Get the latest tech news
Solving Boolean satisfiability and integer programming with Python packaging
You can try the solvers I developed using the methods described below at pipip. Recently, I was quite amused to see this GitHub project: sudoku-in-python-packaging.
Basically you can formulate a Sudoku puzzle as a Python package dependency resolution problem and then call pip (or uv) to solve it. Note that by IP here I mean the decision version, where the goal is to determine if there exists an integer solution that satisfies a given set of linear inequalities. A slightly more interesting example is the file examples/ip/003.txt, which contains the following IP instance: \[ x_1 + x_2 - x_3 \leq 2 \] \[ -x_1 + x_2 + x_3 \leq 1 \] The solver correctly returns feasible, and a solution is \(x_1 = 1, x_2 = 1, x_3 = 1\).
Or read this on Hacker News