On Mon, Nov 10, 2014 at 09:26:06PM +0100, Joachim Breitner wrote: > Hi, > > > Am Montag, den 10.11.2014, 21:19 +0100 schrieb Iustin Pop: > > > The only thing that I’m worried about: What is the best way to ensure > > > that all packages uploaded to experimental are indeed built with the ghc > > > in experimental? > > > > > > One brute-force-method would be to simply build-depend on "ghc (>= 7.8)" > > > in each and every package. But is there something smarter? > > > > I don't know of any other (this doesn't say much), but if this is what > > we can do now, why not? > > well, just checking... > > We’ll also have to adjust our tools, e.g. the binNMU scheduling and the > package plan. Ah, now I see. Sorry, I (still) don't know how all that works, but I would say that moving forward (with 7.8) is important. Up to you whether the overhead of changing the tools is warranted or not. thanks, iustin
Attachment:
signature.asc
Description: Digital signature