(gimp_drawable_transform_affine)
(gimp_drawable_transform_flip)
(gimp_drawable_transform_rotate): also transform the layer mask if we
are transforming a layer and there is no selection (which causes the
entire layer to be transformed).
(gimp_paint_core_get_orig_image)
(gimp_paint_core_get_orig_proj): changed parameters to x,y,width,height.
Update callers accordingly and use x,y,width,height there too except
in gimpperspectiveclone.c which does better with a bounding box.
Use 0666 as permissions instead of 0644 and let the user's umask care
about restricting, so creating a file with open() behaves the same way
as with fopen().
Finally commit the patch from Luidnel Maignan, but don't spit messages
when the effected region is empty (core functions don't spit messages
either). Also got rid of some x2 and y2 variables that are not needed
any longer.
Add gimp_display_shell_sync_config() which copies the prefs settings
to the display shell and call it from gimp_display_shell_new() and
gimp_display_shell_fill().
Check that the cache itself is idle. Previously it would start during
transforms and long pyramid rendering ops and toss writes and large
seeks into the tile cache while it was potentially under heavy pressure.
Also increase its flush rate more to be more in line with modern system
and editing needs.
Although the clean-first strategy gives fast light-load performance,
it also degrades catastrophically under moderate cache pressure. LRU is
not as efficient under light load, but degrades more gracefully under
moderate and heavy load.
Add more profiling In order to analyze efficiency and behavior of the
tile cache. Profiling includes run-time indication of idle swapper
activity.
Also, fixes to original tile profiling code that appears to have
bitrotted some; it was no longer tracking changes made to tile memory
outside of tile.c.
Minor change to TILE_DATA_POINTER that restricts TILE_WIDTH and
TILE_HEIGHT to powers of two, but eliminates two integer divisions
(or, in reality, eliminates the over-complicated assembly resulting
from optimizing out two integer divisions in a C compliant fashion).
at each new GIMP session
Do not store the number of copies in the print-settings resource file,
but keep it in the print-settings that are attached to the image.
The questionable while loops to get GimpCoords.direction into a proper
range were used in two places. This commit fixes the actual problematic
place where, under some rare circumstances, the code used to go into a
loop for a very long time.
Pass the current coords to the function instead of setting them
on the paint core before calling it. Doesn't exactly make the code
in the paint tool simpler, but that needs further refactoring anyway.
Check for SHIFT, CONTROL and MOD1 explicitely so the code doesn't prevent
item activation for esoteric modifiers that are set by whatever X
component (like XKB).
Michael Natterer