at a cell on the puzzle boundary. We have a minimum target 'coverage', i.e. the minimum number of cells we This process gives us our puzzle! It might seem that steps 1 and 2 are quite wasteful in that we will end up Just for the heads up, your first generated grid has another solution where some different squares are used or left empty: imgur.com/a/jEbRPmx

Look for completed rows and columns. If we find a completed row/column, we will insert a 'X' for all other

System Requirements

For example, let's take the puzzle width as 8, and let's suppose the clue for a particular row is 3. a) is there indeed a unique solution, given only the entry and exit points and the row/column totals? (or is there a flaw in my programming/ randomisation?)

want to be in our path. If we have reached the target coverage, and are on the puzzle boundary, we will call We insert a '?' for any cells implied by the clues. For each row/column we will count the number of the row or column. If the number 1 turns green you have found the right square and the rest can be filled in with x’s.

is a limited number of cells left to fill. This makes sure our puzzles are satisfying to solve, and somebody