Some more thoughts about grouping.
Solutions like def summ(a,b): return a+b and def summ(a,b): return b+a are counted as different. Not sure if it's possible to detect such things, but it would be good I think.
Also for Python functions range() and xrange() are counted as different, though usually Python2 users use xrange() as it had some advantages, and Python3 users use range() cause in Python3 it behaves same way as xrange() in Pyton2, while latter is deprecated. So difference between such soutions is just preferred version of language.

Thinking about it, I assume that you could still consider these two things as different, but add some 2-leveled grouping. So that first you group things like now (really-really same solutions, different only in comments, spacing and var names). And then you count kind of index-of-difference, and group closest groups in upper-groups.

To illustrate: solutions using range() vs xrange() or a+b vs b+a will be in different inner-groups, but in same outer-group.

I know it must be a big thing, so it's more like "an idea you could consider to implement in some distant future", not an actual feature-request like "I do really want it" :)

As far as I understand, I only need to add Python to case switch (line 124)... That's easy :)

Good points. The code grouping gem needs to be updated to handle Haskell comments. We will look into it today. If you know Ruby and want to take a stab at it, please check it out yourself. I'm sure there are a number of improvements that could be made to it.

Good points. The code grouping gem needs to be updated to handle Python comments. We will look into it today. If you know Ruby and want to take a stab at it, please check it out yourself. I'm sure there are a number of improvements that could be made to it.

Hi. Python solution grouping is also not perfect.

There are solutions, that are not grouped because of comments, what if all comments are cleaed for grouping?

Other thing is that two solutions with different variables' names are not grouped. Is it possible to group them?

The Haskell solutions grouping is kind of poor.

Two solutions like (f (g x)) and (f $ g x) come up as different solutions, causing most of the first page to just be the same solution with alternating sets of $'s and ( )'s everywhere.

The other thing I've noticed a lot of is that some users (inc. myself) put a linebreak after module ... where before they start the import list, whereas others do not. These are counting as two different solutions ...

Perhaps desugaring (attempting to inline $) the Haskell and removing all whitespace would be a good way to test equivalence.