This repo contains small Python modules that are mostly intended for instructional use. I have no immediate intentions of creating a package, but if you want to use any of the code, it is all in the public domain. I recommend the strategy of simply pulling files directly into your own project.
The basic_bool.py module is a finished product that allows you to Pythonically build an AST for a boolean expression. It is less than 100 lines of code.
Its construction API is basically TRUE, FALSE, and SYMBOL,
and then you can build up larger expressions using code like
x & y, a | b, and ~z.
Then the objects themselves allow the following:
x = SYMBOL("x")
y = SYMBOL("y")
str(~y | x) == "(~y)|(x)"
(~y | x).symbols() == {"x", "y"}
(x | y).eval({"x", "y")) == True
(~x).eval({"x")) == False
(y).eval({"x")) == False
The code is very reliable due to its simplicity, and you can read test_basic.py to see its basic operation.
The code has no optimizations, but the performance should be fine for most use cases. (When you evaluate AND and OR pairs, Python can get lucky and be able to avoid traversing the second subtree.)
The solver.py module is a super-tiny module that finds all satisfying variable assignments for a boolean expression. It does this completely by brute force, and the underlying code is equivalent to just building out a truth table. This code is mostly intended for testing.
Everything else is still in progress. (I am building code to simplify boolean expressions.)
If you have interest in turning some of this code into a package, then please feel free to fork it. The licence puts you under no obligation to give me credit, but I would be grateful for a short mention in a readme file. Thanks!