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.
- keep babl palette formats around in the image in indexed mode
- create drawables with the right format
- as first test, convert indexed drawabled to rgb/gray by simply
calling gegl_buffer_convert()
because it would require really evil hacks to honor these properties
in the gegl projection if they were on the mask, and because they
actually belong to the layer.
Instead of getting the floating selection from the image. Change
gimp_drawable_has_floating_sel() to get_floating_sel() and remove the
"floating_sel" parameter from gimp_drawable_deatch_floating_sel().
This cleanup is one more step towards sanitizing floating selections,
and makes the crash fix in the next commit much much simpler.
Make gimp_image_get_uri() and gimp_image_get_filename() behave as in
the GIMP 2.6 days. Add new functions gimp_image_get_xcf_uri(),
gimp_image_get_exported_uri() and gimp_image_get_imported_uri().
because we temporarily need them after removal from their item trees,
and bad things can happen if they are not kept alive by the undo stack
(like when undo is disabled while loading an image).
and implement it in GimpLayer to unset the removed flag on the layer
mask. Remove layer mask special casing from gimp_image_add_layer().
Make sure that all an item's children get their removed flag unset in
gimp_item_unset_removed().
- replace gimp_display_shell_selection_control() by undraw() and restart()
which actually say what they are doing
- remove enum GimpSelectionControl
- replace GimpImage::selection_control() by ::selection_invalidate()
because none of the other enum values was used any longer
which is the last bit of non-item drawing of stuff that is not
somehow the image itself... wheee!
This involves reverting commit
6bce0641d4 and adding back all the
vectors handlers that were in gimpdisplayshell-callbacks.c before.
Change the callbacks to manage proxy items for all the image's
vectors.
- Add signals GimpImage::guide_added(), removed() and moved()
- Remove singal GimpImage::update_guide()
- Adapt core code to emit the new signals instead of update_guide()
- Have the shell connect to the new signals and update guide canvas
items as needed
- Remove gimp_display_shell_draw_guides()
- Add GimpImage signal "sample-point-moved" and emit it when needed
- Let the shell connect to the sample point add, remove and move signals
and update the canvas items accordingly
- Remove gimp_display_shell_draw_sample_points()
This is needed so that the area covered by the vectors before the
change is exposed. Ideally of course the update signal should
include the changed area.
...just as we do for drawables. Connect to adding, removing, modifying
and toggling visibility of all vectors and emit "update-vectors"
accordingly. Add an update-vectors signal handler to GimpDisplayShell
and remove all other vectors handlers.
Description of undo actions should be marked as action descriptions,
and not as commands. This is required for translation for some
language (like Hebrew) that require a different grammatical tense for
describing actions
