Re: [gtk-list] Re: nonwrapper C++ bindings for GTK (just an idea and a script)
- From: Guido Draheim <guidod gmx de>
- To: gtk-list redhat COM
- Subject: Re: [gtk-list] Re: nonwrapper C++ bindings for GTK (just an idea and a script)
- Date: Thu, 20 Apr 2000 13:13:20 +0200
a) yes, I missed that... (grr, d**n, **** ** * ** **** ****)
b) no, it's not the same.
"sugar" needs some hand-edited files to work on, and it *will*
actually create a C++ wrapper object. It is the same as it
does not need an extra lib (so far as I can see), and it has
typesafe signals (big advantage anyway). You may as well need
an extra handcrafted sugar-header file with each version of
GTK or the widgets that work on top of it.
"cc-headers" can work on *any* object-oriented C-header file
that follows the name-rules of GTK. I tried it last night
on gtkextra. Just try it on your own written Widget-header.
And it does neither use a wrapper object nor is there
actually any member accessor needed.
The point in "cc-headers" is *not* to go 100% C++, instead
you can start off nicely with C++ with its object-namespaced
methods and members, and if you care for, you can incrementally
get back to C since the objects you create and the functions
that work on these are really 100% *the* *same*. It has the
effect to mix C-code and C++-code in just the same file, even
the same procedure.
In the end, it is *not* a wrapper. The generated header files
can be included from a C compiler and they will work just like
the original. Well, I guess an experienced GTK programmer does
not need it anyway since his eyes can easily reduce the c-text
with all its GTK_FOO and gtk_foo_xx. It does help me, someone
only occasionally writing a gtk-box for his favourite C-level
thanks for pointer anyway, ... so, did you gave it a try?
Guillaume Laurent wrote:
> Guido Draheim <firstname.lastname@example.org> writes:
> > and I don't need a real wrapper bloatware for it (as is some
> > C++/GTK-wrapper or even Perl/GTK).
> > I don't know if the idea may have come up earlier, I checked the
> > mailing list archive (about 6 months or so), and I didn't find
> > anything like this.
> Well, you didn't check hard enough :
> To unsubscribe: mail -s unsubscribe email@example.com < /dev/null
mailto:firstname.lastname@example.org "Edel sei der Mensch,
tel: +49 177 5928220 hilfreich und gut" --G.
] [Thread Prev