Partial push of Josh's configure-dive-computer GSoC project

[Jef Driesen]

On 2014-08-14 16:02, Dirk Hohndel wrote:
> On Wed, Aug 13, 2014 at 08:52:16PM -0700, Thiago Macieira wrote:
>> On Wednesday 13 August 2014 21:10:31 Dirk Hohndel wrote:
>> > > Note that this needs libdivecomputer with at least commit
>> > > 56e2dec7266e71e725604a4a8bff5e61f3d62259 (I've cherry-picked that to
>> > > release-0.4 and it built successfully).
>> >
>> > Does it need that to BUILD or to WORK?
>> 
>> Build. Without it, you're going to get undefined references to 
>> hw_ostc3_*
>> functions.
> 
> I would have VERY MUCH appreciated if you had included a #define on the
> libdivecomputer side that had would have allowed us to build this
> conditionally on the Subsurface side.

You can conditionally build on the DC_VERSION_CHECK(0,5,0) macro. That 
would only break if you checkout a much older commit on the master 
branch. But why would anyone want do that? The master branch is an 
unstable development branch for a reason. But as promised I'm not 
planning any backwards incompatible changes (from v0.4.x) until v0.5 is 
released.

New functions can't break backwards compatibility, because they didn't 
exist before. But if you start to use them, that of course means you now 
have a dependency on v0.5 and not v0.4!

> We're right after the release so this isn't as big of a deal because we
> can talk to Jef to do a 0.5 release of libdivecomputer before we 
> release
> 4.3 - but in general I dislike API changes in a library that I cannot
> conditionally compile for.

I plan to release v0.5 in the not too distant future. Probably the end 
of September.

Jef