in reply to Testing: shades of grey

My understanding of TODO tests is they're all expected to fail, and what's unexpected (and reported on) is for them to pass.

As for the "how correct" problem, my approach to that situation would be:

That second one would allow spotting regressions in how correct one's parsing was.

Replies are listed 'Best First'.
Re^2: Testing: shades of grey
by talexb (Chancellor) on Dec 22, 2024 at 16:54 UTC
      My understanding of TODO tests is they're all expected to fail, and what's unexpected (and reported on) is for them to pass.

    I politely disagree with this. The TODO tests are ones that run, and the result is reported, but whether they pass or not is academic. Maybe they passed, and maybe they failed. We don't care -- it's just 'interesting'. :)

    Eventually the stuff that was TODO gets done, and the TODO can be moved into the realm of real tests, and you move on.

    Alex / talexb / Toronto

    For a long time, I had a link in my .sig going to Groklaw. I heard that as of December 2024, this link is dead. Still, thanks to PJ for all your work, we owe you so much. RIP Groklaw -- 2003 to 2013.

[reply]
[d/l]
[select]

      Yes, but in the context of hv's original question, prove does not by default report TODO tests which fail but it does report TODO tests which pass albeit without counting those towards an overall pass/fail of the test file/suite. That's the relevant part, AIUI.

      🦛

[reply]
[d/l]
        Shouldn't verbose mode show them?

        The OP didn't provide us with an SSCCE tho, it's hard to tell what he really needs.

        Cheers Rolf
        (addicted to the Perl Programming Language :)
        see Wikisyntax for the Monastery

[reply]

In Section Seekers of Perl Wisdom