Re: [Gnu-arch-users] more on the merge-fest

[zander]

On Tue, Nov 25, 2003 at 09:03:08PM +0100, Robert Collins wrote:
> On Wed, 2003-11-26 at 06:11, address@hidden wrote:
> 
> 
> > I may misunderstand the point here since I dont know the context, but this
> > seems a bit strange to me.
> 
> In addition to the other good points made, consider what Tom is doing:
> 
> Hes taking an incoming change which (hopefully!) includes test + changed
> code.
> 
> Having reviewed the design & code itself, IF hes changed the external
> behaviour during that process, he needs to update the test to reflect
> this. IOW the test specification is likely wrong in the original patch.

Right, my point was that if Tom looks at the test and finds that the expected
behavior displayed in that code is not what _he_ expects it to be, then it
would be easier to change the test to test what he expects tla to do.
The point being that its easier to define correct behavior from outside then
it is by writing the actual code that causes that behavior.

But I guess this methodology is known, and people have already thought about
this, so there is nothing new for me to add :-)

-- 
Thomas Zander