Re: reforming libgdome object-oriented-fashion with gobject
Luca Padovani <[email protected]> Fri, 14 Feb 2003 18:58:43 +0100 (CET)
| Newsgroups | gmane.comp.gnome.gdome |
|---|---|
| Message-ID | <[email protected]> |
Hi. On Fri, 14 Feb 2003, Kang Jeong-Hee wrote: > as a gtk application developer, I really miss gobject style of object and class system. > do you mind if libgdome reformed with gobject's GType and etc.? > > if you don't, I'll ask W3C local office a full map of DOM classes hierarchy > and make libgdome's objects and vtabs to GType-intergrated objects and classes. > > and may I have your advice for this work? some comments: a) re-implementing libgdome so to use gobjects is a _MAJOR_ project. It implies changing basically everything. b) why do you have to ask a W3C local office a map of DOM classes, wouldn't the DOM technical specification be enough? c) I agree that having gdome with gobjects would be very nice for consistency with other libraries, but what are the other advantages now that the library has been developed and reached a fairly stable point? My point of view is certainly biased by the fact that I can hardly conceive OO programming in C, except in those cases (like Gdome) where having the C layer helps people in writing bindings for different languages. Although I followed gdome2 development, I admit I never (never!) use it directly. I always use some higher-level binding that best suits my needs. If you were to start anew a big project, wouldn't you consider to write it in another language which is not C? But I'm going off-topic here and probably starting again the old flame about the favorite programming language... d) I have no particular advice just I'm not convinced at all this should be the way to go (at least in the short term). I'd also like to hear opinions from other gdome2 users on this. Again, if I were to do something like this, I would consider doing it from scratch or directly in another language, but in any case I see this as a long-term project. -- luca