42a5f01be3
app: rename TempBuf to GimpTempBuf
2012-05-02 17:51:00 +02:00
7441a1f6f7
app: turn the TempBuf's "bytes" into "format" and port everything to it
2012-05-02 17:50:59 +02:00
ff86f85744
app: remove x, y and color parameters from temp_buf_new()
...
Fix the places that passed the color by either temp_buf_data_clear()
or memset(), and assign x and y manually, they are going to vanish
completely soon.
2012-05-02 17:50:58 +02:00
0e9581b956
app: pass the paint area's coords to gimp_brush_core_color_area_with_pixmap()
...
They are no longer stored in the paint area TempBuf.
2012-05-02 17:50:58 +02:00
36fb008197
app: port GimpPaintCore's paste and replase APIs to GeglBuffer
2012-05-02 17:50:54 +02:00
f07f01664e
app: heal the paint_buffer directly, without an intermediate copy
2012-05-02 17:50:54 +02:00
1c60024250
app: port GimpHeal to GEGL, using insane buffer casting tricks
2012-05-02 17:50:53 +02:00
441120b6b5
app: use gimp_gegl_dodgeburn() in GimpDodgeBurn
2012-05-02 17:50:53 +02:00
6efd812d08
app: s/GIMP_GEGL_RECT/GEGL_RECTANGLE/
...
Defining GeglRectangle inline is so useful that it has been added to
GEGL.
2012-05-02 17:50:52 +02:00
a76f44a9de
app: port GimpConvolve to gimp_gegl_convolve()
2012-05-02 17:50:52 +02:00
49dcc3d78a
app: remove obsolete include
2012-05-02 17:50:51 +02:00
c0fdfbd226
app: port GimpInk blob rendering to GeglBufferIterator
2012-05-02 17:50:51 +02:00
c5c555c834
app: add "take_ownerspip" param to gimp_temp_buf_create_buffer()
...
and get rid of some more intermediate buffer creation/destruction, and
some redundant temp_buf_free() plus g_object_unref().
2012-05-02 17:50:51 +02:00
7b5252ac54
app: replace GimpPaintCore::get_paint_area() by ::get_paint_buffer()
...
and remove the public TempBuf based get_paint_area() API.
2012-05-02 17:50:51 +02:00
b4996f2605
app: port GimpSourceCore and its classes to painting to the paint_buffer
2012-05-02 17:50:51 +02:00
0c8cf5ef73
app: more GEGL in gimp_smudge_motion()
2012-05-02 17:50:50 +02:00
60ae55f87b
app: don't return bogus paint buffers if get_paint_area() returned NULL
2012-05-02 17:50:50 +02:00
70c6885e7e
app: replace GimpSmudge's accum buffer by a TempBuf
2012-05-02 17:50:50 +02:00
4455d3e75d
app: more GEGL in gimp_smudge_start()
2012-05-02 17:50:49 +02:00
1037c7e88b
app: more GEGL in GimpDodgeBurn, still uses gimp_lut_process()
2012-05-02 17:50:49 +02:00
0da63f9526
app: port GimpConvolve to GEGL apart from the call to convolve_region()
2012-05-02 17:50:49 +02:00
db0eb0f77c
app: paint ink's color into the paint_buffer not the paint_area
2012-05-02 17:50:47 +02:00
9ac79ddc9c
app: port paintbrush and eraser to paint to GEGL buffers
2012-05-02 17:50:47 +02:00
12ac0bea13
app: add gimp_paint_core_get_paint_buffer()
...
which is the same as gimp_paint_core_get_paint_area() but returns a
GeglBuffer to paint to, not a TempBuf.
2012-05-02 17:50:47 +02:00
ab29daf758
app: remove unneeded includes from gimpairbush.c
2012-05-02 17:50:47 +02:00
1dbadb772d
app: rename GimpPaintCore->canvas_buf to ->paint_area
...
because that's what it's called in public API. This has confused me
long enough now.
2012-05-02 17:50:46 +02:00
93009ef274
app: use gimp_temp_buf_create_buffer() all over the place
...
instead of creating linear buffers manually.
2012-05-02 17:50:46 +02:00
2957795372
app: add gimp_gegl_node_set_matrix() and use it instead of manual fiddling
2012-05-02 17:50:46 +02:00
2af0681587
app: don't create copies in gimp_paint_core_get_orig_image,proj()
...
and instead simply return the paint_core owned buffers. Also, move
graph creation and source buffer fiddling out of perspective clone's
inner loop, and set an area to be processed manually, which makes it
responsive again.
2012-05-02 17:50:46 +02:00
77c34a835b
app: pattern-clone using gegl_buffer_set_pattern()
...
instead of tons and tons of legacy code.
2012-05-02 17:50:45 +02:00
7da52eb917
app: keep the transform graph around during one paint stroke
2012-05-02 17:50:45 +02:00
93864e3082
app: remove legacy includes
2012-05-02 17:50:45 +02:00
7c2c671703
app: make gimp_paint_core_get_orig_image,proj() return GeglBuffers
...
and change their users accordingly, getting rid of temp buffers and
useless copies.
2012-05-02 17:50:44 +02:00
2c4a59c1e5
app: fix buffer copying offsets in gimp_clone_motion()
...
so it works again at the image's borders.
2012-05-02 17:50:44 +02:00
95bf75d7a5
app: in GimpSourceCore's vfuncs, pass GeglBuffers around not PixelRegions
...
which simplifies a lot of code, just don't look at GimpHeal until
it's completely ported ;)
2012-05-02 17:50:44 +02:00
eb2e658408
app: replace GimpPaintCore->canvas_buf by a GeglBuffer
2012-05-02 17:50:43 +02:00
c9d0386e0c
app: replace GimpPaintCore->saved_proj_tiles by a GeglBuffer
2012-05-02 17:50:43 +02:00
50d7242b87
app: remove gimp_image_transform_rgb()
...
and use gimp_rgba_get_pixel() instead.
2012-05-02 17:50:43 +02:00
7f437d7d53
app: remove gimp_image_get_foreground,background()
...
and use gimp_context_get_foreground,background_pixel() instead.
2012-05-02 17:50:43 +02:00
2a853cd1ee
app: remove GimpPickable::get_bytes()
2012-05-02 17:50:42 +02:00
f4a4aa2133
app: use a GEGL graph for perspective clone's transform
...
Horribly slow, but seems to work.
2012-05-02 17:50:41 +02:00
867da8f293
app: add gimp_context_get_foreground,background_pixel()
...
which takes a Babl format to convert from/to.
Include <gegl.h> in a million places.
2012-05-02 17:50:41 +02:00
39cca4aabd
app: more GimpImageType -> Babl
2012-05-02 17:50:40 +02:00
ee97c1f4ff
app: correct usage of babl formats
...
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.
2012-05-02 17:50:38 +02:00
7223af17a2
app: completely port GimpDrawable::apply_region()'s API to GeglBuffer
...
and pass lots of coords that were hidden inside PixelRegions
explicitly.
2012-05-02 17:46:12 +02:00
061e044e0a
app: change GimpDrawable::replace_region() by ::replace_buffer()
2012-05-02 17:46:12 +02:00
3bb973f24b
app: pass the right undo buffer and extents to gimp_drawable_push_undo()
...
the coords need to be in drawable bounds now that we got rid of the
weird sparse logic.
2012-05-02 17:46:11 +02:00
4bcddb04a2
app: turn gimp_drawable_apply_buffer()'s src1_tiles argument into base_buffer
2012-05-02 17:46:11 +02:00
ca734f91b7
app: port gimp_paint_core_paste() to gimp_drawable_apply_buffer()
...
using the new TempBuf-GeglBuffer.
2012-05-02 17:46:11 +02:00
6c59c45612
app: replace GimpPaintCore's undo_tiles by a GeglBuffer
...
and simply use a dumb straight copy of the drawable, it's GEGL's job
to do proper COW.
2012-05-02 17:46:10 +02:00
Michael Natterer