On 05 Feb 2013, at 13:13, Camillo Bruni <camillobruni@gmail.com> wrote:
On 2013-02-05, at 13:05, Igor Stasenko <siguctua@gmail.com> wrote:
On 5 February 2013 12:54, Camillo Bruni <camillobruni@gmail.com> wrote:
On 2013-02-05, at 12:52, Igor Stasenko <siguctua@gmail.com> wrote:
On 4 February 2013 19:19, Stéphane Ducasse <stephane.ducasse@inria.fr> wrote:
Igor
I think that camillo made this because NativeBoost is an essential part of our infrastructure and this is good to know that it is working. So I think that we should thank camillo for the time he spent on that.
It is pretty useless right now. If it would run on all platforms, it would worth having.
then add slaves.... you strike me with laziness again!
i would.. if i would remember that this job exists, and if you would spend some moments explaining why _you_ need it so badly, since you set it up. Because running tests takes like 5 seconds, and i doing it before comitting code, and feel pretty fine with it.
Ask yourself: - why do we have tests? - why do multiple people work together? - why do we want publicly available artifacts? - why do we want these artifacts tested publicly? - why do we write configurations? - why do we try to following coding standards? - why do we program in Smalltalk? - why do we mostly write deterministic code? - why do we work on Pharo? - why do we build a jenkins infrastructure? - why do we write down documentation?
After successfully answering these questions you will understand!
Camillo the Zen Master ! The point of the infrastructure is: you write code in one specific 2.0 image on your mac, the CI infrastructure makes sure you didn't break stuff on other platforms, other/older pharo versions, all automatically. Of course, you need lots of stuff for that, but creating one build job is the first step. Sven