One day I found myself in need of being able to extract the bits-n-pieces of sudoku grids that are latin friendly... or is that latin bits that are sudoku friendly? You know, the pieces that looks like this:
X.. ... ... ... X.. ... ... ... .X. ... ... ..X ... .X. ... ..X ... ... ... ... X.. .X. ... ... ... ..X ...
And I wanted all of them. Seriously. All of them. I was looking for a certain pattern because my curiousity burns like that. Besides, why not? Sounds like fun.
After a pint of 90 proof and banging my head despite the ringing in it, I came up with a way to find all of them. And the best part is that it is not hit-or-miss at all. Rather, I interate with absolute certaintly.
There are 46,656 of them, and this is what the first ten look like:
100000000000100000000000100010000000000010000000000010001000000000001000000000001 100000000000100000000000100010000000000010000000000001001000000000001000000000010 100000000000100000000000010010000000000010000000000100001000000000001000000000001 100000000000100000000000010010000000000010000000000001001000000000001000000000100 100000000000100000000000001010000000000010000000000100001000000000001000000000010 100000000000100000000000001010000000000010000000000010001000000000001000000000100 100000000000100000000000100010000000000001000000000010001000000000010000000000001 100000000000100000000000100010000000000001000000000001001000000000010000000000010 100000000000100000000000010010000000000001000000000100001000000000010000000000001 100000000000100000000000010010000000000001000000000001001000000000010000000000100 ...
Man, now that is beautiful. Absolutely beautiful.
Anybody want to try to exact cover 46,656 lines of 0's and 1's to generate a sudoku grid? Heh.
Anyways, text file zipped up. Ready to be parsed and played with.