Choose-whens are not tested in sequential order about jsx-control-statements

[AlexGilleran]

Thanks @RacingTadpole - is it always when2? I have a feeling it's actually going in reverse sequential order.

[RacingTadpole]

Yes - it looks like it is, maybe due to this line: https://github.com/AlexGilleran/jsx-control-statements/blob/master/src/chooseStatement.js#L72 . When I remove the reverse() all the tests still pass, and the test above passes too. I'll submit a PR for that.

We could just define the reverse behaviour to be correct, but that's pretty different from what you'd expect from the analogy with "if-elseif-else". (Eg. the original discussion of this included an example a lot like this, with the "WhenBlock1" answer.)

[AlexGilleran]

Yeah agree re: the order.

Any thoughts @texttechne ?

[texttechne]

Ahh, I see my mistakes. The most important one is having missed to test ordering 😃

The second mistake is all about reversed ordering.

• The array of children is reduced from the right side (l.17), so the ordering of when blocks gets reversed.
• This is the reason I intended to revert them back in the mentioned line 72.
• Actually this is wrong, because the ternary expression itself is built up from right to left (l.73), which requires reversed ordering...

So @RacingTadpole is completely right with his analysis and bugfix. 👍

[texttechne]

We should leave this issue open until it is released.

[AlexGilleran]

Released as v3.1.1 :)

[texttechne]

👍