abydos.distance._guth.Guth.sim() - Code Metrics - Inspection of "0.4.1" - chrislit/abydos - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#225)

by Chris

created 2019-07-12 05:47 UTC

abydos.distance._guth.Guth.sim() F

↳ Parent: abydos.distance._guth

Complexity

Conditions

Size

Total Lines	100
Code Lines	51

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	51
CRAP Score	33

Importance

Changes

Metric	Value
eloc	51
dl	0
loc	100
ccs	51
cts	51
cp	1
rs	0
c	0
b	0
f	0
cc	33
nop	3
crap	33

How to fix Long Method Complexity

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

# Copyright 2019 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.distance._guth.

Guth matching algorithm
"""

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

from ._distance import _Distance
from ..tokenizer import QGrams

__all__ = ['Guth']


class Guth(_Distance):
    r"""Guth matching.

    Guth matching :cite:`Guth:1976` uses a simple positional matching rule list
    to determine whether two names match. Following the original, the
    :meth:`.sim_score` method returns only 1.0 for matching or 0.0 for
    non-matching.

    The :math:`.sim` mathod instead penalizes more distant matches and never
    outrightly declares two names a non-matching unless no matches can be made
    in the two strings.

    Tokens other than single characters can be matched by specifying a
    tokenizer during initialization or setting the qval parameter.

    .. versionadded:: 0.4.1
    """

    def __init__(self, tokenizer=None, **kwargs):
        """Initialize Guth instance.

        Parameters
        ----------
        tokenizer : _Tokenizer
            A tokenizer instance from the :py:mod:`abydos.tokenizer` package
        **kwargs
            Arbitrary keyword arguments

        Other Parameters
        ----------------
        qval : int
            The length of each q-gram. Using this parameter and tokenizer=None
            will cause the instance to use the QGram tokenizer with this
            q value.


        .. versionadded:: 0.4.1

        """
        super(Guth, self).__init__(**kwargs)

        self.params['tokenizer'] = tokenizer
        if 'qval' in self.params:
            self.params['tokenizer'] = QGrams(
                qval=self.params['qval'], start_stop='$#', skip=0, scaler=None
            )

    def _token_at(self, name, pos):
        """Return the token of name at position pos.

        Parameters
        ----------
        name : str or list
            A string (or list) from which to return a token
        pos : int
            The position of the token to return

        Returns
        -------
        str
            The requested token or None if the position is invalid


        .. versionadded:: 0.4.1

        """
        if pos < 0:
            return None
        if pos >= len(name):
            return None
        return name[pos]

    def sim_score(self, src, tar):
        """Return the Guth matching score of two strings.

        Parameters
        ----------
        src : str
            Source string for comparison
        tar : str
            Target string for comparison

        Returns
        -------
        float
            Guth matching score (1.0 if matching, otherwise 0.0)

        Examples
        --------
        >>> cmp = Guth()
        >>> cmp.sim_score('cat', 'hat')
        1.0
        >>> cmp.sim_score('Niall', 'Neil')
        1.0
        >>> cmp.sim_score('aluminum', 'Catalan')
        0.0
        >>> cmp.sim_score('ATCG', 'TAGC')
        1.0


        .. versionadded:: 0.4.1

        """
        if src == tar:
            return 1.0
        if not src or not tar:
            return 0.0

        if self.params['tokenizer']:
            src = self.params['tokenizer'].tokenize(src).get_list()
            tar = self.params['tokenizer'].tokenize(tar).get_list()

        for pos in range(len(src)):
            s = self._token_at(src, pos)
            t = set(tar[max(0, pos - 1) : pos + 3])
            if s and s in t:
                continue

            s = set(src[max(0, pos - 1) : pos + 3])
            t = self._token_at(tar, pos)
            if t and t in s:
                continue

            s = self._token_at(src, pos + 1)
            t = self._token_at(tar, pos + 1)
            if s and t and s == t:
                continue

            s = self._token_at(src, pos + 2)
            t = self._token_at(tar, pos + 2)
            if s and t and s == t:
                continue

            break
        else:
            return 1.0
        return 0.0

    def sim(self, src, tar):
        """Return the relative Guth similarity of two strings.

        This deviates from the algorithm described in :cite:`Guth:1976` in that
        more distant matches are penalized, so that less similar terms score
        lower that more similar terms.

        If no match is found for a particular token in the source string, this
        does not result in an automatic 0.0 score. Rather, the score is further
        penalized towards 0.0.

        Parameters
        ----------
        src : str
            Source string for comparison
        tar : str
            Target string for comparison

        Returns
        -------
        float
            Relative Guth matching score

        Examples
        --------
        >>> cmp = Guth()
        >>> cmp.sim('cat', 'hat')
        0.8666666666666667
        >>> cmp.sim('Niall', 'Neil')
        0.8800000000000001
        >>> cmp.sim('aluminum', 'Catalan')
        0.4
        >>> cmp.sim('ATCG', 'TAGC')
        0.8


        .. versionadded:: 0.4.1

        """
        if src == tar:
            return 1.0
        if not src or not tar:
            return 0.0

        if self.params['tokenizer']:
            src = self.params['tokenizer'].tokenize(src).get_list()
            tar = self.params['tokenizer'].tokenize(tar).get_list()

        score = 0
        for pos in range(len(src)):
            s = self._token_at(src, pos)
            t = self._token_at(tar, pos)
            if s and t and s == t:
                score += 1.0
                continue

            t = self._token_at(tar, pos + 1)
            if s and t and s == t:
                score += 0.8
                continue

            t = self._token_at(tar, pos + 2)
            if s and t and s == t:
                score += 0.6
                continue

            t = self._token_at(tar, pos - 1)
            if s and t and s == t:
                score += 0.8
                continue

            s = self._token_at(src, pos - 1)
            t = self._token_at(tar, pos)
            if s and t and s == t:
                score += 0.8
                continue

            s = self._token_at(src, pos + 1)
            if s and t and s == t:
                score += 0.8
                continue

            s = self._token_at(src, pos + 2)
            if s and t and s == t:
                score += 0.6
                continue

            s = self._token_at(src, pos + 1)
            t = self._token_at(tar, pos + 1)
            if s and t and s == t:
                score += 0.6
                continue

            s = self._token_at(src, pos + 2)
            t = self._token_at(tar, pos + 2)
            if s and t and s == t:
                score += 0.2
                continue

        return score / len(src)


if __name__ == '__main__':
    import doctest

    doctest.testmod()


1		# -- coding: utf-8 --
2
3		# Copyright 2019 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.distance._guth.
20
21		Guth matching algorithm
22		"""
23
24	1	from __future__ import (
25		absolute_import,
26		division,
27		print_function,
28		unicode_literals,
29		)
30
31	1	from ._distance import _Distance
32	1	from ..tokenizer import QGrams
33
34	1	__all__ = ['Guth']
35
36
37	1	class Guth(_Distance):
38		r"""Guth matching.
39
40		Guth matching :cite:`Guth:1976` uses a simple positional matching rule list
41		to determine whether two names match. Following the original, the
42		:meth:`.sim_score` method returns only 1.0 for matching or 0.0 for
43		non-matching.
44
45		The :math:`.sim` mathod instead penalizes more distant matches and never
46		outrightly declares two names a non-matching unless no matches can be made
47		in the two strings.
48
49		Tokens other than single characters can be matched by specifying a
50		tokenizer during initialization or setting the qval parameter.
51
52		.. versionadded:: 0.4.1
53		"""
54
55	1	def __init__(self, tokenizer=None, **kwargs):
56		"""Initialize Guth instance.
57
58		Parameters
59		----------
60		tokenizer : _Tokenizer
61		A tokenizer instance from the :py:mod:`abydos.tokenizer` package
62		**kwargs
63		Arbitrary keyword arguments
64
65		Other Parameters
66		----------------
67		qval : int
68		The length of each q-gram. Using this parameter and tokenizer=None
69		will cause the instance to use the QGram tokenizer with this
70		q value.
71
72
73		.. versionadded:: 0.4.1
74
75		"""
76	1	super(Guth, self).__init__(**kwargs)
77
78	1	self.params['tokenizer'] = tokenizer
79	1	if 'qval' in self.params:
80	1	self.params['tokenizer'] = QGrams(
81		qval=self.params['qval'], start_stop='$#', skip=0, scaler=None
82		)
83
84	1	def _token_at(self, name, pos):
85		"""Return the token of name at position pos.
86
87		Parameters
88		----------
89		name : str or list
90		A string (or list) from which to return a token
91		pos : int
92		The position of the token to return
93
94		Returns
95		-------
96		str
97		The requested token or None if the position is invalid
98
99
100		.. versionadded:: 0.4.1
101
102		"""
103	1	if pos < 0:
104	1	return None
105	1	if pos >= len(name):
106	1	return None
107	1	return name[pos]
108
109	1	def sim_score(self, src, tar):
110		"""Return the Guth matching score of two strings.
111
112		Parameters
113		----------
114		src : str
115		Source string for comparison
116		tar : str
117		Target string for comparison
118
119		Returns
120		-------
121		float
122		Guth matching score (1.0 if matching, otherwise 0.0)
123
124		Examples
125		--------
126		>>> cmp = Guth()
127		>>> cmp.sim_score('cat', 'hat')
128		1.0
129		>>> cmp.sim_score('Niall', 'Neil')
130		1.0
131		>>> cmp.sim_score('aluminum', 'Catalan')
132		0.0
133		>>> cmp.sim_score('ATCG', 'TAGC')
134		1.0
135
136
137		.. versionadded:: 0.4.1
138
139		"""
140	1	if src == tar:
141	1	return 1.0
142	1	if not src or not tar:
143	1	return 0.0
144
145	1	if self.params['tokenizer']:
146	1	src = self.params['tokenizer'].tokenize(src).get_list()
147	1	tar = self.params['tokenizer'].tokenize(tar).get_list()
148
149	1	for pos in range(len(src)):
150	1	s = self._token_at(src, pos)
151	1	t = set(tar[max(0, pos - 1) : pos + 3])
152	1	if s and s in t:
153	1	continue
154
155	1	s = set(src[max(0, pos - 1) : pos + 3])
156	1	t = self._token_at(tar, pos)
157	1	if t and t in s:
158	1	continue
159
160	1	s = self._token_at(src, pos + 1)
161	1	t = self._token_at(tar, pos + 1)
162	1	if s and t and s == t:
163	1	continue
164
165	1	s = self._token_at(src, pos + 2)
166	1	t = self._token_at(tar, pos + 2)
167	1	if s and t and s == t:
168	1	continue
169
170	1	break
171		else:
172	1	return 1.0
173	1	return 0.0
174
175	1	def sim(self, src, tar):
176		"""Return the relative Guth similarity of two strings.
177
178		This deviates from the algorithm described in :cite:`Guth:1976` in that
179		more distant matches are penalized, so that less similar terms score
180		lower that more similar terms.
181
182		If no match is found for a particular token in the source string, this
183		does not result in an automatic 0.0 score. Rather, the score is further
184		penalized towards 0.0.
185
186		Parameters
187		----------
188		src : str
189		Source string for comparison
190		tar : str
191		Target string for comparison
192
193		Returns
194		-------
195		float
196		Relative Guth matching score
197
198		Examples
199		--------
200		>>> cmp = Guth()
201		>>> cmp.sim('cat', 'hat')
202		0.8666666666666667
203		>>> cmp.sim('Niall', 'Neil')
204		0.8800000000000001
205		>>> cmp.sim('aluminum', 'Catalan')
206		0.4
207		>>> cmp.sim('ATCG', 'TAGC')
208		0.8
209
210
211		.. versionadded:: 0.4.1
212
213		"""
214	1	if src == tar:
215	1	return 1.0
216	1	if not src or not tar:
217	1	return 0.0
218
219	1	if self.params['tokenizer']:
220	1	src = self.params['tokenizer'].tokenize(src).get_list()
221	1	tar = self.params['tokenizer'].tokenize(tar).get_list()
222
223	1	score = 0
224	1	for pos in range(len(src)):
225	1	s = self._token_at(src, pos)
226	1	t = self._token_at(tar, pos)
227	1	if s and t and s == t:
228	1	score += 1.0
229	1	continue
230
231	1	t = self._token_at(tar, pos + 1)
232	1	if s and t and s == t:
233	1	score += 0.8
234	1	continue
235
236	1	t = self._token_at(tar, pos + 2)
237	1	if s and t and s == t:
238	1	score += 0.6
239	1	continue
240
241	1	t = self._token_at(tar, pos - 1)
242	1	if s and t and s == t:
243	1	score += 0.8
244	1	continue
245
246	1	s = self._token_at(src, pos - 1)
247	1	t = self._token_at(tar, pos)
248	1	if s and t and s == t:
249	1	score += 0.8
250	1	continue
251
252	1	s = self._token_at(src, pos + 1)
253	1	if s and t and s == t:
254	1	score += 0.8
255	1	continue
256
257	1	s = self._token_at(src, pos + 2)
258	1	if s and t and s == t:
259	1	score += 0.6
260	1	continue
261
262	1	s = self._token_at(src, pos + 1)
263	1	t = self._token_at(tar, pos + 1)
264	1	if s and t and s == t:
265	1	score += 0.6
266	1	continue
267
268	1	s = self._token_at(src, pos + 2)
269	1	t = self._token_at(tar, pos + 2)
270	1	if s and t and s == t:
271	1	score += 0.2
272	1	continue
273
274	1	return score / len(src)
275
276
277		if __name__ == '__main__':
278		import doctest
279
280		doctest.testmod()
281

chrislit / abydos

Pull Request — master (#225)

abydos.distance._guth.Guth.sim() F

Complexity

Size

Duplication

Code Coverage

Importance

How to fix Long Method Complexity

Long Method

Complexity

Duplication Side-by-Side

Filter issues like