Re: [gtk-list] Re: Proposal: GUBIs gtk.def

[Kenneth Albanowski <kjahds kjahds com>]

On Fri, 21 Nov 1997, Tim Janik wrote:

> On Thu, 20 Nov 1997, Kenneth Albanowski wrote:
> > I hope this works well enough. Remember, the optimal goal is something
> > like Visual Basic or Delphi, where the entire object state can be
> > controlled through a property inspectory -- and that it should be possible
> > to invoke the functions within dynamically loaded code. 
> 
> i'm not sure what you mean by dynamically loaded code, are we talking only
> C here?
> for GUBI i'm thinking about generating one sole function as a multi caller,
> that takes the gtk-function description and it's parameters as argument and
> then calls it.

On platforms with dlopen() (or close relatives) it's not difficult to call
functions at run-time (except for the parameters, come to
think of it, which really need some marshalling code to work reasonably
well.)

> please look at my other mail to owen, we can do much more with slots like
> defining "virtual" widget parameters (e.g. can-focus), defining c-structure
> components for each struct _Gtk*, and we can have a slot describing the
> gtk_widget_set() interface...

Certainly.

> i'm currently implementing a parsing back end (in C ;) that reads in the
> *.defs file and builds an internal tree of it. this can have different front
> ends then (e.g. one  that dumps the function names and their descriptions to
> a text file).
> you could easily implement a front end that suits your needs for this then...

Indeed.

-- 
Kenneth Albanowski (kjahds@kjahds.com, CIS: 70705,126)