Every message/string, which needs to be decoded, had been encoded before. So, question to jhancock: how can you get such an encoded message in Paul Cipher, which has small letters in it? In my opinion, there's a contradiction with the task, and that contradiction has to be fixed. Nonetheless, this kata is great!

The most popular 6-kyu kata? Are you kidding me?

I have one problem - I've written my code on C#, it works perfectly on Visual Studio, but Codewars gives me an error. After each line of numbers I add to the string "/n", but Codewars doesn't convert it to a new line and makes it like this "1/n 121/n". I have tried also "/r/n" and System.Environment.NewLine, but it still doesn't work. Have you faced that problem and what should I do?

No, it never enters the else branch.

If there are no vowels, my code returns the sum of all consonants, doesn't it?

Remember to mark your posts as having spoiler content when they do. What happens in your code if there is no vowel?
@KenKamau, check the random testing suite, it throws errors (not always) and it wasn't for the user's code.

System.InvalidOperationException : Sequence contains no elements
 Stack Trace
   at System.Linq.Enumerable.Max(IEnumerable`1 source)
   at Solution.Tests.SolutionSolver(String s)
   at Solution.Tests.RandomTests()

can you post your code? someone will help.

I think my code is without mistakes but every random test returns a wrong answer. However, sample tests are correct) Has anyone met such problem on C#? What to do in order to make my random tests run without returning mistaken values?