output args / hack-status

[Andy Wingo]

Hey Jan+Rotty,

On Wed, 2004-10-27 at 17:23 +0200, Andreas Rottmann wrote:
> In G-Wrap, there is support for output arguments, which is excercised
> via so-called type rules for the .defs in the wrapset specification,
> e.g. (from glib-spec.scm):

And also in dbus-spec.scm. The only problem this has is e.g. gchar*: is
it an output char or an array? This problem is especially acute in the
libecal (evolution-data-server's calendar component) API. For some
brain-dead reason they want to return values only by output parameters.
Take a look at e-cal-component.h for an example. Furthermore, if it's an
array, is it null-terminated, or is there a length parameter somewhere
else?

I don't know what the solution is. It would be exceedingly clever if
these patterns could be captured somewhere in g-wrap, though.

> > I can't commit to gna yet, maybe you can include my patch for the next
> > release?  I also like to wrap gtk-menu-popup to allow menu positioning
> > function.
> >
> I'll have look at merging your changesets.

To recap for Jan: I don't have a stable internet connection. It's been
out for most of this week. I downloaded these emails, but can't do
anything as I reply. So I'll add you as soon as https will go through,
but in the meantime just be patient :-)

Then there's the other thing. I won't really be able to hack from
mid-november until february, and also I'll be out next week. (I'm moving
from Namibia to Spain through various parts of the US.) If either of
you, Andreas and Jan, want to get on the 2.10 platform bindings
bandwagon, you should step up. I'll get out a 2.8.0 release within the
next couple weeks, but the platform bindings needs more deadline-loving
than I can give it.

To be clear, I don't think we're really ready for it. Being a "beta"
bindings set would be achievable, though. (See the rules linked to in
the original mail for details.)

Cheers, and happy hacking,
-- 
Andy Wingo <address@hidden>
http://ambient.2y.net/wingo/