Simple but powerful

Thanks for responding so quickly. Honestly, I think the test case shouldn't deal with case insensitivity, rather the solutinons we code should, however I appreciate you addressing my concern.

By the way, I did really enjoy the challenge. Thanks!

You're solution is almost right, but for the first Light test it failed because it generates the given object :

> groupAnagrams(["rat", "tar", "star"])
[ [ 'rat', 'tar' ], 
  [ 'star' ], 
  undefined ]

The undefined should not be part of the output for the test to succeed.

Cheers

I've fixed the assertSimilarUnsorted to compare lowercase string versions. Thanks for noticing it !

I feel like the tests cases have invalid failures since it's case senistive. As a result "Rats" is not an anagram of "rats" which it most certainly is.

Light and edge tests do not validate my correct solution (tested for different inputs in console).
Here is my solution: http://pastebin.com/LXRnTSTd

I wrote this 4 days ago:

I tried to avoid rounding problems by returning the max speed as an integer... missed! Usually I truncate rather than rounding to the nearest.
In each of the 8 languages when I calculate (1.92-1.47) * 3600 / 20 I find 80.99999999999999 which, when truncated, gives 80 but ((1.92-1.47) / 20) * 3600 gives 81. One knows that the order of operations matters but here we have a difficult case:-) and for the time being I don't know what to do! I did the multiplication first and you did the division first. Maybe the only solution will be to suppress or to modify this case.

Following that I modified the tests in each language (by introducing some "fuzziness") except in Java and Clojure because Codewars didn't let me re-publish Java and Clojure versions. I will try again and if it is not possible to re-publish I will suppress Java and Clojure versions. I'm very sorry for the problems of re-publishing at Codewars!
I understand the case of 81 versus 80 but not 2384 versus 44. Do you have the input for the last case?

PS: I tried again to re-publish but it is impossible.
7 guys passed the tests in Java without problems and 2 in Clojure. Try this order in calculation: (3600 * (x[k + 1] - x[k]) / s) for Java, and (/ (* 3600 (- a b)) (float s)) in Clojure. Tell me if this works.

I was able to submit finally, but the issue seems like it might be intermittent.

Clicking Submit was working then suddenly I started receiving an error, red bar, and the Submit button was grayed out.

Seeing same thing on Clojure side

The Java version seems to have an issue. The math required to convert the 3rd test (resulting in 80) actually results in 81, and the random tests on submission do not seem to be expecting the correct results (expected 2384 actual 44).