4 Aug
2016
4 Aug
'16
7:38 a.m.
Unfortunately we cannot do much better with SmalltalkHub/Monticello, because it
doesn't have any real support for parallel development, as the latest version is
whoever happens to be winning the branch race condition, and you will see if your code
passed all tests only after you've broken the master build, because there is no simple
way to test it before integration. Add the mind-bending complexity of configurations and
you have fun for a lifetime.
So yes, we will waste some time from time to time if the build breaks in such proportions,
but I don't really see a better solution without a better infrastructure and tools.
P.s.: One more option that came to my mind was having CI automatically update
ConfigurationOfRoassal if the build passed, but I don't know how to do that (yet).
Peter
On Wed, Aug 03, 2016 at 08:33:12PM -0400, Johan Fabry wrote:
Thanks Peter for fixing this!
And yes I am aware that I am on a development branch so some instability can happen and
functionality can change. So I am OK with that. That being said, committers should ensure
that all tests always pass. Situations like this cause time to be wasted and are bad
publicity for Roassal.
--
Does this mail seem too brief? Sorry for that, I don’t mean to be rude! Please see
http://emailcharter.org .
Johan Fabry - http://pleiad.cl/~jfabry
PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile
On Aug 3, 2016, at 14:27, Peter Uhnak
<i.uhnak(a)gmail.com> wrote:
On Wed, Aug 03, 2016 at 01:49:30PM -0400, Johan Fabry wrote:
_______________________________________________
Moose-dev mailing list
Moose-dev(a)list.inf.unibe.ch
https://www.list.inf.unibe.ch/listinfo/moose-dev
latest version
This is the core of the issue.
We can either use #bleedingEdge and slit our wrists with it from time to time (because
there are no guarantees), or change ConfigurationOf after every single functional change
(so you commit your changes, see the build to pass, then you go back and edit the code
again).
(Or use git, but people apparently like the approach above.)
Peter
_______________________________________________
Moose-dev mailing list
Moose-dev(a)list.inf.unibe.ch
https://www.list.inf.unibe.ch/listinfo/moose-dev