Hi,
On Thu, Jul 09, 2015 at 04:52:12PM +0200, intrigeri wrote:
>
> bertagaz wrote (09 Jul 2015 13:28:23 GMT) :
> > On Wed, Jul 01, 2015 at 07:19:04PM +0200, intrigeri wrote:
> >> bertagaz wrote (25 Jun 2015 09:41:23 GMT) :
>
> Indeed I find it too vague so I've rephrased this paragraph (414e4f3),
> and added (414e4f3) another requirement:
>
> * if applicable, the commit at which the base branch was at, when it
> was merged into the branch being built;
Great, fine with me.
> (because we need to run the test suite not from the commit on the
> topic branch, but on the result of merging its base branch into the
> topic branch at that commit)
>
> IIRC I've made it so the build log contains that information, and if
> the build scripts need adjustements to make it easier to retrieve from
> Jenkins (e.g. logging it in a machine-parsable format), let me know,
> I'll be happy to follow-up on this.
Noted. Another way to be sure to test the right {topic,base} branch HEAD
commits could also be to simply copy the workspace from the isobuilder
to the isotester. That's something Jenkins knows how to do.
> Now, it would be good to have feedback from other contributors, in
> particular those who will be directly affected. Let's schedule
> a session to look into this at the summit, shall we?
Yes, we had one but not so much concerned people attended.
So please contributors, take some time reading and commenting on this
blueprint. Most of it is a copy of the automated builds one, as it is
the next step in the chain and the previous one defined already most of
our design.
Given it will be deployed in < 2 monthes, it's quite the last moment to
do so, and to make it clear, you won't be able to complain afterward. :)
bert.
