blob: 1d87a91e51cbccb56e4ab4da291f7924cd2ebf0f (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
The solver can only solve when at least one cell can be solved with certainty
at each step of the process until repeating this eventually results in a solved
board.
The next step here is to allow exploration/trail of possible solutions.
Presumably, would want to keep a copy of the board just before each "guess"
that's taken to be able to easily roll back if failure.
This would probably work really nicely as a recursive call:
board = get_fresh_board(board)
solve_all_certainties(board)
for each possible board as cand_board:
solution = solve_all_certainties(cand_board)
if complete(solution)
success
/* we should have solution presuming board was solvable */
|
