On 7 October 2010 11:36, Fernando olivero <fernando.olivero@usi.ch> wrote:
Also, the missing class is NBGLCurveRenderer, which i moved to NBOpenGL-Rome package. I did a snapshot of it. But it depends on RomeCanvas class(es) , which is part of Athens.. But just ignore that for now and load it. For demo, you need only NBGLCurveRenderer which not rely on any Rome-specific stuff.
Just for curiosity, why this dependance on Rome? To end up providing a Canvas were all the rendering is performed by OpenGL and you text rendering class ?
A NBGLCurveRenderer is actually an experiment, which should be a part of something bigger (read Rome). So, that's why i moved it there, breaking demo :)
The demo works just fine now!
I plan to implement an OpenGL backend for canvas, which implements Rome protocol. And then implement Morphic which using rome-canvas for rendering.. And then implement brand-new zoomable UI in it..
(i have a big plans)... :)
I was hoping you did!
Questions: 1.Once you have a NBOpenGLRomeCanvas , how do you plan to implement the zoomable UI? On which layer of abstraction? (Canvas or  Morphic ?) Is there going to be a "camera" abstraction? ( just like Lumiere ) or who will know  its current z transformation?
first i want to implement morphic rendering pipeline which will use Rome canvas, instead of Canvas. Next, i want to replace PasteUpMorph (World) with own subclass, which will actually control the camera, zoom level etc. So, it will be a global desktop-level zoom&translation. A submorphs will be not aware of it. In this way, its a less work and less changes to existing Morphic (i hope). And of course, i want a clear separate layer (Rome canvas) which can be used by any other UI/graphics framework.
2. Regarding Text, will you reuse Paragraph and Editor already present in the system, or use an external library like Pango? Â I was looking at the True Type font rendering demo, and everything seems pretty much into place, and now the "paragraph" layer must be implemented.
Initially, i will just do the simplest possible thing. Reuse what morpic provides. Text/paragraph rendering is a separate and very complex problem. Of course it would be cool to do things right, but not at initial iteration :)
How can i help? (Eventually...)
Anything you want. If you want something to be there, don't hesitate and dont wait for me - start coding :) My projects are public, they are open for contributions from anyone. Also, it would be cool to sync with Juan and his work on Morphic 3. I'd like to provide a GL backend for Morphic 3. But its unclear to me, what plans Juans having about it. Recently he said that he won't make any public releases. But i really hope that we could work together on it, because such ambitious projects need more than a single man in order to complete.
Fernando _______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
-- Best regards, Igor Stasenko AKA sig.