Crate cspsolver 

CSP Solver

A constraint satisfaction problem (CSP) solver library.

This library provides tools for solving constraint satisfaction problems, including constraint propagation and search algorithms.

Features

• Constraint propagation
• Backtracking search
• Domain filtering
• Support for various constraint types: +, -, *, /, ==, !=, <, <=, >, >=, all_different
• Type of variables: float, int, mixed (int and float)

Basic Example

use cspsolver::prelude::*;

// Create a new model
let mut model = Model::default();

// Create a variable x in [1, 10]
let v = model.new_var_int(1, 10);

// Add constraint: x > 2.5
model.greater_than(v, float(2.5));

// Solve the problem minimizing x
let solution = model.minimize(v).unwrap();
if let Val::ValI(x) = solution[v] {
    assert_eq!(x, 3);
}

Variables with Predefined Values

use cspsolver::prelude::*;

// Create a new model  
let mut model = Model::default();

// Create variables with specific allowed values
let even_var = model.new_var_with_values(vec![2, 4, 6, 8]);
let odd_var = model.new_var_with_values(vec![1, 3, 5, 7]);

// Add constraint: variables must be different
model.not_equals(even_var, odd_var);

// Solve the problem
let solution = model.solve().unwrap();
if let (Val::ValI(even), Val::ValI(odd)) = (solution[even_var], solution[odd_var]) {
    assert!(even % 2 == 0);  // even number
    assert!(odd % 2 == 1);   // odd number  
    assert_ne!(even, odd);   // different values
}

Modules

model

prelude

Prelude module for CSP solver

solution

vars