Love this

Not at all, feel free to rewrite.

For which language? I just clicked on "Show Test Cases" and the code that people complained about is commented out:

# Possible nasty rounding issues around 4:24:59 vs 4:25:00
# Test.expect(clock_hands(17.5) == ["12:03:10", "01:08:38", 
#   "02:14:05", "03:19:32", "04:25:00", "05:30:27", "06:35:54", 
#   "07:41:21", "08:46:49", "09:52:16", "10:57:43"])

What rounding problem do you see?

Actually by adding a test case that N=0 golfers should be handled, it broke every single one of the nine solutions people submitted, so I think I should go back to not handling that at all.

Wow, thank you so much for your feedback. I have corrected the test case typos, and clarified the behavior for emptiness and even added a visibile test case for it.

As for observation (3) are you suggestion the problem description calls out a requirement that the sizes of the groups must be the same every day?

Hello, still new on authoring. Looking for rank assessments and additional tags on this one, too. I hope it's found to be fun. This is just the validation part of an algorithm which Wolfram MathWorld said is an unsolved problem. I gave an initial assessment at 6th kyu. Python has some fancy functions that help here; without them I think it would be much higher. Looking for opinions, though. :)

I agree that the test cases are wrong. There are four that fail, expecting nil, that should have a clear winner.

The interpretation is actually

• 1st place has score 1
• 2nd place - there is no 2nd place, it's a tie for first, with score 1
• 3rd place has score 2
• 4th place - there is no 4th place, it's a tie for third, with score 2
• 5th place - there is no 5th place, it's a tie for third, with score 2
• 6th place has score 6

This is proving to be difficult. The response "DO NOT TRY TO USE RETURN" was a little misleading. Maybe it is supposed to be?

Yes, rounding is very problematic, which makes me think tests should be more tolerant, perhaps. One example is that for an angle of 17.5, if you compute (seconds*43200.0)/11.0 you will eventually end up at 04:24:59.999999 which rounds down to 04:24:59, while if you do seconds*(43200.0/11.0) you will get 04:25:00. So integers-only is correct for the test. Staying with integers and doing the divs and mods by 60 is, I think, better. The 43200.0/11.0 is fun because it is a one-liner. Yes I did approved the translations so have fun with those.