I think you don't need if(dec[i] > 0)

Added a stringizer class.

I see.. thanks!

It's a known problem with testing framework for C++ used by Codewars. See the issue I raised above.

Reported by user pandektis below:

In C++, method return type is not supported by stringizer, what results in very confusing assertion messages. Testing framework needs implementation of a stringizer for std::pair <std::string, std::string> (should suffice) or std::vector<std::pair <std::string, std::string>> to display assertion messages correctly.

This chapter of documentation for Snowhouse framework explains how to add such stringizer class. AFAIK adding similar implementation to preloaded section solved the problem in other kata.

Well I am trying this kata in C++ and I get this:

Expected: equal to [ [unsupported type], [unsupported type], [unsupported type] ]
Actual: [ [unsupported type], [unsupported type], [unsupported type] ]

I print to the output, and it seems ok. Am I missing something?

Edit: Ok, I found out it was an off-by-one error, that got a leading space inserted in there. But [unsupported type] got me curious...

Totally forgot about re.. Nice one! ;)