On 29/01/2015 3:26 p.m., Andrei Alexandrescu wrote:
On 1/28/15 5:34 PM, Jeremy DeHaan wrote:
What about that Aurora project? Wasn't that supposed to fill this kind
of role?
I think Aurora is dead in spite of significant goodwill and karma
expenditure. I think there might be a lesson in there, but I don't know
exactly which. -- Andrei
It's scope was definitely a little on the large side.
Also why with Devisualization I have split things up into very
manageable small chunks. With very specific scope.
Before I started it I had a very specific game plan for it.
Devisualization.Window is quite literally at the bottom and hence first
to be made.
I'm now at the point that I may not have the best image implementation.
But the abstraction is relatively pretty decent.
But small summary:
- window doesn't see many changes these days. Mostly documentation and
fixes for updates to image definition
- image, bare minimum png export and a relatively decent png import with
an ok mutable type
- font, rasterizes BDF fonts fine. No underline/strike through
- scenegraph, sort of the same situation for window. Although could have
more implementations of the scenegraph interfaces.
- mesh, ehh is at the very very beginning of life, has Wavefront Object
model loader
- util, has a bunch of sub projects
- core, needs more work especially the linegraph (given line/curves get
all points based upon the step)
- opengl, huge number of opengl functions need wrapping up in D types
for easy calling. Also more work on the wrapping of Opengl concepts like
buffers.
- photoshop_aco same situation as window
- I want a basic glu implementation but its too much work
- gmaterial, GUI toolkit based upon Google Material Design. Long long
long way off. I've only just worked out how to make half decent shadows
for papersheets. Which was a killer on how to do it.
Reward for doing small amounts of work. It's working half decently for me.
