this is all good news. I yesterday was the first day since I got back
wednesday that I had any time to work on gegl. I have some GeglBuffer
and GeglBufferDouble stuff written, and I will try to get that into
place soon. I assume you want me to check in _working_ versions of this
stuff only?
Here is my plan of attack:
Write GeglBuffer (abstract) and GeglBufferDouble (concrete).
Write SampleModel (abstact), ComponentSampleModel (abstract) and
PixelInterleavedSampleModel (concrete).
Write ColorModel (abstract), ComponentColorModel(concrete)
Write GeglTile (the jai analog of this is Raster) using the above three
classes.
Write GeglImage as a collection of GeglTiles
after I get all of those in place, me or someone else can start
converting the ops to the new api's, while I or someone else implements
new sample models, color models and buffer types and seeing if the api's
are flexable enough to do what we need to do. I am still thinking about
iterators, but I am waiting to write a few ops to see the best ways to
iterate.
Also, I emailed sven and got commit access.
Calvin Williamson wrote:
Daniel:
I checked in some code that should make the areas you want to
work on a little easier to deal with.
Following our discussions, I moved out a bunch of ops,
removed that silly ChannelSpace stuff, introduced a
single color object which will store the data as
doubles internally (actually currently it uses
floats but we can change this to doubles shortly
after I do some additional work on the pixel value
types).
You should be able to make some progress for Images,
DataBuffers, Tiles, Iterators, SampleModel (or
Storage, we can even change this name too if
you want. Also the way the ops get the data pointers,
etc.
Let me know if you get some stuff working...Ill
email Yosh about a cvs account unless you already
did...
I can even remove a couple more ops if it still
feels like too many to work with. I left a couple
of the Unarys since they have different
parameters/properties.
We could trim it down even more.
Im also not necessarily wedded to the current ImageOp
classes and subclasses structure either, if you can
imagine a better breakdown of those classes.
I am focussing still on general input, output validation
issues (like inserting color or other data conversions
during evaluation), and general property/parameter
passing issues.
If I need to clean up even more areas for you to
make progress, let me know....
Calvin
