python-boolean

A few neat functions under a liberal license to read in a string representing a Boolean expression and return various handy forms.

It uses pyparsing to do the heavy lifting of the parsing, and uses a Quine-McCluskey algorithm written by George Prekas based on code by Robert Dick and Pat Maupin (wholly contained within qm.py).

Mostly the functions of interest are within logic.py and are self explanatory with examples, but the interesting stuff is as below.

Functions

parse_boolean_function_string(boolean_string, boolean_variables)

Parses the string using the passed iterable of boolean variables and constructs a callable that implements the boolean function that the string describes.

The four logical operands that are allowed, in order of precedence, are:

• NOT: 'NOT', '~', '!'
• AND: 'AND', '.', '&'
• OR: 'OR', '+', '|'
• XOR: 'XOR'

The word form of each of the above is not case dependent.

Parantheses can be used to explicitly denote precedence.

The callable takes a dictionary of variables to their logical values and returns the result of applying the boolean function to those variables.

>>> func = parse_boolean_function_string('A XOR B', ['A', 'B'])
>>> func({'A':True, 'B':False})
True
>>> func({'A':True, 'B':True})
False

>>> func = parse_boolean_function_string('not (A and B)', ['A', 'B'])
>>> func({'A':False, 'B':False})
True
>>> func({'A':True, 'B':True})
False
>>> func({'A':False, 'B':True})
True

>>> func = parse_boolean_function_string('A.B+~B', ['A', 'B'])
>>> print(func)
((A . B) + ~B)
>>> func({'A':True, 'B':False})
True
>>> func({'A':False, 'B':False})
True
>>> func({'A':False, 'B':True})
False

get_minimal_sop_from_string(boolean_string, boolean_variables)

Get the minimal sum of products as a list of lists which correspond to the sum and products respectively.

Each entry in the inner list corresponds to a boolean variable given in the boolean_variables argument, and in the same order.

The value of each entry in the inner list is either True, False or None, according to whether that variable should be True, False or "Don't care" in generating a True output.

An empty string will return all "Don't care".

>>> get_minimal_sop_from_string('A AND B', ['A', 'B', 'C'])
[[True, True, None]]
>>> get_minimal_sop_from_string('A OR C', ['A', 'B', 'C'])
[[True, None, None], [None, None, True]]
>>> get_minimal_sop_from_string('', ['A', 'B', 'C'])
[[None, None, None]]

get_canonical_minimal_sop_from_string(boolean_string, boolean_variables)

Returns the full minimal canonical form for all the variables passed, given the string, or simply '0' if the string always evaluates to False.

This function is used by get_minimal_sop_from_string.

>>> get_canonical_minimal_sop_from_string('A AND (B OR C)', ['A', 'B', 'C'])
[(3, 4), (5, 2)]