v0.2.1

What's Changed

• Set EQUAL_EDGE_COST=false, making diagonal moves cost more than cardinal ones. This reduces path artifacts and typically speeds up search
• Corrected success condition for approximate pathfinding (pathing to neighbor of goal) with EQUAL_EDGE_COST=false
• Added support for loading and running Moving AI benchmarks using the new utility crate grid_pathfinding_benchmark and integrated these into standard benchmarks
  Full Changelog: v0.2.0...v0.2.1

v0.2.0

What's Changed

• Added support for restricting diagonals (#3), making it possible to find paths on grids with and without diagonal neighbors. This can be done by appropriately setting the allow_diagonal_move field of PathingGrid. By default, allow_diagonal_move = true
• Added improved pruning rules toggle using improved_pruning
• Set the heuristic scaling factor to a default of 1.0 instead of 1.2. This means that by default solutions are now optimal. The value can be changed by setting heuristic_factor for faster and sub-optimal search
• Corrected node costs of the neighborhood of a pruned node when using improved pruning, which gave sub-optimal solutions before (#2, #5)
• Rewrote jumping logic to be iterative instead of recursive, increasing performance and avoiding the recursion limit and potential stack overflows on large maps
• Switched to fxhash hashing in the basic A* routine, making hashing faster

Full Changelog: v0.1.2...v0.2.0

v0.1.2

• Removed info! logging in connected component generation and pathfinding to avoid slowdowns when running at a dynamic logging level
• Added comments describing logic in place of these info!'s
• Updated dependencies

v0.1.1

Finished metadata and bumped version

v0.1.0

Made link to examples/ dir