DESIGN.md

-> main.py - what the user should run

• main() running during runtime /Quoridor -> game.py

• get_available_action() returns the list of available actions

• get_available_states() returns the states resulting from all possible actions (for use by AI)

• take_action() action of user/agent can be inputed

• log() returns a dict with information of the game

-> test.py

• simulate(**kwargs) inputs AI version used for both players; board representation; pathfinding mode; set of moves to play (control the moves made)
• compare(**kwargs) inputs number of games; board representations; pathfinding modes

---

-> bitboard.py -> graph.py

• get_available_action() returns the list of available actions as a boolean array with 128+12 values

First 64 values (063) represent the horizontal wall placements and whether they are possible or not - a1h, b1h and so on Next 64 values (64127) represent the vertical wall placements and whether they are possible or not - a1v, b1v and so on Next 4 values (128131) represent the possible moves by the player - in order of N, E, S, W Next 8 values (132139) reprsent the possible moves by the player - in order of NN, NE, EE, SE, SS, SW, WW, NW

• get_available_states() returns the states resulting from all possible actions (for use by AI)
• display_beautiful() outputs the UNICODE representation to terminal
• take_action() performs the input move - only called when game isn't over
• is_over() returns if the game is over
• winner() returns the winner of the game - only called when game is over

---