Skip to content
Navigation Menu
{{ message }}
forked from ilanschnell/bitarray
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathutil.py
More file actions
354 lines (291 loc) · 11.7 KB
/
Copy pathutil.py
File metadata and controls
354 lines (291 loc) · 11.7 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
# Copyright (c) 2019 - 2025, Ilan Schnell; All Rights Reserved
# bitarray is published under the PSF license.
#
# Author: Ilan Schnell
"""
Useful utilities for working with bitarrays.
"""
from __future__ import absolute_import
import os
import sys
from bitarray import bitarray, bits2bytes
from bitarray._util import (
zeros, ones, count_n, parity, xor_indices,
count_and, count_or, count_xor, any_and, subset,
correspond_all, byteswap,
serialize, deserialize,
ba2hex, hex2ba,
ba2base, base2ba,
sc_encode, sc_decode,
vl_encode, vl_decode,
canonical_decode,
)
__all__ = [
'zeros', 'ones', 'urandom',
'pprint', 'strip', 'count_n',
'parity', 'xor_indices',
'count_and', 'count_or', 'count_xor', 'any_and', 'subset',
'correspond_all', 'byteswap', 'intervals',
'ba2hex', 'hex2ba',
'ba2base', 'base2ba',
'ba2int', 'int2ba',
'serialize', 'deserialize',
'sc_encode', 'sc_decode',
'vl_encode', 'vl_decode',
'huffman_code', 'canonical_huffman', 'canonical_decode',
]
def urandom(__length, endian=None):
"""urandom(length, /, endian=None) -> bitarray
Return a bitarray of `length` random bits (uses `os.urandom`).
"""
a = bitarray(os.urandom(bits2bytes(__length)), endian)
del a[__length:]
return a
def pprint(__a, stream=None, group=8, indent=4, width=80):
"""pprint(bitarray, /, stream=None, group=8, indent=4, width=80)
Prints the formatted representation of object on `stream` (which defaults
to `sys.stdout`). By default, elements are grouped in bytes (8 elements),
and 8 bytes (64 elements) per line.
Non-bitarray objects are printed by the standard library
function `pprint.pprint()`.
"""
if stream is None:
stream = sys.stdout
if not isinstance(__a, bitarray):
import pprint as _pprint
_pprint.pprint(__a, stream=stream, indent=indent, width=width)
return
group = int(group)
if group < 1:
raise ValueError('group must be >= 1')
indent = int(indent)
if indent < 0:
raise ValueError('indent must be >= 0')
width = int(width)
if width <= indent:
raise ValueError('width must be > %d (indent)' % indent)
gpl = (width - indent) // (group + 1) # groups per line
epl = group * gpl # elements per line
if epl == 0:
epl = width - indent - 2
type_name = type(__a).__name__
# here 4 is len("'()'")
multiline = len(type_name) + 4 + len(__a) + len(__a) // group >= width
if multiline:
quotes = "'''"
elif __a:
quotes = "'"
else:
quotes = ""
stream.write("%s(%s" % (type_name, quotes))
for i, b in enumerate(__a):
if multiline and i % epl == 0:
stream.write('\n%s' % (indent * ' '))
if i % group == 0 and i % epl != 0:
stream.write(' ')
stream.write(str(b))
if multiline:
stream.write('\n')
stream.write("%s)\n" % quotes)
stream.flush()
def strip(__a, mode='right'):
"""strip(bitarray, /, mode='right') -> bitarray
Return a new bitarray with zeros stripped from left, right or both ends.
Allowed values for mode are the strings: `left`, `right`, `both`
"""
if not isinstance(mode, str):
raise TypeError("str expected for mode, got '%s'" %
type(__a).__name__)
if mode not in ('left', 'right', 'both'):
raise ValueError("mode must be 'left', 'right' or 'both', got %r" %
mode)
start = None if mode == 'right' else __a.find(1)
if start == -1:
return __a[:0]
stop = None if mode == 'left' else __a.find(1, right=1) + 1
return __a[start:stop]
def intervals(__a):
"""intervals(bitarray, /) -> iterator
Compute all uninterrupted intervals of 1s and 0s, and return an
iterator over tuples `(value, start, stop)`. The intervals are guaranteed
to be in order, and their size is always non-zero (`stop - start > 0`).
"""
try:
value = __a[0] # value of current interval
except IndexError:
return
n = len(__a)
stop = 0 # "previous" stop - becomes next start
while stop < n:
start = stop
# assert __a[start] == value
try: # find next occurrence of opposite value
stop = __a.index(not value, start)
except ValueError:
stop = n
yield int(value), start, stop
value = not value # next interval has opposite value
def ba2int(__a, signed=False):
"""ba2int(bitarray, /, signed=False) -> int
Convert the given bitarray to an integer.
The bit-endianness of the bitarray is respected.
`signed` indicates whether two's complement is used to represent the integer.
"""
if not isinstance(__a, bitarray):
raise TypeError("bitarray expected, got '%s'" % type(__a).__name__)
length = len(__a)
if length == 0:
raise ValueError("non-empty bitarray expected")
if __a.padbits:
pad = zeros(__a.padbits, __a.endian)
__a = __a + pad if __a.endian == "little" else pad + __a
res = int.from_bytes(__a.tobytes(), byteorder=__a.endian)
if signed and res >> length - 1:
res -= 1 << length
return res
def int2ba(__i, length=None, endian=None, signed=False):
"""int2ba(int, /, length=None, endian=None, signed=False) -> bitarray
Convert the given integer to a bitarray (with given bit-endianness,
and no leading (big-endian) / trailing (little-endian) zeros), unless
the `length` of the bitarray is provided. An `OverflowError` is raised
if the integer is not representable with the given number of bits.
`signed` determines whether two's complement is used to represent the integer,
and requires `length` to be provided.
"""
if not isinstance(__i, int):
raise TypeError("int expected, got '%s'" % type(__i).__name__)
if length is not None:
if not isinstance(length, int):
raise TypeError("int expected for argument 'length'")
if length <= 0:
raise ValueError("length must be > 0")
if signed:
if length is None:
raise TypeError("signed requires argument 'length'")
m = 1 << length - 1
if not (-m <= __i < m):
raise OverflowError("signed integer not in range(%d, %d), "
"got %d" % (-m, m, __i))
if __i < 0:
__i += 1 << length
else: # unsigned
if length and __i >> length:
raise OverflowError("unsigned integer not in range(0, %d), "
"got %d" % (1 << length, __i))
a = bitarray(0, endian)
b = __i.to_bytes(bits2bytes(__i.bit_length()), byteorder=a.endian)
a.frombytes(b)
le = a.endian == 'little'
if length is None:
return strip(a, 'right' if le else 'left') if a else a + '0'
if len(a) > length:
return a[:length] if le else a[-length:]
if len(a) == length:
return a
# len(a) < length, we need padding
pad = zeros(length - len(a), a.endian)
return a + pad if le else pad + a
# ------------------------------ Huffman coding -----------------------------
def _huffman_tree(__freq_map):
"""_huffman_tree(dict, /) -> Node
Given a dict mapping symbols to their frequency, construct a Huffman tree
and return its root node.
"""
from heapq import heappush, heappop
class Node(object):
"""
There are to tyes of Node instances (both have 'freq' attribute):
* leaf node: has 'symbol' attribute
* parent node: has 'child' attribute (tuple with both children)
"""
def __lt__(self, other):
# heapq needs to be able to compare the nodes
return self.freq < other.freq
minheap = []
# create all leaf nodes and push them onto the queue
for sym, f in __freq_map.items():
leaf = Node()
leaf.symbol = sym
leaf.freq = f
heappush(minheap, leaf)
# repeat the process until only one node remains
while len(minheap) > 1:
# take the two nodes with lowest frequencies from the queue
# to construct a new parent node and push it onto the queue
parent = Node()
parent.child = heappop(minheap), heappop(minheap)
parent.freq = parent.child[0].freq + parent.child[1].freq
heappush(minheap, parent)
# the single remaining node is the root of the Huffman tree
return minheap[0]
def huffman_code(__freq_map, endian=None):
"""huffman_code(dict, /, endian=None) -> dict
Given a frequency map, a dictionary mapping symbols to their frequency,
calculate the Huffman code, i.e. a dict mapping those symbols to
bitarrays (with given bit-endianness). Note that the symbols are not limited
to being strings. Symbols may be any hashable object.
"""
if not isinstance(__freq_map, dict):
raise TypeError("dict expected, got '%s'" % type(__freq_map).__name__)
if len(__freq_map) < 2:
if len(__freq_map) == 0:
raise ValueError("cannot create Huffman code with no symbols")
# Only one symbol: Normally if only one symbol is given, the code
# could be represented with zero bits. However here, the code should
# be at least one bit for the .encode() and .decode() methods to work.
# So we represent the symbol by a single code of length one, in
# particular one 0 bit. This is an incomplete code, since if a 1 bit
# is received, it has no meaning and will result in an error.
sym = list(__freq_map)[0]
return {sym: bitarray('0', endian)}
result = {}
def traverse(nd, prefix=bitarray(0, endian)):
try: # leaf
result[nd.symbol] = prefix
except AttributeError: # parent, so traverse each child
traverse(nd.child[0], prefix + '0')
traverse(nd.child[1], prefix + '1')
traverse(_huffman_tree(__freq_map))
return result
def canonical_huffman(__freq_map):
"""canonical_huffman(dict, /) -> tuple
Given a frequency map, a dictionary mapping symbols to their frequency,
calculate the canonical Huffman code. Returns a tuple containing:
0. the canonical Huffman code as a dict mapping symbols to bitarrays
1. a list containing the number of symbols of each code length
2. a list of symbols in canonical order
Note: the two lists may be used as input for `canonical_decode()`.
"""
if not isinstance(__freq_map, dict):
raise TypeError("dict expected, got '%s'" % type(__freq_map).__name__)
if len(__freq_map) < 2:
if len(__freq_map) == 0:
raise ValueError("cannot create Huffman code with no symbols")
# Only one symbol: see note above in huffman_code()
sym = list(__freq_map)[0]
return {sym: bitarray('0', 'big')}, [0, 1], [sym]
code_length = {} # map symbols to their code length
def traverse(nd, length=0):
# traverse the Huffman tree, but (unlike in huffman_code() above) we
# now just simply record the length for reaching each symbol
try: # leaf
code_length[nd.symbol] = length
except AttributeError: # parent, so traverse each child
traverse(nd.child[0], length + 1)
traverse(nd.child[1], length + 1)
traverse(_huffman_tree(__freq_map))
# We now have a mapping of symbols to their code length, which is all we
# need to construct a list of tuples (symbol, code length) sorted by
# code length:
table = sorted(code_length.items(), key=lambda item: item[1])
maxbits = table[-1][1]
codedict = {}
count = (maxbits + 1) * [0]
code = 0
for i, (sym, length) in enumerate(table):
codedict[sym] = int2ba(code, length, 'big')
count[length] += 1
if i + 1 < len(table):
code += 1
code <<= table[i + 1][1] - length
return codedict, count, [item[0] for item in table]
You can’t perform that action at this time.