Reraised as issue

I made the output format explicit in the description.

I don't see the problem, actually... :o

the sample tests clearly show you're supposed to return a 2D-tuple, so why would you do otherwise?

(maybe the output format should be specified in the descriptionn, tho)

"already approved some time ago"

Closing because there is a more recent issue report on this same issue (which also mentions C# as well). I might consider fixing C and C# versions soon.

?

You could define it as static, then it would have internal linkage.

Post your code here marked as a spoiler.

Same for PHP, I expected to return a JSON string from pass1, but an array was expected.

'y' represents the column... and 'x' represents the line... then, the result of 'line' x 'columns' evaluate the shift to line represented in array. The result plus column represents the right position I want. I would can make a another matrix, but I tried invent.

apart from the base 2, it would be quite easy to implement random tests for the other bases up to 16, without big modifications of the test suite, tho. That would be a good addition to your kata!

mmm... I'm not sure to follow your train of thoughts, here. But one thing seems almost sure, you're missing the good idea that will simplify your problem by a lot. ;o

there are 150 of them... ;s

Warning, you're on the hard path! ;o

Still doable, but really extrem. There is a simpler way to solve it. ;)

I did notice this when implementing the encoder, but I didn't use it :)