The test structure of both sample tests and actual tests need to be modified: it's calling user solution twice per test for no reasons.

The number of random tests is not enough; 20 gives some chances for slightly incorrect solutions to pass by chance.

Isn't this a duplicate?

Why this isn't already published?

I've translated this kata into Python. I dont have a clue of how merge works here, so here is the link and I guess it needs your aproval. Greetings

Glad you like it. :] In MarkDown, a couple of spaces can go a long way.

Good luck on the companion kata!

Great, the description looks much better now.

And thanks for the tips. I'll try the other alignments at some point.

Note that the description is in GitHub Flavoured MarkDown, not HTML - though you can use HTML tags. You had forgotten a line break before the last assumption, and then I redid the lists markup a bit also.

I don't know for certain how many ratings and rankings you need. Not every rating or ranking counts the same, but five seems low. Nothing you can do about that anyway, just wait. Resolve issues if and when they crop up. Best thing you can probably do is publish high quality Align right and Justify kata, so people will solve all of them. Just remember, you want to avoid "Unsatisfied" ratings at all cost, so try to get everything right before you publish, and resolve issues sooner rather than later, because people might not come back to adjust their rating.

Excellent, thanks very much for your help!

I've made the suggested changes re specifying alphanumeric characters (and underscores).

Yes, I was expecting your solution might be significantly more elegant than mine.... Very nice.. I've rearranged my comments so that they fit better - let me know if they're still too wide. I'll try to make my solution shorter when I've got time.

So the kata needs to get five approval votes now?

.

May I say your example solution is looooong? If code needs that much commentary, is't not self-explantory enough. Try to fit it on one page. I'd like to read it, but you're not making it easy.

.

Good job, looks like. :]

Any added hyphens must be at least two letters from both the start and the end of the word being broken

(my emphasis)

You're breaking on numeric digits as well. Not a problem, but it should be specified correctly. You're apparently treating digits (and _) as letters, which is subtly different from the current wording.

I'll try to update my solution for the new rules. I agree the new rules make sense and are better, but did you really have to do that? :P

Hi!

So I think I've sorted out the problems you were having last week... I'd be very grateful if you could take another look.

I've taken the bleeping diacritics out. :-)

I've formatted the submit tests in the same way as the example tests, so that they print the expected results and the results given.

I've sorted out the problems with the tests with 'niche' and 'things' that you pointed out. In doing so, I've also added a new rule to account for words that are already hyphenated, and then altered my code to account for this: lines can now be broken after hyphens in words that are already hyphenated, if appropriate.

I've added a couple more tests to the example tests, one to test hyphenated words are being processed correctly, and one to test that spaced en-dashes are being processed correctly.

And I've carefully manually checked that all the tests, including the random tests, are using the right expected answers.

So I'm really hoping that everything is fixed now!

Let me know how it goes...

Cheers