GIMP was doing evil hacks lying to GEGL about it's pixels being in a linear
color space when they are not. This causes incorrect rendering, makes gaussian
blur misbehave etc.
The legacy projection modes should be implemented using the same 2.2 gamma
formats that are correct to specify for sRGB data. (for proper color
management in higher bitdepths; icc backend babl formats should be used.)
For the old image modes correct babl formats are:
R'G'B'A u8 - 8 bit RGB with 2.2 gamma (sRGB) with linear alpha component
R'G'B' u8 - 8 bit RGB with 2.2 gamma (sRGB)
Y'A u8 - 8 bit Grayscale with 2.2 gamma with linear alpha component
Y' u8 - 8 bit Grayscale with 2.2 gamma
Y u8 - 8 bit linear data, used for masks/channels
A u8 - 8 bit linear alpha
-----------------------------------------------
RGBA float - 32bit floating point linear light RGB
RaGaBaA float - 32bit floating point linear light RGB, premultiplied alpha
to be used for processing that needs to scale by the alpha,
(blurs, resampling etc)
R'G'B'A float - 32bit floating point sRGB with gamma, to be used where
the result depends on being closer to perceptual when
processing, can be used a cheaper alternative to CIE Lab
based modes.
-----------------------------------------------
The legacy layer modes should use the formats with gamma 2.2 only for loading
and rendering legacy XCF files correctly, in the brave new world compositing
should most likely be done in linear light with "RGBA float" and even better
"RaGaBaA float" like GEGL does for porter duff and other compositing modes.
The ability to chose the legacy layer modes should probably be hidden from the
user unless an old .xcf has been opened.
It's likely that the removed call and function were causing
random drawing artefacts, but as this was once in a blue
moon bug, I have just a guess to go by however, testing
revealed no regressions, so pushing.
such as masks and outlines. The cache is currently very stupid and
only cacheds the last transformed object. Add caches to GimpBrush for
its mask, its pixmap and its boundary, and remove the same caches and
a ton of members from GimpBrushCore. This involves adding lots of
const qualifiers because GimpBrush returns const pointers now for
trasnformed stuff.
Rebased/fixed to go on top of current master. Next commit will add cleanup.
Had to change author tag because gnome is not accepting random stuff
in email fields. Original author is tarai, from gimp painter project
in sourceforge.
Add a transform matrix to GimpCanvasBoundary and get rid of the whole
BoundSeg transform code in boundary.c and gimpbrushcore.c, it was
impossible to get this right on that level. Also fix te extents of
GimpCanvasBoundary os it leaves no artifacts.
- GimpCanvasBoundary takes unsorted BoundSeg arrays now and uses
gimp_display_shell_transform_boundary() and gimp_cairo_add_boundary().
- Nobody calls boundary_sort() any longer for the purpose of displaying
a boundary.
- gimp_display_shell_transform_boundary() got offset parameters
so it can transform things that are not in the image's coordinate
system.
Michael Natterer