[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gzz] The Fenfire/Loom test system
From: |
Tuomas Lukka |
Subject: |
Re: [Gzz] The Fenfire/Loom test system |
Date: |
Fri, 28 Mar 2003 14:13:36 +0200 |
User-agent: |
Mutt/1.4.1i |
On Sun, Mar 23, 2003 at 03:57:54PM +0100, Benja Fallenstein wrote:
> Tuomas Lukka wrote:
> >On Sat, Mar 22, 2003 at 05:50:13PM +0100, Benja Fallenstein wrote:
> >
> >>FYI-- the new test system in Fenfire's CVS makes .test files real Python
> >>modules, which requires that you write a setUp() method if you want code
> >>to be run before every test. I've just finally docced the system; see
> >>below (not much new there, tho).
> >
> >
> >Could we put this in navidoc so that it can also be used in libvob?
> >
> >Having cross-dependencies between libvob and fenfire is not good ;)
>
> Hm, it currently knows about graphics apis IIRC, because of the
> 'fail: AWT' stuff. If we can abstract this out, no problem.
I think it should be abstracted to a generic "fail" mechanism:
the test can say
fail: AWT
fail-bug: Kaffe
meaning that the test always fails on AWT because the things are intentionally
not implemented there, and it fails on kaffe, not intentionally, because of some
bug, that needs fixing.
For the test framework, these would be just keywords.
What do you think?
Tuomas