preparing for a 4.6.4

Dirk Hohndel dirk at
Wed Apr 19 07:43:39 PDT 2017

On Wed, Apr 19, 2017 at 03:08:42PM +0200, Anton Lundin wrote:
> On 18 April, 2017 - Dirk Hohndel wrote:
> > I know, we've spent all this time on Subsurface-mobile and the bugs there
> > still aren't fully understood (let alone fixed). But I can't forget about
> > our main app again - and we really need to get an update out to our users
> > with the fixed libdivecomputer.
> > 
> > I have two options. I can do a quick new build that uses the new
> > libdivecomputer and the Subsurface that we used for 4.6.3. Or I can use
> > the current master and stabilize this.
> > 
> > The advantage of the first option is that we had really really complete
> > translations for that. The advantage of the second option is that we'd get
> > the other new things out to our users, including all the fixes to the
> > planner, the manual entry of depth and duration, etc.
> > 
> > I'm kinda leaning towards the second option but that means we need to
> > spend a week or two, focused on testing the latest master and finishing
> > the translations, again. If people have strong feelings one way or
> > another, please speak up.
> > 
> > I'll do new "daily" builds next and also make sure that all the source
> > strings are pushed to Transifex, just in case.
> > 
> Why not do both?
> Roll a (or something) Now, with just libdivecomputer patches,
> and when we got something stable, roll a 4.6.4?

That would be a 4.6.4 and 4.6.5 (because... reasons - actually, because we
use the fourth digit for the number of commits since a release).

The reason I don't just "do a release" is that while this may be invisible
to most of you, there is actually a good bit of work involved for me to
make a release. And so I was trying to see if the translators would be up
to doing a 4.6.4 or if there were other considerations to take into
account. "why don't you just do all that work twice, Dirk" wasn't really
part of my plan...


More information about the subsurface mailing list