How do I create a constant in Python?

Question

Is there a way to declare a constant in Python? In Java we can create constant values in this manner:

public static final String CONST_NAME = "Name";

What is the equivalent of the above Java constant declaration in Python?

Answer 1

I write a util lib for python const: kkconst - pypi support str, int, float, datetime

the const field instance will keep its base type behavior.

For example:

from __future__ import print_function
from kkconst import (
    BaseConst,
    ConstFloatField,
)

class MathConst(BaseConst):
    PI = ConstFloatField(3.1415926, verbose_name=u"Pi")
    E = ConstFloatField(2.7182818284, verbose_name=u"mathematical constant")  # Euler's number"
    GOLDEN_RATIO = ConstFloatField(0.6180339887, verbose_name=u"Golden Ratio")

magic_num = MathConst.GOLDEN_RATIO
assert isinstance(magic_num, ConstFloatField)
assert isinstance(magic_num, float)

print(magic_num)  # 0.6180339887
print(magic_num.verbose_name)  # Golden Ratio

more details usage you can read the pypi url: pypi or github

Answer 2

We can create a descriptor object:

class Constant:
  def __init__(self,value=None):
    self.value = value
  def __get__(self,instance,owner):
    return self.value
  def __set__(self,instance,value):
    raise ValueError("You can't change a constant")

class A:
  NULL = Constant()
  NUM = Constant(0xFF)

class B:
  NAME = Constant('bar')
  LISTA = Constant([0,1,'INFINITY'])

>>> obj=A()
>>> print(obj.NUM)  #=> 255
>>> obj.NUM =100

Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ValueError: You can't change a constant

Answer 3

You can wrap a constant in a numpy array, flag it write only, and always call it by index zero.

import numpy as np

# declare a constant
CONSTANT = 'hello'

# put constant in numpy and make read only
CONSTANT = np.array([CONSTANT])
CONSTANT.flags.writeable = False
# alternatively: CONSTANT.setflags(write=0)

# call our constant using 0 index    
print 'CONSTANT %s' % CONSTANT[0]

# attempt to modify our constant with try/except
new_value = 'goodbye'
try:
    CONSTANT[0] = new_value
except:
    print "cannot change CONSTANT to '%s' it's value '%s' is immutable" % (
        new_value, CONSTANT[0])

# attempt to modify our constant producing ValueError
CONSTANT[0] = new_value



>>>
CONSTANT hello
cannot change CONSTANT to 'goodbye' it's value 'hello' is immutable
Traceback (most recent call last):
  File "shuffle_test.py", line 15, in <module>
    CONSTANT[0] = new_value
ValueError: assignment destination is read-only

of course this only protects the contents of the numpy, not the variable "CONSTANT" itself; you can still do:

CONSTANT = 'foo'

and CONSTANT would change, however that would quickly throw an TypeError the first time CONSTANT[0] is later called in the script.

although... I suppose if you at some point changed it to

CONSTANT = [1,2,3]

now you wouldn't get the TypeError anymore. hmmmm....

https://docs.scipy.org/doc/numpy/reference/generated/numpy.ndarray.setflags.html

Answer 4

In my case, I needed immutable bytearrays for an implementation of a crypto library containing many literal numbers I wanted to ensure were constant.

This answer works but attempted reassignment of bytearray elements does not raise an error.

def const(func):
    '''implement const decorator'''
    def fset(self, val):
        '''attempting to set a const raises `ConstError`'''
        class ConstError(TypeError):
            '''special exception for const reassignment'''
            pass

        raise ConstError

    def fget(self):
        '''get a const'''
        return func()

    return property(fget, fset)


class Consts(object):
    '''contain all constants'''

    @const
    def C1():
        '''reassignment to C1 fails silently'''
        return bytearray.fromhex('deadbeef')

    @const
    def pi():
        '''is immutable'''
        return 3.141592653589793

Constants are immutable, but constant bytearray assignment fails silently:

>>> c = Consts()
>>> c.pi = 6.283185307179586  # (https://en.wikipedia.org/wiki/Tau_(2%CF%80))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "consts.py", line 9, in fset
    raise ConstError
__main__.ConstError
>>> c.C1[0] = 0
>>> c.C1[0]
222
>>> c.C1
bytearray(b'\xde\xad\xbe\xef')

A more powerful, simple, and perhaps even more 'pythonic' approach involves the use of memoryview objects (buffer objects in <= python-2.6).

import sys

PY_VER = sys.version.split()[0].split('.')

if int(PY_VER[0]) == 2:
    if int(PY_VER[1]) < 6:
        raise NotImplementedError
    elif int(PY_VER[1]) == 6:
        memoryview = buffer

class ConstArray(object):
    '''represent a constant bytearray'''
    def __init__(self, init):
        '''
        create a hidden bytearray and expose a memoryview of that bytearray for
        read-only use
        '''
        if int(PY_VER[1]) == 6:
            self.__array = bytearray(init.decode('hex'))
        else:
            self.__array = bytearray.fromhex(init)

        self.array = memoryview(self.__array)

    def __str__(self):
        return str(self.__array)

    def __getitem__(self, *args, **kwargs):
       return self.array.__getitem__(*args, **kwargs)

ConstArray item assignment is a TypeError:

>>> C1 = ConstArray('deadbeef')
>>> C1[0] = 0
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: 'ConstArray' object does not support item assignment
>>> C1[0]
222

Answer 5

There is a cleaner way to do this with namedtuple:

from collections import namedtuple


def make_consts(name, **kwargs):
    return namedtuple(name, kwargs.keys())(**kwargs)

Usage Example

CONSTS = make_consts("baz1",
                     foo=1,
                     bar=2)

With this exactly approach you can namespace your constants.

Answer 6

I'm probably missing a trick here, but this seems to work for me:

class CONST(object):
    FOO = 1234

    def __setattr__(self, *_):
        pass

CONST = CONST()

#----------

print CONST.FOO    # 1234

CONST.FOO = 4321
CONST.BAR = 5678

print CONST.FOO    # Still 1234!
print CONST.BAR    # Oops AttributeError

Creating the instance allows the magic __setattr__ method to kick in and intercept attempts to set the FOO variable. You could throw an exception here if you wanted to. Instantiating the instance over the class name prevents access directly via the class.

It's a total pain for one value, but you could attach lots to your CONST object. Having an upper class, class name also seems a bit grotty, but I think it's quite succinct overall.