Date: Mon, 13 Apr 2015 16:45:31 -0400 From: Julio Merino <jmmv@meroh.net> To: Alan Somers <asomers@freebsd.org> Cc: "freebsd-testing@freebsd.org" <freebsd-testing@freebsd.org> Subject: Re: Python unittest backend for Kyua Message-ID: <A84C5AFA-5CC8-4EEF-94CA-80A258C833C6@meroh.net> In-Reply-To: <CAOtMX2hvfpskz2aTy5kwS3ewXPgNLeWeoS3AcSG4EUSK2WTPpA@mail.gmail.com> References: <CAOtMX2g35ybAzFHriVuQqYMGq5Snm0%2BEcUWNhQgEr%2Bqx1xYpcA@mail.gmail.com> <20150214215750.GA5065@rodrigc-laptop1> <CAOtMX2jFi89wt1fC5OAAGZhEDHKpLZza=tAXU7WM5y-pwL4Yqw@mail.gmail.com> <C288DCCA-6031-4C03-8163-8826FBCE12A7@meroh.net> <CAOtMX2hvfpskz2aTy5kwS3ewXPgNLeWeoS3AcSG4EUSK2WTPpA@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[Here I am apologizing for delays again... that's it, I'm re-enabling email notifications for my OSS email account; splitting it off didn't work as intended.] On Mar 2, 2015, at 19:07, Alan Somers <asomers@freebsd.org> wrote: > Well, maybe I should dust off my old branch and get back to work. But > I don't know how my python tester would work with your executor > branch. If I understand correctly, the executor branch dispenses with > the *-tester binaries and calls the test programs directly from the > kyua executable, correct? The python tester relied on the tester > being a separate binary. The tester was actually implemented in > Python, which made it very easy for the tester to interact with > individual test cases. How would I accomplish that on the executor > branch? I think your approach of writing the tester in Python was a good one, as that allows the introspection features you mention. To plug it into the executor code, we'd need to explicitly write a new backend type that invokes external programs. I have not thought much about it, but it's possible to make this as generic as the previous code was. Or maybe we shouldn't make it generic, and instead we should just have a native pyUnit primitive. If you have interest in moving this forward, we can discuss. But the executor code is about to receive some other big changes to properly fix one regression that was introduced, so I'd wait a little bit longer. Cheers
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?A84C5AFA-5CC8-4EEF-94CA-80A258C833C6>