This library provides an object type which efficiently represents an array of booleans. Bitarrays are sequence types and behave very much like usual lists. Eight bits are represented by one byte in a contiguous block of memory. The user can select between two representations: little-endian and big-endian. All functionality is implemented in C. Methods for accessing the machine representation are provided, including the ability to import and export buffers. This allows creating bitarrays that are mapped to other objects, including memory-mapped files.
In 2024 (probably around August), we are planning the release of bitarray 3.0. The 3.0 release will:
- Remove Python 2.7 support.
- Rename
.itersearch()
to.search()
and.iterdecode()
to.decode()
(and remove their non-iterator counterpart). - Remove
util.rindex()
, use.index(..., right=1)
instead - Remove
util.make_endian()
, usebitarray(..., endian=...)
instead - Remove hackish support for
bitarray()
handling unpickling, see detailed explaination in #207. This will close #206.
- The bit-endianness can be specified for each bitarray object, see below.
- Sequence methods: slicing (including slice assignment and deletion),
operations
+
,*
,+=
,*=
, thein
operator,len()
- Bitwise operations:
~
,&
,|
,^
,<<
,>>
(as well as their in-place versions&=
,|=
,^=
,<<=
,>>=
). - Fast methods for encoding and decoding variable bit length prefix codes.
- Bitarray objects support the buffer protocol (both importing and exporting buffers).
- Packing and unpacking to other binary data formats, e.g.
numpy.ndarray
. - Pickling and unpickling of bitarray objects.
- Immutable
frozenbitarray
objects which are hashable - Sequential search
- Type hinting
- Extensive test suite with about 500 unittests.
- Utility module
bitarray.util
:- conversion to and from hexadecimal strings
- (de-) serialization
- pretty printing
- conversion to and from integers
- creating Huffman codes
- compression of sparse bitarrays
- various count functions
- other helpful functions
Python wheels are are available on PyPI for all mayor platforms and Python versions. Which means you can simply:
$ pip install bitarray
In addition, conda packages are available (both the default Anaconda repository as well as conda-forge support bitarray):
$ conda install bitarray
Once you have installed the package, you may want to test it:
$ python -c 'import bitarray; bitarray.test()'
bitarray is installed in: /Users/ilan/bitarray/bitarray
bitarray version: 2.9.2
sys.version: 3.10.14 (main, Oct 25 2022) [Clang 16.0.6]
sys.prefix: /Users/ilan/miniforge3
pointer size: 64 bit
sizeof(size_t): 8
sizeof(bitarrayobject): 80
HAVE_BUILTIN_BSWAP64: 1
default bit-endianness: big
machine byte-order: little
DEBUG: 0
.........................................................................
.........................................................................
................................................................
----------------------------------------------------------------------
Ran 502 tests in 0.187s
OK
The test()
function is part of the API. It will return
a unittest.runner.TextTestResult
object, such that one can verify that
all tests ran successfully by:
import bitarray
assert bitarray.test().wasSuccessful()
As mentioned above, bitarray objects behave very much like lists, so there is not too much to learn. The biggest difference from list objects (except that bitarray are obviously homogeneous) is the ability to access the machine representation of the object. When doing so, the bit-endianness is of importance; this issue is explained in detail in the section below. Here, we demonstrate the basic usage of bitarray objects:
>>> from bitarray import bitarray
>>> a = bitarray() # create empty bitarray
>>> a.append(1)
>>> a.extend([1, 0])
>>> a
bitarray('110')
>>> x = bitarray(2 ** 20) # bitarray of length 1048576 (initialized to 0)
>>> len(x)
1048576
>>> bitarray('1001 011') # initialize from string (whitespace is ignored)
bitarray('1001011')
>>> lst = [1, 0, False, True, True]
>>> a = bitarray(lst) # initialize from iterable
>>> a
bitarray('10011')
>>> a[2] # indexing a single item will always return an integer
0
>>> a[2:4] # whereas indexing a slice will always return a bitarray
bitarray('01')
>>> a[2:3] # even when the slice length is just one
bitarray('0')
>>> a.count(1)
3
>>> a.remove(0) # removes first occurrence of 0
>>> a
bitarray('1011')
Like lists, bitarray objects support slice assignment and deletion:
>>> a = bitarray(50)
>>> a.setall(0) # set all elements in a to 0
>>> a[11:37:3] = 9 * bitarray('1')
>>> a
bitarray('00000000000100100100100100100100100100000000000000')
>>> del a[12::3]
>>> a
bitarray('0000000000010101010101010101000000000')
>>> a[-6:] = bitarray('10011')
>>> a
bitarray('000000000001010101010101010100010011')
>>> a += bitarray('000111')
>>> a[9:]
bitarray('001010101010101010100010011000111')
In addition, slices can be assigned to booleans, which is easier (and faster) than assigning to a bitarray in which all values are the same:
>>> a = 20 * bitarray('0')
>>> a[1:15:3] = True
>>> a
bitarray('01001001001001000000')
This is easier and faster than:
>>> a = 20 * bitarray('0')
>>> a[1:15:3] = 5 * bitarray('1')
>>> a
bitarray('01001001001001000000')
Note that in the latter we have to create a temporary bitarray whose length must be known or calculated. Another example of assigning slices to Booleans, is setting ranges:
>>> a = bitarray(30)
>>> a[:] = 0 # set all elements to 0 - equivalent to a.setall(0)
>>> a[10:25] = 1 # set elements in range(10, 25) to 1
>>> a
bitarray('000000000011111111111111100000')
As of bitarray version 2.8, indices may also be lists of arbitrary indices (like in NumPy), or bitarrays that are treated as masks, see Bitarray indexing.
Bitarray objects support the bitwise operators ~
, &
, |
, ^
,
<<
, >>
(as well as their in-place versions &=
, |=
, ^=
,
<<=
, >>=
). The behavior is very much what one would expect:
>>> a = bitarray('101110001')
>>> ~a # invert
bitarray('010001110')
>>> b = bitarray('111001011')
>>> a ^ b
bitarray('010111010')
>>> a &= b
>>> a
bitarray('101000001')
>>> a <<= 2 # in-place left shift by 2
>>> a
bitarray('100000100')
>>> b >> 1
bitarray('011100101')
The C language does not specify the behavior of negative shifts and of left shifts larger or equal than the width of the promoted left operand. The exact behavior is compiler/machine specific. This Python bitarray library specifies the behavior as follows:
- the length of the bitarray is never changed by any shift operation
- blanks are filled by 0
- negative shifts raise
ValueError
- shifts larger or equal to the length of the bitarray result in bitarrays with all values 0
It is worth noting that (regardless of bit-endianness) the bitarray left
shift (<<
) always shifts towards lower indices, and the right
shift (>>
) always shifts towards higher indices.
Unless explicitly converting to machine representation, using
the .tobytes()
, .frombytes()
, .tofile()
and .fromfile()
methods, as well as using memoryview
, the bit-endianness will have no
effect on any computation, and one can skip this section.
Since bitarrays allows addressing individual bits, where the machine represents 8 bits in one byte, there are two obvious choices for this mapping: little-endian and big-endian.
When dealing with the machine representation of bitarray objects, it is recommended to always explicitly specify the endianness.
By default, bitarrays use big-endian representation:
>>> a = bitarray()
>>> a.endian()
'big'
>>> a.frombytes(b'A')
>>> a
bitarray('01000001')
>>> a[6] = 1
>>> a.tobytes()
b'C'
Big-endian means that the most-significant bit comes first.
Here, a[0]
is the lowest address (index) and most significant bit,
and a[7]
is the highest address and least significant bit.
When creating a new bitarray object, the endianness can always be specified explicitly:
>>> a = bitarray(endian='little')
>>> a.frombytes(b'A')
>>> a
bitarray('10000010')
>>> a.endian()
'little'
Here, the low-bit comes first because little-endian means that increasing
numeric significance corresponds to an increasing address.
So a[0]
is the lowest address and least significant bit,
and a[7]
is the highest address and most significant bit.
The bit-endianness is a property of the bitarray object. The endianness cannot be changed once a bitarray object is created. When comparing bitarray objects, the endianness (and hence the machine representation) is irrelevant; what matters is the mapping from indices to bits:
>>> bitarray('11001', endian='big') == bitarray('11001', endian='little')
True
Bitwise operations (|
, ^
, &=
, |=
, ^=
, ~
) are
implemented efficiently using the corresponding byte operations in C, i.e. the
operators act on the machine representation of the bitarray objects.
Therefore, it is not possible to perform bitwise operators on bitarrays
with different endianness.
When converting to and from machine representation, using
the .tobytes()
, .frombytes()
, .tofile()
and .fromfile()
methods, the endianness matters:
>>> a = bitarray(endian='little')
>>> a.frombytes(b'\x01')
>>> a
bitarray('10000000')
>>> b = bitarray(endian='big')
>>> b.frombytes(b'\x80')
>>> b
bitarray('10000000')
>>> a == b
True
>>> a.tobytes() == b.tobytes()
False
As mentioned above, the endianness can not be changed once an object is created. However, you can create a new bitarray with different endianness:
>>> a = bitarray('111000', endian='little')
>>> b = bitarray(a, endian='big')
>>> b
bitarray('111000')
>>> a == b
True
Bitarray objects support the buffer protocol. They can both export their own buffer, as well as import another object's buffer. To learn more about this topic, please read buffer protocol. There is also an example that shows how to memory-map a file to a bitarray: mmapped-file.py
The .encode()
method takes a dictionary mapping symbols to bitarrays
and an iterable, and extends the bitarray object with the encoded symbols
found while iterating. For example:
>>> d = {'H':bitarray('111'), 'e':bitarray('0'),
... 'l':bitarray('110'), 'o':bitarray('10')}
...
>>> a = bitarray()
>>> a.encode(d, 'Hello')
>>> a
bitarray('111011011010')
Note that the string 'Hello'
is an iterable, but the symbols are not
limited to characters, in fact any immutable Python object can be a symbol.
Taking the same dictionary, we can apply the .decode()
method which will
return a list of the symbols:
>>> a.decode(d)
['H', 'e', 'l', 'l', 'o']
>>> ''.join(a.decode(d))
'Hello'
Since symbols are not limited to being characters, it is necessary to return
them as elements of a list, rather than simply returning the joined string.
The above dictionary d
can be efficiently constructed using the function
bitarray.util.huffman_code()
. I also wrote Huffman coding in Python
using bitarray for more
background information.
When the codes are large, and you have many decode calls, most time will
be spent creating the (same) internal decode tree objects. In this case,
it will be much faster to create a decodetree
object, which can be
passed to bitarray's .decode()
and .iterdecode()
methods, instead
of passing the prefix code dictionary to those methods itself:
>>> from bitarray import bitarray, decodetree
>>> t = decodetree({'a': bitarray('0'), 'b': bitarray('1')})
>>> a = bitarray('0110')
>>> a.decode(t)
['a', 'b', 'b', 'a']
>>> ''.join(a.iterdecode(t))
'abba'
The sole purpose of the immutable decodetree
object is to be passed
to bitarray's .decode()
and .iterdecode()
methods.
A frozenbitarray
object is very similar to the bitarray object.
The difference is that this a frozenbitarray
is immutable, and hashable,
and can therefore be used as a dictionary key:
>>> from bitarray import frozenbitarray
>>> key = frozenbitarray('1100011')
>>> {key: 'some value'}
{frozenbitarray('1100011'): 'some value'}
>>> key[3] = 1
Traceback (most recent call last):
...
TypeError: frozenbitarray is immutable
bitarray version: 2.9.2 -- change log
In the following, item
and value
are usually a single bit -
an integer 0 or 1.
Also, sub_bitarray
refers to either a bitarray, or an item
.
bitarray(initializer=0, /, endian='big', buffer=None)
-> bitarrayReturn a new bitarray object whose items are bits initialized from the optional initial object, and endianness. The initializer may be of the following types:
int
: Create a bitarray of given integer length. The initial values are all0
.str
: Create bitarray from a string of0
and1
.iterable
: Create bitarray from iterable or sequence of integers 0 or 1.Optional keyword arguments:
endian
: Specifies the bit-endianness of the created bitarray object. Allowed values arebig
andlittle
(the default isbig
). The bit-endianness effects the buffer representation of the bitarray.buffer
: Any object which exposes a buffer. When provided,initializer
cannot be present (or has to beNone
). The imported buffer may be read-only or writable, depending on the object type.New in version 2.3: optional
buffer
argument.
all()
-> bool- Return True when all bits in bitarray are True.
Note that
a.all()
is faster thanall(a)
. any()
-> bool- Return True when any bit in bitarray is True.
Note that
a.any()
is faster thanany(a)
. append(item, /)
- Append
item
to the end of the bitarray. buffer_info()
-> tupleReturn a tuple containing:
- memory address of buffer
- buffer size (in bytes)
- bit-endianness as a string
- number of pad bits
- allocated memory for the buffer (in bytes)
- memory is read-only
- buffer is imported
- number of buffer exports
bytereverse(start=0, stop=<end of buffer>, /)
For each byte in byte-range(start, stop) reverse bits in-place. The start and stop indices are given in terms of bytes (not bits). Also note that this method only changes the buffer; it does not change the endianness of the bitarray object. Padbits are left unchanged such that two consecutive calls will always leave the bitarray unchanged.
New in version 2.2.5: optional start and stop arguments.
clear()
Remove all items from the bitarray.
New in version 1.4.
copy()
-> bitarray- Return a copy of the bitarray.
count(value=1, start=0, stop=<end>, step=1, /)
-> intNumber of occurrences of
value
bitarray within[start:stop:step]
. Optional argumentsstart
,stop
andstep
are interpreted in slice notation, meaninga.count(value, start, stop, step)
equalsa[start:stop:step].count(value)
. Thevalue
may also be a sub-bitarray. In this case non-overlapping occurrences are counted within[start:stop]
(step
must be 1).New in version 1.1.0: optional start and stop arguments.
New in version 2.3.7: optional step argument.
New in version 2.9: add non-overlapping sub-bitarray count.
decode(code, /)
-> list- Given a prefix code (a dict mapping symbols to bitarrays, or
decodetree
object), decode content of bitarray and return it as a list of symbols. encode(code, iterable, /)
- Given a prefix code (a dict mapping symbols to bitarrays), iterate over the iterable object with symbols, and extend bitarray with corresponding bitarray for each symbol.
endian()
-> str- Return the bit-endianness of the bitarray as a string (
little
orbig
). extend(iterable, /)
- Append all items from
iterable
to the end of the bitarray. If the iterable is a string, each0
and1
are appended as bits (ignoring whitespace and underscore). fill()
-> int- Add zeros to the end of the bitarray, such that the length will be a multiple of 8, and return the number of bits added [0..7].
find(sub_bitarray, start=0, stop=<end>, /, right=False)
-> intReturn lowest (or rightmost when
right=True
) index where sub_bitarray is found, such that sub_bitarray is contained within[start:stop]
. Return -1 when sub_bitarray is not found.New in version 2.1.
New in version 2.9: add optional keyword argument
right
.frombytes(bytes, /)
Extend bitarray with raw bytes from a bytes-like object. Each added byte will add eight bits to the bitarray.
New in version 2.5.0: allow bytes-like argument.
fromfile(f, n=-1, /)
- Extend bitarray with up to
n
bytes read from file objectf
(or any other binary stream what supports a.read()
method, e.g.io.BytesIO
). Each read byte will add eight bits to the bitarray. Whenn
is omitted or negative, all bytes until EOF are read. Whenn
is non-negative but exceeds the data available,EOFError
is raised (but the available data is still read and appended). index(sub_bitarray, start=0, stop=<end>, /, right=False)
-> intReturn lowest (or rightmost when
right=True
) index where sub_bitarray is found, such that sub_bitarray is contained within[start:stop]
. RaisesValueError
when the sub_bitarray is not present.New in version 2.9: add optional keyword argument
right
.insert(index, value, /)
- Insert
value
into bitarray beforeindex
. invert(index=<all bits>, /)
Invert all bits in bitarray (in-place). When the optional
index
is given, only invert the single bit at index.New in version 1.5.3: optional index argument.
iterdecode(code, /)
-> iterator- Given a prefix code (a dict mapping symbols to bitarrays, or
decodetree
object), decode content of bitarray and return an iterator over the symbols. itersearch(sub_bitarray, start=0, stop=<end>, /, right=False)
-> iteratorReturn iterator over indices where sub_bitarray is found, such that sub_bitarray is contained within
[start:stop]
. The indices are iterated in ascending order (from lowest to highest), unlessright=True
, which will iterate in descending oder (starting with rightmost match).New in version 2.9: optional start and stop arguments - add optional keyword argument
right
.pack(bytes, /)
Extend bitarray from a bytes-like object, where each byte corresponds to a single bit. The byte
b'\x00'
maps to bit 0 and all other bytes map to bit 1.This method, as well as the
.unpack()
method, are meant for efficient transfer of data between bitarray objects to other Python objects (for example NumPy's ndarray object) which have a different memory view.New in version 2.5.0: allow bytes-like argument.
pop(index=-1, /)
-> item- Remove and return item at
index
(default last). RaisesIndexError
if index is out of range. remove(value, /)
- Remove the first occurrence of
value
. RaisesValueError
if value is not present. reverse()
- Reverse all bits in bitarray (in-place).
search(sub_bitarray, limit=<none>, /)
-> list- Searches for given sub_bitarray in self, and return list of start positions. The optional argument limits the number of search results to the integer specified. By default, all search results are returned.
setall(value, /)
- Set all elements in bitarray to
value
. Note thata.setall(value)
is equivalent toa[:] = value
. sort(reverse=False)
- Sort all bits in bitarray (in-place).
to01()
-> str- Return a string containing '0's and '1's, representing the bits in the bitarray.
tobytes()
-> bytes- Return the bitarray buffer in bytes (pad bits are set to zero).
tofile(f, /)
- Write byte representation of bitarray to file object f.
tolist()
-> listReturn bitarray as list of integer items.
a.tolist()
is equal tolist(a)
.Note that the list object being created will require 32 or 64 times more memory (depending on the machine architecture) than the bitarray object, which may cause a memory error if the bitarray is very large.
unpack(zero=b'\x00', one=b'\x01')
-> bytes- Return bytes containing one character for each bit in the bitarray, using specified mapping.
Data descriptors were added in version 2.6.
nbytes
-> int- buffer size in bytes
padbits
-> int- number of pad bits
readonly
-> bool- bool indicating whether buffer is read-only
frozenbitarray(initializer=0, /, endian='big', buffer=None)
-> frozenbitarrayReturn a
frozenbitarray
object. Initialized the same way abitarray
object is initialized. Afrozenbitarray
is immutable and hashable, and may therefore be used as a dictionary key.New in version 1.1.
decodetree(code, /)
-> decodetreeGiven a prefix code (a dict mapping symbols to bitarrays), create a binary tree object to be passed to
.decode()
or.iterdecode()
.New in version 1.6.
bits2bytes(n, /)
-> int- Return the number of bytes necessary to store n bits.
get_default_endian()
-> strReturn the default endianness for new bitarray objects being created. Unless
_set_default_endian('little')
was called, the default endianness isbig
.New in version 1.3.
test(verbosity=1)
-> TextTestResult- Run self-test, and return unittest.runner.TextTestResult object.
This sub-module was added in version 1.2.
zeros(length, /, endian=None)
-> bitarray- Create a bitarray of length, with all values 0, and optional endianness, which may be 'big', 'little'.
ones(length, /, endian=None)
-> bitarrayCreate a bitarray of length, with all values 1, and optional endianness, which may be 'big', 'little'.
New in version 2.9.
urandom(length, /, endian=None)
-> bitarrayReturn a bitarray of
length
random bits (usesos.urandom
).New in version 1.7.
pprint(bitarray, /, stream=None, group=8, indent=4, width=80)
Prints the formatted representation of object on
stream
(which defaults tosys.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 functionpprint.pprint()
.New in version 1.8.
make_endian(bitarray, /, endian)
-> bitarrayWhen the endianness of the given bitarray is different from
endian
, return a new bitarray, with endiannessendian
and the same elements as the original bitarray. Otherwise (endianness is alreadyendian
) the original bitarray is returned unchanged.New in version 1.3.
New in version 2.9: deprecated - use
bitarray()
.rindex(bitarray, sub_bitarray=1, start=0, stop=<end>, /)
-> intReturn rightmost (highest) index where sub_bitarray (or item - defaults to 1) is found in bitarray (
a
), such that sub_bitarray is contained withina[start:stop]
. RaisesValueError
when the sub_bitarray is not present.New in version 2.3.0: optional start and stop arguments.
New in version 2.9: deprecated - use
.index(..., right=1)
.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
count_n(a, n, value=1, /)
-> intReturn lowest index
i
for whicha[:i].count(value) == n
. RaisesValueError
whenn
exceeds total count (a.count(value)
).New in version 2.3.6: optional value argument.
parity(a, /)
-> intReturn parity of bitarray
a
.parity(a)
is equivalent toa.count() % 2
but more efficient.New in version 1.9.
count_and(a, b, /)
-> int- Return
(a & b).count()
in a memory efficient manner, as no intermediate bitarray object gets created. count_or(a, b, /)
-> int- Return
(a | b).count()
in a memory efficient manner, as no intermediate bitarray object gets created. count_xor(a, b, /)
-> intReturn
(a ^ b).count()
in a memory efficient manner, as no intermediate bitarray object gets created.This is also known as the Hamming distance.
any_and(a, b, /)
-> boolEfficient implementation of
any(a & b)
.New in version 2.7.
subset(a, b, /)
-> bool- Return
True
if bitarraya
is a subset of bitarrayb
.subset(a, b)
is equivalent toa | b == b
(and equallya & b == a
) but more efficient as no intermediate bitarray object is created and the buffer iteration is stopped as soon as one mismatch is found. intervals(bitarray, /)
-> iteratorCompute 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
).New in version 2.7.
ba2hex(bitarray, /)
-> hexstr- Return a string containing the hexadecimal representation of the bitarray (which has to be multiple of 4 in length).
hex2ba(hexstr, /, endian=None)
-> bitarray- Bitarray of hexadecimal representation. hexstr may contain any number (including odd numbers) of hex digits (upper or lower case).
ba2base(n, bitarray, /)
-> strReturn a string containing the base
n
ASCII representation of the bitarray. Allowed values forn
are 2, 4, 8, 16, 32 and 64. The bitarray has to be multiple of length 1, 2, 3, 4, 5 or 6 respectively. Forn=32
the RFC 4648 Base32 alphabet is used, and forn=64
the standard base 64 alphabet is used.See also: Bitarray representations
New in version 1.9.
base2ba(n, asciistr, /, endian=None)
-> bitarrayBitarray of base
n
ASCII representation. Allowed values forn
are 2, 4, 8, 16, 32 and 64. Forn=32
the RFC 4648 Base32 alphabet is used, and forn=64
the standard base 64 alphabet is used.See also: Bitarray representations
New in version 1.9.
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. int2ba(int, /, length=None, endian=None, signed=False)
-> bitarray- Convert the given integer to a bitarray (with given endianness,
and no leading (big-endian) / trailing (little-endian) zeros), unless
the
length
of the bitarray is provided. AnOverflowError
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 requireslength
to be provided. serialize(bitarray, /)
-> bytesReturn a serialized representation of the bitarray, which may be passed to
deserialize()
. It efficiently represents the bitarray object (including its bit-endianness) and is guaranteed not to change in future releases.See also: Bitarray representations
New in version 1.8.
deserialize(bytes, /)
-> bitarrayReturn a bitarray given a bytes-like representation such as returned by
serialize()
.See also: Bitarray representations
New in version 1.8.
New in version 2.5.0: allow bytes-like argument.
sc_encode(bitarray, /)
-> bytesCompress a sparse bitarray and return its binary representation. This representation is useful for efficiently storing sparse bitarrays. Use
sc_decode()
for decompressing (decoding).See also: Compression of sparse bitarrays
New in version 2.7.
sc_decode(stream)
-> bitarrayDecompress binary stream (an integer iterator, or bytes-like object) of a sparse compressed (
sc
) bitarray, and return the decoded bitarray. This function consumes only one bitarray and leaves the remaining stream untouched. Usesc_encode()
for compressing (encoding).See also: Compression of sparse bitarrays
New in version 2.7.
vl_encode(bitarray, /)
-> bytesReturn variable length binary representation of bitarray. This representation is useful for efficiently storing small bitarray in a binary stream. Use
vl_decode()
for decoding.See also: Variable length bitarray format
New in version 2.2.
vl_decode(stream, /, endian=None)
-> bitarrayDecode binary stream (an integer iterator, or bytes-like object), and return the decoded bitarray. This function consumes only one bitarray and leaves the remaining stream untouched. Use
vl_encode()
for encoding.See also: Variable length bitarray format
New in version 2.2.
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 endianness). Note that the symbols are not limited
to being strings. Symbols may may be any hashable object (such as
None
). canonical_huffman(dict, /)
-> tupleGiven a frequency map, a dictionary mapping symbols to their frequency, calculate the canonical Huffman code. Returns a tuple containing:
- the canonical Huffman code as a dict mapping symbols to bitarrays
- a list containing the number of symbols of each code length
- a list of symbols in canonical order
Note: the two lists may be used as input for
canonical_decode()
.See also: Canonical Huffman Coding
New in version 2.5.
canonical_decode(bitarray, count, symbol, /)
-> iteratorDecode bitarray using canonical Huffman decoding tables where
count
is a sequence containing the number of symbols of each length andsymbol
is a sequence of symbols in canonical order.See also: Canonical Huffman Coding
New in version 2.5.