[Error-Handling]: instructions are unclear, and wrong programs pass the tests

I just did the error-handling exercise, and had several issues:

• It was unclear at first what to do with inputs. Looking at the tests, it seemed that trying to convert the input to int was the right thing to try, but this is not mentioned in the readme or the stubs.
• The test for whether a method throws an exception will pass any error that might occur, not just if it’s intentionally raised:

def handle_error_by_throwing_exception():
    raise "(╯°□°）╯︵ ┻━┻"  # Raising a string is a TypeError, but passes the test

• When returning a tuple, which element should be the result and which one the status? What should the result element be when it fails? No suggestion
• No mention of advantages or disadvantages of the different kinds of error handling, or what is idiomatic Python.

[cmccandless]

The entire error-handling exercise is a bit problematic in its current implementation. It will likely be replaced with a proper Concept Exercise in v3.

[BethanyG]

@cmccandless – I think this would be good to leave open as a reminder. Otherwise, it might get lost in the shuffle.

Ultimately, we want to log both the error-handling concept exercise issues and a re-write of this practice exercise as its own issue, but I don’t have time atm to do so – so I think the best bet is to leave this pending.