Re: Notes, April 21, 2016

On Fri, Apr 22, 2016 at 11:43 AM, Dominique Hazael-Massieux <dom@w3.org>
wrote:

> On 21/04/2016 17:04, Harald Alvestrand wrote:
> >> Mediacapture-main
> >> =============
> > NOte: Travis seems to error out on warnings. Should it?
>
> It normally should (as a -Wall type of sanity check); but with the
> recent change where respec warns on old-school WebIDL, I have had to
> disable it.
>
> As a side note, Vivien and I have reworked quite a bit the Travis set up
> over the past two days due to a recent non-backwards compatible change
> in the respec toolchain; things should be OK now, but do let us know if
> anything weird occurs.
>
> If anybody use webrtc-respec-ci on their local version of the spec, they
> should look at updating it (and run a make update).
>
> > Conflict in the changelog. Merge-ready.
> > Suggestion: Tell people to suggest a changelog entry in the PR rather
> > than adding it themselves.
>
> FWIW, I usually don't provide changelog entries; but in this case,
> master was in a Travis-unhappy state because of an empty
> <ul><li></li></ul> in the would-be changelog since April, and the only
> way to make it happy on PRs was to add something in that <li>.
>
> So I would add another suggestion: never leave master in a
> Travis-unhappy state :)
>

Dom, this suggestion was more a general one because of the "multiple
changes to the same place cause a conflict issue", regardless of the state
master.  And I already fessed up to leaving it in the bad state -- I meant
to leave an $&nbsp; so it wouldn't complain.  I didn't realize that the
current Ttravis setup would remove all my elements, generate a warning, and
fail!  Anyway, I will try not to screw up master again :)

>
> Dom
>
>
>