Re: Scripting choices [Was: 2.4 Module List - zenity]

[Rodrigo Moya <rodrigo gnome-db org>]

On Fri, 2003-03-28 at 05:21, Biswapesh Chattopadhyay wrote:
> > > this makes me think that it could be a good idea to offer an integrated
> > > scripting engine, as is done in Mac OS, for instance. I don't know
> > > exactly how this could be accomplished, but I guess making obvious to
> > > every GNOME user that they can write nice little scripts could be a good
> > > idea.
> > 
> > This is something I've privately advocated for a while.  The problem is
> > really *which* language to use.  Aside from holy wars, there's licensing
> > issues, and in the end user-friendliness.  People who code C all day
> > look at Python and call it user-friendly, but someone who's never coded
> > before might not agree - but then, I've never conducted a non-programmer
> > scripting usability study, so I can't personally make any conclusions
> > about it.  ;-)
> 
> I did a sort of informal survey about this on the lists a few months
> back and the consensus was that:
> 
> 1) We *really* do need a common (official) scripting framework.
>
I started such a thing in the libgnomeoffice CVS module. Right now, it
only supports scripting via .NET (Mono) assemblies (not really fully
tested, but it should basically work. But, although only supporting tat,
I did it with the idea to support any scripting language/framework.

Please have a look at it. I think it can be a good starting point.

> 2) Python should be the official scripting language.
> 
as said above, it should be a generic scripting engine, for any
language, and not be tied to a specific one.

> Another point that was discussed was a language-agnostic scripting
> framework. However, this was felt to be mostly unnecessary, too much
> hassle for too little gain + leading to confusion, inconsistency, etc.
> 
see the libgnomeoffice code, and tell me if it's too much hassle.

> After the discussion, I remember someone saying that he intended to
> start a module in GNOME CVS for a common scripting framework. However,
> that's where the thread trailed off, and I moved on to other things
> (since there was a lot to do before we could think of implementing
> script - a lot of basic functionality is still missing in anjuta).
> 
I suppose that was me :-)

cheers