abydos.distance._compression.sim_ncd_bz2() - Code Metrics - Inspection of "0.3.6" - chrislit/abydos - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#141)

by Chris

created 2018-11-08 03:44 UTC

abydos.distance._compression.sim_ncd_bz2() A

↳ Parent: abydos.distance._compression

Complexity

Conditions

Size

Total Lines	24
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	2
CRAP Score	1

Importance

Changes

Metric	Value
eloc	2
dl	0
loc	24
ccs	2
cts	2
cp	1
rs	10
c	0
b	0
f	0
cc	1
nop	2
crap	1

# -*- 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.distance.compression.

The distance.compression module implements compression distance measures.
"""

from __future__ import division, unicode_literals

import bz2
import zlib

from ._distance import Distance
from ..compression import Arithmetic, BWT, RLE

try:
    import lzma
except ImportError:  # pragma: no cover
    # If the system lacks the lzma library, that's fine, but lzma compression
    # similarity won't be supported.
    lzma = None

__all__ = [
    'NCDzlib',
    'dist_ncd_zlib',
    'sim_ncd_zlib',
    'NCDbz2',
    'dist_ncd_bz2',
    'sim_ncd_bz2',
    'NCDlzma',
    'dist_ncd_lzma',
    'sim_ncd_lzma',
    'NCDbwtrle',
    'dist_ncd_bwtrle',
    'sim_ncd_bwtrle',
    'NCDrle',
    'dist_ncd_rle',
    'sim_ncd_rle',
    'NCDarith',
    'dist_ncd_arith',
    'sim_ncd_arith',
]


class NCDarith(Distance):

    """Normalized Compression Distance using Arithmetic Coding.

    Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
    """

    _coder = None

    def __init__(self):
        """Initialize the arithmetic coder object."""
        self._coder = Arithmetic()

    def dist(self, src, tar, probs=None):

        """Return the NCD between two strings using arithmetic coding.

        Args:
            src (str): Source string for comparison
            tar (str): Target string for comparison
            probs (dict): A dictionary trained with :py:meth:`Arithmetic.train`

        Returns:
            float: Compression distance

        Examples:
            >>> cmp = NCDarith()
            >>> cmp.dist('cat', 'hat')
            0.5454545454545454
            >>> cmp.dist('Niall', 'Neil')
            0.6875
            >>> cmp.dist('aluminum', 'Catalan')
            0.8275862068965517
            >>> cmp.dist('ATCG', 'TAGC')
            0.6923076923076923

        """
        if src == tar:
            return 0.0

        if probs is None:
            # lacking a reasonable dictionary, train on the strings themselves
            self._coder.train(src + tar)
        else:
            self._coder.set_probs(probs)

        src_comp = self._coder.encode(src)[1]
        tar_comp = self._coder.encode(tar)[1]
        concat_comp = self._coder.encode(src + tar)[1]
        concat_comp2 = self._coder.encode(tar + src)[1]

        return (
            min(concat_comp, concat_comp2) - min(src_comp, tar_comp)
        ) / max(src_comp, tar_comp)


def dist_ncd_arith(src, tar, probs=None):
    """Return the NCD between two strings using arithmetic coding.

    This is a wrapper for :py:meth:`NCDarith.dist`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison
        probs (dict): A dictionary trained with :py:meth:`Arithmetic.train`

    Returns:
        float: Compression distance

    Examples:
        >>> dist_ncd_arith('cat', 'hat')
        0.5454545454545454
        >>> dist_ncd_arith('Niall', 'Neil')
        0.6875
        >>> dist_ncd_arith('aluminum', 'Catalan')
        0.8275862068965517
        >>> dist_ncd_arith('ATCG', 'TAGC')
        0.6923076923076923

    """
    return NCDarith().dist(src, tar, probs)


def sim_ncd_arith(src, tar, probs=None):
    """Return the NCD similarity between two strings using arithmetic coding.

    This is a wrapper for :py:meth:`NCDarith.sim`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison
        probs (dict): A dictionary trained with :py:meth:`Arithmetic.train`

    Returns:
        float: Compression similarity

    Examples:
        >>> sim_ncd_arith('cat', 'hat')
        0.4545454545454546
        >>> sim_ncd_arith('Niall', 'Neil')
        0.3125
        >>> sim_ncd_arith('aluminum', 'Catalan')
        0.1724137931034483
        >>> sim_ncd_arith('ATCG', 'TAGC')
        0.3076923076923077

    """
    return NCDarith().sim(src, tar, probs)


class NCDrle(Distance):

    """Normalized Compression Distance using RLE.

    Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
    """

    _rle = RLE()

    def dist(self, src, tar):

        """Return the NCD between two strings using RLE.

        Args:
            src (str): Source string for comparison
            tar (str): Target string for comparison

        Returns:
            float: Compression distance

        Examples:
            >>> cmp = NCDrle()
            >>> cmp.dist('cat', 'hat')
            1.0
            >>> cmp.dist('Niall', 'Neil')
            1.0
            >>> cmp.dist('aluminum', 'Catalan')
            1.0
            >>> cmp.dist('ATCG', 'TAGC')
            1.0

        """
        if src == tar:
            return 0.0

        src_comp = self._rle.encode(src)
        tar_comp = self._rle.encode(tar)
        concat_comp = self._rle.encode(src + tar)
        concat_comp2 = self._rle.encode(tar + src)

        return (
            min(len(concat_comp), len(concat_comp2))
            - min(len(src_comp), len(tar_comp))
        ) / max(len(src_comp), len(tar_comp))


def dist_ncd_rle(src, tar):
    """Return the NCD between two strings using RLE.

    This is a wrapper for :py:meth:`NCDrle.dist`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression distance

    Examples:
        >>> dist_ncd_rle('cat', 'hat')
        1.0
        >>> dist_ncd_rle('Niall', 'Neil')
        1.0
        >>> dist_ncd_rle('aluminum', 'Catalan')
        1.0
        >>> dist_ncd_rle('ATCG', 'TAGC')
        1.0

    """
    return NCDrle().dist(src, tar)


def sim_ncd_rle(src, tar):
    """Return the NCD similarity between two strings using RLE.

    This is a wrapper for :py:meth:`NCDrle.sim`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression similarity

    Examples:
        >>> sim_ncd_rle('cat', 'hat')
        0.0
        >>> sim_ncd_rle('Niall', 'Neil')
        0.0
        >>> sim_ncd_rle('aluminum', 'Catalan')
        0.0
        >>> sim_ncd_rle('ATCG', 'TAGC')
        0.0

    """
    return NCDrle().sim(src, tar)


class NCDbwtrle(NCDrle):

    """Normalized Compression Distance using BWT plus RLE.

    Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
    """

    _bwt = BWT()

    def dist(self, src, tar):
        """Return the NCD between two strings using BWT plus RLE.

        Args:
            src (str): Source string for comparison
            tar (str): Target string for comparison

        Returns:
            float: Compression distance

        Examples:
            >>> cmp = NCDbwtrle()
            >>> cmp.dist('cat', 'hat')
            0.75
            >>> cmp.dist('Niall', 'Neil')
            0.8333333333333334
            >>> cmp.dist('aluminum', 'Catalan')
            1.0
            >>> cmp.dist('ATCG', 'TAGC')
            0.8

        """
        if src == tar:
            return 0.0

        src_comp = self._rle.encode(self._bwt.encode(src))
        tar_comp = self._rle.encode(self._bwt.encode(tar))
        concat_comp = self._rle.encode(self._bwt.encode(src + tar))
        concat_comp2 = self._rle.encode(self._bwt.encode(tar + src))

        return (
            min(len(concat_comp), len(concat_comp2))
            - min(len(src_comp), len(tar_comp))
        ) / max(len(src_comp), len(tar_comp))


def dist_ncd_bwtrle(src, tar):
    """Return the NCD between two strings using BWT plus RLE.

    This is a wrapper for :py:meth:`NCDbwtrle.dist`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression distance

    Examples:
        >>> dist_ncd_bwtrle('cat', 'hat')
        0.75
        >>> dist_ncd_bwtrle('Niall', 'Neil')
        0.8333333333333334
        >>> dist_ncd_bwtrle('aluminum', 'Catalan')
        1.0
        >>> dist_ncd_bwtrle('ATCG', 'TAGC')
        0.8

    """
    return NCDbwtrle().dist(src, tar)


def sim_ncd_bwtrle(src, tar):
    """Return the NCD similarity between two strings using BWT plus RLE.

    This is a wrapper for :py:meth:`NCDbwtrle.sim`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression similarity

    Examples:
        >>> sim_ncd_bwtrle('cat', 'hat')
        0.25
        >>> sim_ncd_bwtrle('Niall', 'Neil')
        0.16666666666666663
        >>> sim_ncd_bwtrle('aluminum', 'Catalan')
        0.0
        >>> sim_ncd_bwtrle('ATCG', 'TAGC')
        0.19999999999999996

    """
    return NCDbwtrle().sim(src, tar)


class NCDzlib(Distance):

    """Normalized Compression Distance using zlib compression.

    Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
    """

    _compressor = None

    def __init__(self, level=zlib.Z_DEFAULT_COMPRESSION):
        """Initialize zlib compressor.

        Args:
            level (int): The compression level (0 to 9)
        """
        self._compressor = zlib.compressobj(level)

    def dist(self, src, tar):

        """Return the NCD between two strings using zlib compression.

        Args:
            src (str): Source string for comparison
            tar (str): Target string for comparison

        Returns:
            float: Compression distance

        Examples:
            >>> cmp = NCDzlib()
            >>> cmp.dist('cat', 'hat')
            0.3333333333333333
            >>> cmp.dist('Niall', 'Neil')
            0.45454545454545453
            >>> cmp.dist('aluminum', 'Catalan')
            0.5714285714285714
            >>> cmp.dist('ATCG', 'TAGC')
            0.4

        """
        if src == tar:
            return 0.0

        src = src.encode('utf-8')
        tar = tar.encode('utf-8')

        self._compressor.compress(src)
        src_comp = self._compressor.flush(zlib.Z_FULL_FLUSH)
        self._compressor.compress(tar)
        tar_comp = self._compressor.flush(zlib.Z_FULL_FLUSH)
        self._compressor.compress(src + tar)
        concat_comp = self._compressor.flush(zlib.Z_FULL_FLUSH)
        self._compressor.compress(tar + src)
        concat_comp2 = self._compressor.flush(zlib.Z_FULL_FLUSH)

        return (
            min(len(concat_comp), len(concat_comp2))
            - min(len(src_comp), len(tar_comp))
        ) / max(len(src_comp), len(tar_comp))


def dist_ncd_zlib(src, tar):
    """Return the NCD between two strings using zlib compression.

    This is a wrapper for :py:meth:`NCDzlib.dist`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression distance

    Examples:
        >>> dist_ncd_zlib('cat', 'hat')
        0.3333333333333333
        >>> dist_ncd_zlib('Niall', 'Neil')
        0.45454545454545453
        >>> dist_ncd_zlib('aluminum', 'Catalan')
        0.5714285714285714
        >>> dist_ncd_zlib('ATCG', 'TAGC')
        0.4

    """
    return NCDzlib().dist(src, tar)


def sim_ncd_zlib(src, tar):
    """Return the NCD similarity between two strings using zlib compression.

    This is a wrapper for :py:meth:`NCDzlib.sim`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression similarity

    Examples:
        >>> sim_ncd_zlib('cat', 'hat')
        0.6666666666666667
        >>> sim_ncd_zlib('Niall', 'Neil')
        0.5454545454545454
        >>> sim_ncd_zlib('aluminum', 'Catalan')
        0.4285714285714286
        >>> sim_ncd_zlib('ATCG', 'TAGC')
        0.6

    """
    return NCDzlib().sim(src, tar)


class NCDbz2(Distance):

    """Normalized Compression Distance using bz2 compression.

    Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
    """

    _level = 9

    def __init__(self, level=9):
        """Initialize zlib compressor.

        Args:
            level (int): The compression level (0 to 9)
        """
        self._level = level

    def dist(self, src, tar):

        """Return the NCD between two strings using bz2 compression.

        Args:
            src (str): Source string for comparison
            tar (str): Target string for comparison

        Returns:
            float: Compression distance

        Examples:
            >>> cmp = NCDbz2()
            >>> cmp.dist('cat', 'hat')
            0.06666666666666667
            >>> cmp.dist('Niall', 'Neil')
            0.03125
            >>> cmp.dist('aluminum', 'Catalan')
            0.17647058823529413
            >>> cmp.dist('ATCG', 'TAGC')
            0.03125

        """
        if src == tar:
            return 0.0

        src = src.encode('utf-8')
        tar = tar.encode('utf-8')

        src_comp = bz2.compress(src, self._level)[10:]
        tar_comp = bz2.compress(tar, self._level)[10:]
        concat_comp = bz2.compress(src + tar, self._level)[10:]
        concat_comp2 = bz2.compress(tar + src, self._level)[10:]

        return (
            min(len(concat_comp), len(concat_comp2))
            - min(len(src_comp), len(tar_comp))
        ) / max(len(src_comp), len(tar_comp))


def dist_ncd_bz2(src, tar):
    """Return the NCD between two strings using bz2 compression.

    This is a wrapper for :py:meth:`NCDbz2.dist`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression distance

    Examples:
        >>> dist_ncd_bz2('cat', 'hat')
        0.06666666666666667
        >>> dist_ncd_bz2('Niall', 'Neil')
        0.03125
        >>> dist_ncd_bz2('aluminum', 'Catalan')
        0.17647058823529413
        >>> dist_ncd_bz2('ATCG', 'TAGC')
        0.03125

    """
    return NCDbz2().dist(src, tar)


def sim_ncd_bz2(src, tar):
    """Return the NCD similarity between two strings using bz2 compression.

    This is a wrapper for :py:meth:`NCDbz2.sim`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression similarity

    Examples:
        >>> sim_ncd_bz2('cat', 'hat')
        0.9333333333333333
        >>> sim_ncd_bz2('Niall', 'Neil')
        0.96875
        >>> sim_ncd_bz2('aluminum', 'Catalan')
        0.8235294117647058
        >>> sim_ncd_bz2('ATCG', 'TAGC')
        0.96875

    """
    return NCDbz2().sim(src, tar)


class NCDlzma(Distance):

    """Normalized Compression Distance using lzma compression.

    Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
    """

    def dist(self, src, tar):

        """Return the NCD between two strings using lzma compression.

        Args:
            src (str): Source string for comparison
            tar (str): Target string for comparison

        Returns:
            float: Compression distance

        Raises:
            ValueError: Install the PylibLZMA module in order to use lzma

        Examples:
            >>> cmp = NCDlzma()
            >>> cmp.dist('cat', 'hat')
            0.08695652173913043
            >>> cmp.dist('Niall', 'Neil')
            0.16
            >>> cmp.dist('aluminum', 'Catalan')
            0.16
            >>> cmp.dist('ATCG', 'TAGC')
            0.08695652173913043

        """
        if src == tar:
            return 0.0

        src = src.encode('utf-8')
        tar = tar.encode('utf-8')

        if lzma is not None:
            src_comp = lzma.compress(src)[14:]
            tar_comp = lzma.compress(tar)[14:]
            concat_comp = lzma.compress(src + tar)[14:]
            concat_comp2 = lzma.compress(tar + src)[14:]
        else:  # pragma: no cover
            raise ValueError(
                'Install the PylibLZMA module in order to use lzma'
            )

        return (
            min(len(concat_comp), len(concat_comp2))
            - min(len(src_comp), len(tar_comp))
        ) / max(len(src_comp), len(tar_comp))


def dist_ncd_lzma(src, tar):
    """Return the NCD between two strings using lzma compression.

    This is a wrapper for :py:meth:`NCDlzma.dist`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression distance

    Examples:
        >>> dist_ncd_lzma('cat', 'hat')
        0.08695652173913043
        >>> dist_ncd_lzma('Niall', 'Neil')
        0.16
        >>> dist_ncd_lzma('aluminum', 'Catalan')
        0.16
        >>> dist_ncd_lzma('ATCG', 'TAGC')
        0.08695652173913043

    """
    return NCDlzma().dist(src, tar)


def sim_ncd_lzma(src, tar):
    """Return the NCD similarity between two strings using lzma compression.

    This is a wrapper for :py:meth:`NCDlzma.sim`.

    Args:
        src (str): Source string for comparison
        tar (str): Target string for comparison

    Returns:
        float: Compression similarity

    Examples:
        >>> sim_ncd_lzma('cat', 'hat')
        0.9130434782608696
        >>> sim_ncd_lzma('Niall', 'Neil')
        0.84
        >>> sim_ncd_lzma('aluminum', 'Catalan')
        0.84
        >>> sim_ncd_lzma('ATCG', 'TAGC')
        0.9130434782608696

    """
    return NCDlzma().sim(src, tar)


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.distance.compression.
20
21		The distance.compression module implements compression distance measures.
22		"""
23
24	1	from __future__ import division, unicode_literals
25
26	1	import bz2
27	1	import zlib
28
29	1	from ._distance import Distance
30	1	from ..compression import Arithmetic, BWT, RLE
31
32	1	try:
33	1	import lzma
34		except ImportError: # pragma: no cover
35		# If the system lacks the lzma library, that's fine, but lzma compression
36		# similarity won't be supported.
37		lzma = None
38
39	1	__all__ = [
40		'NCDzlib',
41		'dist_ncd_zlib',
42		'sim_ncd_zlib',
43		'NCDbz2',
44		'dist_ncd_bz2',
45		'sim_ncd_bz2',
46		'NCDlzma',
47		'dist_ncd_lzma',
48		'sim_ncd_lzma',
49		'NCDbwtrle',
50		'dist_ncd_bwtrle',
51		'sim_ncd_bwtrle',
52		'NCDrle',
53		'dist_ncd_rle',
54		'sim_ncd_rle',
55		'NCDarith',
56		'dist_ncd_arith',
57		'sim_ncd_arith',
58		]
59
60
61	1	class NCDarith(Distance):
		0 ignored issues – show Unused Code introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report The variable `__class__` seems to be unused. Loading history...
62		"""Normalized Compression Distance using Arithmetic Coding.
63
64		Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
65		"""
66
67	1	_coder = None
68
69	1	def __init__(self):
70		"""Initialize the arithmetic coder object."""
71	1	self._coder = Arithmetic()
72
73	1	def dist(self, src, tar, probs=None):
		0 ignored issues – show Bug introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Parameters differ from overridden 'dist' method Loading history...
74		"""Return the NCD between two strings using arithmetic coding.
75
76		Args:
77		src (str): Source string for comparison
78		tar (str): Target string for comparison
79		probs (dict): A dictionary trained with :py:meth:`Arithmetic.train`
80
81		Returns:
82		float: Compression distance
83
84		Examples:
85		>>> cmp = NCDarith()
86		>>> cmp.dist('cat', 'hat')
87		0.5454545454545454
88		>>> cmp.dist('Niall', 'Neil')
89		0.6875
90		>>> cmp.dist('aluminum', 'Catalan')
91		0.8275862068965517
92		>>> cmp.dist('ATCG', 'TAGC')
93		0.6923076923076923
94
95		"""
96	1	if src == tar:
97	1	return 0.0
98
99	1	if probs is None:
100		# lacking a reasonable dictionary, train on the strings themselves
101	1	self._coder.train(src + tar)
102		else:
103	1	self._coder.set_probs(probs)
104
105	1	src_comp = self._coder.encode(src)[1]
106	1	tar_comp = self._coder.encode(tar)[1]
107	1	concat_comp = self._coder.encode(src + tar)[1]
108	1	concat_comp2 = self._coder.encode(tar + src)[1]
109
110	1	return (
111		min(concat_comp, concat_comp2) - min(src_comp, tar_comp)
112		) / max(src_comp, tar_comp)
113
114
115	1	def dist_ncd_arith(src, tar, probs=None):
116		"""Return the NCD between two strings using arithmetic coding.
117
118		This is a wrapper for :py:meth:`NCDarith.dist`.
119
120		Args:
121		src (str): Source string for comparison
122		tar (str): Target string for comparison
123		probs (dict): A dictionary trained with :py:meth:`Arithmetic.train`
124
125		Returns:
126		float: Compression distance
127
128		Examples:
129		>>> dist_ncd_arith('cat', 'hat')
130		0.5454545454545454
131		>>> dist_ncd_arith('Niall', 'Neil')
132		0.6875
133		>>> dist_ncd_arith('aluminum', 'Catalan')
134		0.8275862068965517
135		>>> dist_ncd_arith('ATCG', 'TAGC')
136		0.6923076923076923
137
138		"""
139	1	return NCDarith().dist(src, tar, probs)
140
141
142	1	def sim_ncd_arith(src, tar, probs=None):
143		"""Return the NCD similarity between two strings using arithmetic coding.
144
145		This is a wrapper for :py:meth:`NCDarith.sim`.
146
147		Args:
148		src (str): Source string for comparison
149		tar (str): Target string for comparison
150		probs (dict): A dictionary trained with :py:meth:`Arithmetic.train`
151
152		Returns:
153		float: Compression similarity
154
155		Examples:
156		>>> sim_ncd_arith('cat', 'hat')
157		0.4545454545454546
158		>>> sim_ncd_arith('Niall', 'Neil')
159		0.3125
160		>>> sim_ncd_arith('aluminum', 'Catalan')
161		0.1724137931034483
162		>>> sim_ncd_arith('ATCG', 'TAGC')
163		0.3076923076923077
164
165		"""
166	1	return NCDarith().sim(src, tar, probs)
167
168
169	1	class NCDrle(Distance):
		0 ignored issues – show Unused Code introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report The variable `__class__` seems to be unused. Loading history...
170		"""Normalized Compression Distance using RLE.
171
172		Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
173		"""
174
175	1	_rle = RLE()
176
177	1	def dist(self, src, tar):
		0 ignored issues – show Bug introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Parameters differ from overridden 'dist' method Loading history...
178		"""Return the NCD between two strings using RLE.
179
180		Args:
181		src (str): Source string for comparison
182		tar (str): Target string for comparison
183
184		Returns:
185		float: Compression distance
186
187		Examples:
188		>>> cmp = NCDrle()
189		>>> cmp.dist('cat', 'hat')
190		1.0
191		>>> cmp.dist('Niall', 'Neil')
192		1.0
193		>>> cmp.dist('aluminum', 'Catalan')
194		1.0
195		>>> cmp.dist('ATCG', 'TAGC')
196		1.0
197
198		"""
199	1	if src == tar:
200	1	return 0.0
201
202	1	src_comp = self._rle.encode(src)
203	1	tar_comp = self._rle.encode(tar)
204	1	concat_comp = self._rle.encode(src + tar)
205	1	concat_comp2 = self._rle.encode(tar + src)
206
207	1	return (
208		min(len(concat_comp), len(concat_comp2))
209		- min(len(src_comp), len(tar_comp))
210		) / max(len(src_comp), len(tar_comp))
211
212
213	1	def dist_ncd_rle(src, tar):
214		"""Return the NCD between two strings using RLE.
215
216		This is a wrapper for :py:meth:`NCDrle.dist`.
217
218		Args:
219		src (str): Source string for comparison
220		tar (str): Target string for comparison
221
222		Returns:
223		float: Compression distance
224
225		Examples:
226		>>> dist_ncd_rle('cat', 'hat')
227		1.0
228		>>> dist_ncd_rle('Niall', 'Neil')
229		1.0
230		>>> dist_ncd_rle('aluminum', 'Catalan')
231		1.0
232		>>> dist_ncd_rle('ATCG', 'TAGC')
233		1.0
234
235		"""
236	1	return NCDrle().dist(src, tar)
237
238
239	1	def sim_ncd_rle(src, tar):
240		"""Return the NCD similarity between two strings using RLE.
241
242		This is a wrapper for :py:meth:`NCDrle.sim`.
243
244		Args:
245		src (str): Source string for comparison
246		tar (str): Target string for comparison
247
248		Returns:
249		float: Compression similarity
250
251		Examples:
252		>>> sim_ncd_rle('cat', 'hat')
253		0.0
254		>>> sim_ncd_rle('Niall', 'Neil')
255		0.0
256		>>> sim_ncd_rle('aluminum', 'Catalan')
257		0.0
258		>>> sim_ncd_rle('ATCG', 'TAGC')
259		0.0
260
261		"""
262	1	return NCDrle().sim(src, tar)
263
264
265	1	class NCDbwtrle(NCDrle):
		0 ignored issues – show Unused Code introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report The variable `__class__` seems to be unused. Loading history...
266		"""Normalized Compression Distance using BWT plus RLE.
267
268		Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
269		"""
270
271	1	_bwt = BWT()
272
273	1	def dist(self, src, tar):
274		"""Return the NCD between two strings using BWT plus RLE.
275
276		Args:
277		src (str): Source string for comparison
278		tar (str): Target string for comparison
279
280		Returns:
281		float: Compression distance
282
283		Examples:
284		>>> cmp = NCDbwtrle()
285		>>> cmp.dist('cat', 'hat')
286		0.75
287		>>> cmp.dist('Niall', 'Neil')
288		0.8333333333333334
289		>>> cmp.dist('aluminum', 'Catalan')
290		1.0
291		>>> cmp.dist('ATCG', 'TAGC')
292		0.8
293
294		"""
295	1	if src == tar:
296	1	return 0.0
297
298	1	src_comp = self._rle.encode(self._bwt.encode(src))
299	1	tar_comp = self._rle.encode(self._bwt.encode(tar))
300	1	concat_comp = self._rle.encode(self._bwt.encode(src + tar))
301	1	concat_comp2 = self._rle.encode(self._bwt.encode(tar + src))
302
303	1	return (
304		min(len(concat_comp), len(concat_comp2))
305		- min(len(src_comp), len(tar_comp))
306		) / max(len(src_comp), len(tar_comp))
307
308
309	1	def dist_ncd_bwtrle(src, tar):
310		"""Return the NCD between two strings using BWT plus RLE.
311
312		This is a wrapper for :py:meth:`NCDbwtrle.dist`.
313
314		Args:
315		src (str): Source string for comparison
316		tar (str): Target string for comparison
317
318		Returns:
319		float: Compression distance
320
321		Examples:
322		>>> dist_ncd_bwtrle('cat', 'hat')
323		0.75
324		>>> dist_ncd_bwtrle('Niall', 'Neil')
325		0.8333333333333334
326		>>> dist_ncd_bwtrle('aluminum', 'Catalan')
327		1.0
328		>>> dist_ncd_bwtrle('ATCG', 'TAGC')
329		0.8
330
331		"""
332	1	return NCDbwtrle().dist(src, tar)
333
334
335	1	def sim_ncd_bwtrle(src, tar):
336		"""Return the NCD similarity between two strings using BWT plus RLE.
337
338		This is a wrapper for :py:meth:`NCDbwtrle.sim`.
339
340		Args:
341		src (str): Source string for comparison
342		tar (str): Target string for comparison
343
344		Returns:
345		float: Compression similarity
346
347		Examples:
348		>>> sim_ncd_bwtrle('cat', 'hat')
349		0.25
350		>>> sim_ncd_bwtrle('Niall', 'Neil')
351		0.16666666666666663
352		>>> sim_ncd_bwtrle('aluminum', 'Catalan')
353		0.0
354		>>> sim_ncd_bwtrle('ATCG', 'TAGC')
355		0.19999999999999996
356
357		"""
358	1	return NCDbwtrle().sim(src, tar)
359
360
361	1	class NCDzlib(Distance):
		0 ignored issues – show Unused Code introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report The variable `__class__` seems to be unused. Loading history...
362		"""Normalized Compression Distance using zlib compression.
363
364		Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
365		"""
366
367	1	_compressor = None
368
369	1	def __init__(self, level=zlib.Z_DEFAULT_COMPRESSION):
370		"""Initialize zlib compressor.
371
372		Args:
373		level (int): The compression level (0 to 9)
374		"""
375	1	self._compressor = zlib.compressobj(level)
376
377	1	def dist(self, src, tar):
		0 ignored issues – show Bug introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Parameters differ from overridden 'dist' method Loading history...
378		"""Return the NCD between two strings using zlib compression.
379
380		Args:
381		src (str): Source string for comparison
382		tar (str): Target string for comparison
383
384		Returns:
385		float: Compression distance
386
387		Examples:
388		>>> cmp = NCDzlib()
389		>>> cmp.dist('cat', 'hat')
390		0.3333333333333333
391		>>> cmp.dist('Niall', 'Neil')
392		0.45454545454545453
393		>>> cmp.dist('aluminum', 'Catalan')
394		0.5714285714285714
395		>>> cmp.dist('ATCG', 'TAGC')
396		0.4
397
398		"""
399	1	if src == tar:
400	1	return 0.0
401
402	1	src = src.encode('utf-8')
403	1	tar = tar.encode('utf-8')
404
405	1	self._compressor.compress(src)
406	1	src_comp = self._compressor.flush(zlib.Z_FULL_FLUSH)
407	1	self._compressor.compress(tar)
408	1	tar_comp = self._compressor.flush(zlib.Z_FULL_FLUSH)
409	1	self._compressor.compress(src + tar)
410	1	concat_comp = self._compressor.flush(zlib.Z_FULL_FLUSH)
411	1	self._compressor.compress(tar + src)
412	1	concat_comp2 = self._compressor.flush(zlib.Z_FULL_FLUSH)
413
414	1	return (
415		min(len(concat_comp), len(concat_comp2))
416		- min(len(src_comp), len(tar_comp))
417		) / max(len(src_comp), len(tar_comp))
418
419
420	1	def dist_ncd_zlib(src, tar):
421		"""Return the NCD between two strings using zlib compression.
422
423		This is a wrapper for :py:meth:`NCDzlib.dist`.
424
425		Args:
426		src (str): Source string for comparison
427		tar (str): Target string for comparison
428
429		Returns:
430		float: Compression distance
431
432		Examples:
433		>>> dist_ncd_zlib('cat', 'hat')
434		0.3333333333333333
435		>>> dist_ncd_zlib('Niall', 'Neil')
436		0.45454545454545453
437		>>> dist_ncd_zlib('aluminum', 'Catalan')
438		0.5714285714285714
439		>>> dist_ncd_zlib('ATCG', 'TAGC')
440		0.4
441
442		"""
443	1	return NCDzlib().dist(src, tar)
444
445
446	1	def sim_ncd_zlib(src, tar):
447		"""Return the NCD similarity between two strings using zlib compression.
448
449		This is a wrapper for :py:meth:`NCDzlib.sim`.
450
451		Args:
452		src (str): Source string for comparison
453		tar (str): Target string for comparison
454
455		Returns:
456		float: Compression similarity
457
458		Examples:
459		>>> sim_ncd_zlib('cat', 'hat')
460		0.6666666666666667
461		>>> sim_ncd_zlib('Niall', 'Neil')
462		0.5454545454545454
463		>>> sim_ncd_zlib('aluminum', 'Catalan')
464		0.4285714285714286
465		>>> sim_ncd_zlib('ATCG', 'TAGC')
466		0.6
467
468		"""
469	1	return NCDzlib().sim(src, tar)
470
471
472	1	class NCDbz2(Distance):
		0 ignored issues – show Unused Code introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report The variable `__class__` seems to be unused. Loading history...
473		"""Normalized Compression Distance using bz2 compression.
474
475		Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
476		"""
477
478	1	_level = 9
479
480	1	def __init__(self, level=9):
481		"""Initialize zlib compressor.
482
483		Args:
484		level (int): The compression level (0 to 9)
485		"""
486	1	self._level = level
487
488	1	def dist(self, src, tar):
		0 ignored issues – show Bug introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Parameters differ from overridden 'dist' method Loading history...
489		"""Return the NCD between two strings using bz2 compression.
490
491		Args:
492		src (str): Source string for comparison
493		tar (str): Target string for comparison
494
495		Returns:
496		float: Compression distance
497
498		Examples:
499		>>> cmp = NCDbz2()
500		>>> cmp.dist('cat', 'hat')
501		0.06666666666666667
502		>>> cmp.dist('Niall', 'Neil')
503		0.03125
504		>>> cmp.dist('aluminum', 'Catalan')
505		0.17647058823529413
506		>>> cmp.dist('ATCG', 'TAGC')
507		0.03125
508
509		"""
510	1	if src == tar:
511	1	return 0.0
512
513	1	src = src.encode('utf-8')
514	1	tar = tar.encode('utf-8')
515
516	1	src_comp = bz2.compress(src, self._level)[10:]
517	1	tar_comp = bz2.compress(tar, self._level)[10:]
518	1	concat_comp = bz2.compress(src + tar, self._level)[10:]
519	1	concat_comp2 = bz2.compress(tar + src, self._level)[10:]
520
521	1	return (
522		min(len(concat_comp), len(concat_comp2))
523		- min(len(src_comp), len(tar_comp))
524		) / max(len(src_comp), len(tar_comp))
525
526
527	1	def dist_ncd_bz2(src, tar):
528		"""Return the NCD between two strings using bz2 compression.
529
530		This is a wrapper for :py:meth:`NCDbz2.dist`.
531
532		Args:
533		src (str): Source string for comparison
534		tar (str): Target string for comparison
535
536		Returns:
537		float: Compression distance
538
539		Examples:
540		>>> dist_ncd_bz2('cat', 'hat')
541		0.06666666666666667
542		>>> dist_ncd_bz2('Niall', 'Neil')
543		0.03125
544		>>> dist_ncd_bz2('aluminum', 'Catalan')
545		0.17647058823529413
546		>>> dist_ncd_bz2('ATCG', 'TAGC')
547		0.03125
548
549		"""
550	1	return NCDbz2().dist(src, tar)
551
552
553	1	def sim_ncd_bz2(src, tar):
554		"""Return the NCD similarity between two strings using bz2 compression.
555
556		This is a wrapper for :py:meth:`NCDbz2.sim`.
557
558		Args:
559		src (str): Source string for comparison
560		tar (str): Target string for comparison
561
562		Returns:
563		float: Compression similarity
564
565		Examples:
566		>>> sim_ncd_bz2('cat', 'hat')
567		0.9333333333333333
568		>>> sim_ncd_bz2('Niall', 'Neil')
569		0.96875
570		>>> sim_ncd_bz2('aluminum', 'Catalan')
571		0.8235294117647058
572		>>> sim_ncd_bz2('ATCG', 'TAGC')
573		0.96875
574
575		"""
576	1	return NCDbz2().sim(src, tar)
577
578
579	1	class NCDlzma(Distance):
		0 ignored issues – show Unused Code introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report The variable `__class__` seems to be unused. Loading history...
580		"""Normalized Compression Distance using lzma compression.
581
582		Normalized compression distance (NCD) :cite:`Cilibrasi:2005`.
583		"""
584
585	1	def dist(self, src, tar):
		0 ignored issues – show Bug introduced 2018-11-04 08:02 UTC by Report Bug Copy Issue Report Parameters differ from overridden 'dist' method Loading history...
586		"""Return the NCD between two strings using lzma compression.
587
588		Args:
589		src (str): Source string for comparison
590		tar (str): Target string for comparison
591
592		Returns:
593		float: Compression distance
594
595		Raises:
596		ValueError: Install the PylibLZMA module in order to use lzma
597
598		Examples:
599		>>> cmp = NCDlzma()
600		>>> cmp.dist('cat', 'hat')
601		0.08695652173913043
602		>>> cmp.dist('Niall', 'Neil')
603		0.16
604		>>> cmp.dist('aluminum', 'Catalan')
605		0.16
606		>>> cmp.dist('ATCG', 'TAGC')
607		0.08695652173913043
608
609		"""
610		if src == tar:
611		return 0.0
612
613		src = src.encode('utf-8')
614		tar = tar.encode('utf-8')
615
616		if lzma is not None:
617		src_comp = lzma.compress(src)[14:]
618		tar_comp = lzma.compress(tar)[14:]
619		concat_comp = lzma.compress(src + tar)[14:]
620		concat_comp2 = lzma.compress(tar + src)[14:]
621		else: # pragma: no cover
622		raise ValueError(
623		'Install the PylibLZMA module in order to use lzma'
624		)
625
626		return (
627		min(len(concat_comp), len(concat_comp2))
628		- min(len(src_comp), len(tar_comp))
629		) / max(len(src_comp), len(tar_comp))
630
631
632	1	def dist_ncd_lzma(src, tar):
633		"""Return the NCD between two strings using lzma compression.
634
635		This is a wrapper for :py:meth:`NCDlzma.dist`.
636
637		Args:
638		src (str): Source string for comparison
639		tar (str): Target string for comparison
640
641		Returns:
642		float: Compression distance
643
644		Examples:
645		>>> dist_ncd_lzma('cat', 'hat')
646		0.08695652173913043
647		>>> dist_ncd_lzma('Niall', 'Neil')
648		0.16
649		>>> dist_ncd_lzma('aluminum', 'Catalan')
650		0.16
651		>>> dist_ncd_lzma('ATCG', 'TAGC')
652		0.08695652173913043
653
654		"""
655		return NCDlzma().dist(src, tar)
656
657
658	1	def sim_ncd_lzma(src, tar):
659		"""Return the NCD similarity between two strings using lzma compression.
660
661		This is a wrapper for :py:meth:`NCDlzma.sim`.
662
663		Args:
664		src (str): Source string for comparison
665		tar (str): Target string for comparison
666
667		Returns:
668		float: Compression similarity
669
670		Examples:
671		>>> sim_ncd_lzma('cat', 'hat')
672		0.9130434782608696
673		>>> sim_ncd_lzma('Niall', 'Neil')
674		0.84
675		>>> sim_ncd_lzma('aluminum', 'Catalan')
676		0.84
677		>>> sim_ncd_lzma('ATCG', 'TAGC')
678		0.9130434782608696
679
680		"""
681		return NCDlzma().sim(src, tar)
682
683
684		if __name__ == '__main__':
685		import doctest
686
687		doctest.testmod()
688

chrislit / abydos

Pull Request — master (#141)

abydos.distance._compression.sim_ncd_bz2() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like