused. Loading history...
39		"""Metaphone.
40
41		Based on Lawrence Philips' Pick BASIC code from 1990 :cite:`Philips:1990`,
42		as described in :cite:`Philips:1990b`.
43		This incorporates some corrections to the above code, particularly
44		some of those suggested by Michael Kuhn in :cite:`Kuhn:1995`.
45		"""
46
47	1	_frontv = {'E', 'I', 'Y'}
48	1	_varson = {'C', 'G', 'P', 'S', 'T'}
49
50	1	def encode(self, word, max_length=-1):
		0 ignored issues – show Bug introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Parameters differ from overridden 'encode' method Loading history...
51		"""Return the Metaphone code for a word.
52
53		Based on Lawrence Philips' Pick BASIC code from 1990
54		:cite:`Philips:1990`, as described in :cite:`Philips:1990b`.
55		This incorporates some corrections to the above code, particularly
56		some of those suggested by Michael Kuhn in :cite:`Kuhn:1995`.
57
58		Parameters
59		----------
60		word : str
61		The word to transform
62		max_length : int
63		The maximum length of the returned Metaphone code (defaults to 64,
64		but in Philips' original implementation this was 4)
65
66		Returns
67		-------
68		str
69		The Metaphone value
70
71		Examples
72		--------
73		>>> pe = Metaphone()
74		>>> pe.encode('Christopher')
75		'KRSTFR'
76		>>> pe.encode('Niall')
77		'NL'
78		>>> pe.encode('Smith')
79		'SM0'
80		>>> pe.encode('Schmidt')
81		'SKMTT'
82
83		"""
84		# Require a max_length of at least 4
85	1	if max_length != -1:
86	1	max_length = max(4, max_length)
87		else:
88	1	max_length = 64
89
90		# As in variable sound--those modified by adding an "h"
91	1	ename = ''.join(c for c in word.upper() if c.isalnum())
92	1	ename = ename.replace('ß', 'SS')
93
94		# Delete non-alphanumeric characters and make all caps
95	1	if not ename:
96	1	return ''
97	1	if ename[0:2] in {'PN', 'AE', 'KN', 'GN', 'WR'}:
98	1	ename = ename[1:]
99	1	elif ename[0] == 'X':
100	1	ename = 'S' + ename[1:]
101	1	elif ename[0:2] == 'WH':
102	1	ename = 'W' + ename[2:]
103
104		# Convert to metaphone
105	1	elen = len(ename) - 1
106	1	metaph = ''
107	1	for i in range(len(ename)):
		0 ignored issues – show unused-code introduced 2018-08-02 19:04 UTC by Report Bug Copy Issue Report Consider using enumerate instead of iterating with range and len Loading history...
108	1	if len(metaph) >= max_length:
109	1	break
110	1	if (
111		ename[i] not in {'G', 'T'}
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
112		and i > 0
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
113		and ename[i - 1] == ename[i]
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
114		):
115	1	continue
116
117	1	if ename[i] in self._uc_v_set and i == 0:
118	1	metaph = ename[i]
119
120	1	elif ename[i] == 'B':
121	1	if i != elen or ename[i - 1] != 'M':
122	1	metaph += ename[i]
123
124	1	elif ename[i] == 'C':
125	1	if not (
126		i > 0
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
127		and ename[i - 1] == 'S'
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
128		and ename[i + 1 : i + 2] in self._frontv
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
129		):
130	1	if ename[i + 1 : i + 3] == 'IA':
131	1	metaph += 'X'
132	1	elif ename[i + 1 : i + 2] in self._frontv:
133	1	metaph += 'S'
134	1	elif i > 0 and ename[i - 1 : i + 2] == 'SCH':
135	1	metaph += 'K'
136	1	elif ename[i + 1 : i + 2] == 'H':
137	1	if (
138		i == 0
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
139		and i + 1 < elen
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
140		and ename[i + 2 : i + 3] not in self._uc_v_set
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
141		):
142	1	metaph += 'K'
143		else:
144	1	metaph += 'X'
145		else:
146	1	metaph += 'K'
147
148	1	elif ename[i] == 'D':
149	1	if (
150		ename[i + 1 : i + 2] == 'G'
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
151		and ename[i + 2 : i + 3] in self._frontv
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
152		):
153	1	metaph += 'J'
154		else:
155	1	metaph += 'T'
156
157	1	elif ename[i] == 'G':
158	1	if ename[i + 1 : i + 2] == 'H' and not (
159		i + 1 == elen or ename[i + 2 : i + 3] not in self._uc_v_set
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
160		):
161	1	continue
162	1	elif i > 0 and (
163		(i + 1 == elen and ename[i + 1] == 'N')
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
164		or (i + 3 == elen and ename[i + 1 : i + 4] == 'NED')
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
165		):
166	1	continue
167	1	elif (
168		i - 1 > 0
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
169		and i + 1 <= elen
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
170		and ename[i - 1] == 'D'
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
171		and ename[i + 1] in self._frontv
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
172		):
173	1	continue
174	1	elif ename[i + 1 : i + 2] == 'G':
175	1	continue
176	1	elif ename[i + 1 : i + 2] in self._frontv:
177	1	if i == 0 or ename[i - 1] != 'G':
178	1	metaph += 'J'
179		else:
180	1	metaph += 'K'
181		else:
182	1	metaph += 'K'
183
184	1	elif ename[i] == 'H':
185	1	if (
186		i > 0
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
187		and ename[i - 1] in self._uc_v_set
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
188		and ename[i + 1 : i + 2] not in self._uc_v_set
		0 ignored issues – show Coding Style introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
189		):
190	1	continue
191	1	elif i > 0 and ename[i - 1] in self._varson:
192	1	continue
193		else:
194	1	metaph += 'H'
195
196	1	elif ename[i] in {'F', 'J', 'L', 'M', 'N', 'R'}:
197	1	metaph += ename[i]
198
199	1	elif ename[i] == 'K':
200	1	if i > 0 and ename[i - 1] == 'C':
201	1	continue
202		else:
203	1	metaph += 'K'
204
205	1	elif ename[i] == 'P':
206	1	if ename[i + 1 : i + 2] == 'H':
207	1	metaph += 'F'
208		else:
209	1	metaph += 'P'
210
211	1	elif ename[i] == 'Q':
212	1	metaph += 'K'
213
214	1	elif ename[i] == 'S':
215	1	if (
216		i > 0
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
217		and i + 2 <= elen
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
218		and ename[i + 1] == 'I'
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
219		and ename[i + 2] in 'OA'
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
220		):
221	1	metaph += 'X'
222	1	elif ename[i + 1 : i + 2] == 'H':
223	1	metaph += 'X'
224		else:
225	1	metaph += 'S'
226
227	1	elif ename[i] == 'T':
228	1	if (
229		i > 0
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
230		and i + 2 <= elen
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
231		and ename[i + 1] == 'I'
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
232		and ename[i + 2] in {'A', 'O'}
		0 ignored issues – show Coding Style introduced 2018-10-24 06:00 UTC by Report Bug Copy Issue Report Wrong hanging indentation before block (add 4 spaces). Loading history...
233		):
234	1	metaph += 'X'
235	1	elif ename[i + 1 : i + 2] == 'H':
236	1	metaph += '0'
237	1	elif ename[i + 1 : i + 3] != 'CH':
238	1	if ename[i - 1 : i] != 'T':
239	1	metaph += 'T'
240
241	1	elif ename[i] == 'V':
242	1	metaph += 'F'
243
244	1	elif ename[i] in 'WY':
245	1	if ename[i + 1 : i + 2] in self._uc_v_set:
246	1	metaph += ename[i]
247
248	1	elif ename[i] == 'X':
249	1	metaph += 'KS'
250
251	1	elif ename[i] == 'Z':
252	1	metaph += 'S'
253
254	1	return metaph
255
256
257	1	def metaphone(word, max_length=-1):
258		"""Return the Metaphone code for a word.
259
260		This is a wrapper for :py:meth:`Metaphone.encode`.
261
262		Parameters
263		----------
264		word : str
265		The word to transform
266		max_length : int
267		The maximum length of the returned Metaphone code (defaults to 64, but
268		in Philips' original implementation this was 4)
269
270		Returns
271		-------
272		str
273		The Metaphone value
274
275		Examples
276		--------
277		>>> metaphone('Christopher')
278		'KRSTFR'
279		>>> metaphone('Niall')
280		'NL'
281		>>> metaphone('Smith')
282		'SM0'
283		>>> metaphone('Schmidt')
284		'SKMTT'
285
286		"""
287	1	return Metaphone().encode(word, max_length)
288
289
290		if __name__ == '__main__':
291		import doctest
292
293		doctest.testmod()
294

chrislit / abydos

Push — master ( f43547...71985b )

abydos.phonetic._metaphone.metaphone() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like