Re: marshalling merges

[Tim Janik <Tim Janik Hamburg Netsurf DE>]

On 21 Jul 1998, Owen Taylor wrote:

> Tim Janik <timj@gtk.org> writes:

> > 3) what was wrong about the automatic marshaller lookup approach, where
> >    the programmer is supposed to pass the marshaller as NULL to the signal
> >    creation function, and the marshallers get looked up through a hash table
> >    of statically registered functions?
> 
> I think this is sort of an extra. I don't think there is a real
> reason to do automatic marshaller lookups in the internal GTK+
> code, but I think we should definitely add that facility for
> people who are writing GTK+ apps and creating new widgets.

hm, i night get around to implement an automatic marshaller lookup
in the next few days.
but still, why wouldn't you want to use that in gtk itself?
it would impose an extra hashtable lookup upon class_init()'s sgnal_new()
execution,
and i'm pretty convinced that these lookups are not going to be noticable.
in any case, we will probably require a C style proggy to perform the desired
task (for type conversions so we only investigate their fundamental portions).
so since i *might* get around to implemeniting this, i need to announce
that elliot's recent chages should be considered *experimental*, as it is
the case with all of the new features in the -1.1 branch (though, pretty many
issues have been settled already, e.g. the key binding stuff (i got the parsing
side working already and am about to commit that tomorrow) and the recent
object/child argument changes- and hopefully -also the new container resizing
code.

> 
> Regards,
>                                         Owen
> 

---
ciaoTJ