phpt

[till]

cc @helgi @cweiske

Wanted to get some feedback. It seems like the phpt in Pyrus is a pretty limited framework/implementation. It's hard to do anything — so it seems and there are lots of gotchas and caveats.

So e.g.: I added to a --SKIPIF-- section and typo'd something in there. Turns out, that when there is a parse error, it doesn't skip or do anything at all. It doesn't even tell you that there is an error. I'm sure there are a ton of other issues which I haven't yet seen.

Maybe we should start accepting phpunit-based testcases?

[saltybeagle]

There are probably quite a few bugs in the PHPT runner. Those bugs can be filed on the Pyrus_Developer package.

To the question of accepting phpunit-based testcases — At one point we had around 70% LoC coverage with the existing phpts, so it seems to me that the test framework is not the problem. I do recognize that some contributors aren't familiar with it, though.

I would prefer phpt and will continue to write my tests using phpt, but have no energy to fight contributions from others. And, PHPUnit can run the existing phpt tests, so I'm not completely against phpunit contributions.

[till]

I think LOC are not the best indicator for good tests though. But ok, if we can do both, that's great.