[cairo] cairomm 1.6.0 fails with cairo 1.6.4 + quartz on Mac OS X

[Murray Cumming]

On Fri, 2008-04-25 at 12:59 -0400, Benjamin Reed wrote:
> On Fri, Apr 25, 2008 at 11:24 AM, Benjamin Reed <rangerrick at gmail.com> wrote:
> 
> >  I don't know this is solveable without somehow splitting up the logic
> >  into platform-specific bits in the cairomm code and then linking it
> >  all together, from what I can tell.
> 
> I've made a patch that splits those bits out into separate .cc files,
> although I'm not much of a C++ guy, so it could be the wrong way to go
> about it.  ;)
> 
> http://svn.finkproject.org/svn/users/ranger/trunk/experimental/common/main/finkinfo/graphics/cairomm1.patch
> 
> Also, I'm not sure the best way to go about symbol-visibility issues
> for that, should I use the GCC pragma or something else to make them
> "private"?  It didn't seem appropriate to actually make them a part of
> the Context object...
> 
> Anyways, the concept works, at least, I was able to compile cleanly
> with both the quartz and xlib bits enabled.

It seems like a generally good solution. I'd prefer these new functions
to be in a Cairo::Private namespace, and I'd prefer them to be named
wrap_surface_*() if we don't have some similar wrap* functions already.

And the headers for these should not be installed.

I'll get around to that eventually if nobody else does.

In general, please put patches in bugzilla in future. Many thanks.

-- 
murrayc at murrayc.com
www.murrayc.com
www.openismus.com