Re: AnyGUI/Narya Client

Magnus Lie Hetland <[email protected]> Sat, 16 Nov 2002 15:44:14 +0100
Newsgroups gmane.comp.python.anygui.devel
Message-ID <[email protected]>
Terry Hancock <[email protected]>:
>
> Hi All,

Hi!

I'm really sorry that I haven't anwered you before.

> I just joined the list, and I think I'd probably like to help with
> at least a couple of backends for AnyGUI.   So, I'm looking for
> where to start.

Right... Considering the relatively low activity at the moment, I'm
tempted to say "dive right in", but I guess it would be helpful for
you to understand a bit about how things work first.

The current architecture is underdocumented at the moment, so the best
source of info is the code. Take a look at some of the back-ends for
starters.

> My motivation is this:
> 
> In my proposal about Python Browser Clients:
> http://www.anansispaceworks.net/NaryaInfo/client_html
> I referred to "Tkinter", but I wonder if substituting AnyGUI would
> make my job easier.

Maybe. At the moment, doing simple things in Anygui is very easy, but
doing many complex things isn't even possible. Hopefully this
situation will improve in time, but things aren't exactly running at
top speed :)

> To run in popular stock browsers like Netscape 4.x or IE 5, I'll need to run 
> in Jython/Java-1 (not Java-2), which has AWT, but not Swing. I've heard Swing 
> will run on Java-1, though, so I'm wondering if it's possible to delivering 
> it with my applet instead of relying on it to be included with the 
> interpreter.  Is that even possible?

That I do not know.

> Or should I be focusing on a ground-up implementation using only
> AWT?

You could, of course, write an Anygui back-end for AWT based on the
current one (which is based on Swing). I don't think it should be too
hard -- just replace some API calls and some of the classes used etc.
Might be a nice way of figuring out the internals :)

> I think I'm also going to need to develop an SDL/PyGame based
> backend if I want to follow this up.  Less sure about that part,
> though.

I've already tentatively begun a back-end for PyUI (based on PyGame),
called pyuigui (found in the cvs distribution). Feel free to work on
it, as I probably won't for a while.

> I also noted the PyUI project

Right. It uses (among other rendererd) PyGame.

> -- which sounds like it might be faster to write a backend
> (renderer) for, especially in the environments I'm talking about.
> Would that make more sense perhaps? (I note that a PyUI backend for
> AnyGUI is "planned" but not working).

Exactly. (I'm reading in a linear fashion here, as you can see ;)

> So is there some kind of "Start Here" for someone who wants to implement a 
> new (or extend an old) backend for AnyGUI?

Well... You should probably start by taking a look at the empty
skeleton back-end skelgui.py. It refers to the document IRFC 14; that
document does explain quite a few of the concepts in the architecture,
but it isn't 100% up to date with the latest developments.

Feel free to ask if you have any specific questions.

> Thanks for any pointers,
> Terry

-- 
Magnus Lie Hetland        Practical Python          The Anygui Project
http://hetland.org        http://ppython.com        http://anygui.org


-------------------------------------------------------
This sf.net email is sponsored by: To learn the basics of securing 
your web site with SSL, click here to get a FREE TRIAL of a Thawte 
Server Certificate: http://www.gothawte.com/rd524.html