[Twisted-Python] trial setUp/tearDown behavior

Jonathan Lange jml at mumak.net
Tue Nov 15 17:02:53 MST 2005


On 16/11/05, Jean-Paul Calderone <exarkun at divmod.com> wrote:
> To summarize the thread "Re: [Twisted-commits] r15100 - So I really do want this stuff":
>

Thanks for the summary -- it's helpful and I like it.

> me:
> It does not seem surprising that setUp is for setting things up.
>
> chorus:
> It is surprising that setUp is for setting things up.
>

It seems surprising that you chose this way of summarising.
Rather, I'd say "It does not seem surprising that Trial shares state
between test cases." / "It is surprising that Trial shares state
between test cases"

> A main problem to coming to some agreement here seems to be that a lot of people are assuming they know how trial deals with shared state (or how trial should deal with shared state), but I haven't seen anyone explicitly spell this out yet.  Doing this might be a good first step to making further progress.

A worthy thought, and most likely right.

Thanks again for the summary.
jml




More information about the Twisted-Python mailing list