Public SourceMage Discussion List

[Eric Sandall <eric AT sandall.us>]

On Mon, 2006-12-18 at 18:27 -0800, Andrew Stitt wrote:
<snip>
> To be clear, I agree with you that it would be advantageous to automate
> this process somehow (I think that was your underlying idea). I just am
> not (currently) convinced it should be implemented inside sorcery. So
> its not that I dont endorse your idea, I do, I just was asking for
> justification for a particular implementation detail of that idea. I
> think that is quite different from wholesale disagreement. I should have
> been more clear on that, I apologize.
> 
> I think the counter-point was that testing should be done before the
> spells are "distributed" to the "user" (whatever we decide that means).
> As opposed to not testing and letting sorcery work out the answer
> independently on every users box. The end result is the same (the spell
> works). Theres certainly a design tradeoff between the two solutions. My
> contention (and probably other's) was merely on that design tradeoff. Not
> that we shouldn't automate gcc version compatibility testing.
> 
> -Andrew

We could have a box (or boxes) dedicated to running prometheus on all
git commits against each spell modified. This, I think, would be a
better approach than modifying Sorcery to do so (IMO Sorcery is the
wrong tool for bug testing).

-sandalle

-- 
Eric Sandall                     |  Source Mage GNU/Linux Developer
eric AT sandall.us                  |  http://www.sourcemage.org/
http://eric.sandall.us/          |  SysAdmin @ Shock Physics @ WSU
http://counter.li.org/  #196285  |  http://www.shock.wsu.edu/

Attachment: signature.asc
Description: This is a digitally signed message part