Hi all,
I'm present my self : i am the developper of monkey-bubble (http://home.gna.org/home/monkeybubble). MonkeyBubble use svg gfx from the begining but i am not happy of my display architecture : i use librsvg to load svg to a bitmap. The main disadvantage of this, is that i can't do sophisticated transformation of vector object like rotation, zoom ..etc
So i have read somewhere that inkscape plan to implement a vectorial canvas lib. I am very interesed in this. is the work already started ? where can i know more about this ?
regards, Laurent Belmonte
On Thu, 25 Mar 2004, laurent belmonte wrote:
Hi all,
I'm present my self : i am the developper of monkey-bubble (http://home.gna.org/home/monkeybubble). MonkeyBubble use svg gfx from the begining but i am not happy of my display architecture : i use librsvg to load svg to a bitmap. The main disadvantage of this, is that i can't do sophisticated transformation of vector object like rotation, zoom ..etc
Cool. I myself got involved in Inkscape (actually Sodipodi at the time) while working on an SVG-based game, and in need of an SVG canvas.
So i have read somewhere that inkscape plan to implement a vectorial canvas lib. I am very interesed in this. is the work already started ? where can i know more about this ?
Work has not started on implementing it; about a year ago a couple developers started an attempt to extract the canvas but ran into a variety of problems with being able to cut it loose from underlying libs, as you can imagine. So after taking a step back and looking at it again, it was recognized that the codebase would need to be refactored and several of the component libraries either replaced with standard libs (Pango, etc.) or written into a more modular form (livarot, AST, etc.) Better documentation is also needed, especially for identifying the API.
These prepatory steps are underway currently. Once these steps are far enough along, then the work of extracting the canvas should be much more tractible.
I think what we need the most right now is some conceptual-level brainstorming on how this canvas should work. I.e., what sort of functions and data structures should it expose in its API? Do you have some ideas?
Bryce
participants (2)
-
Bryce Harrington
-
laurent belmonte