CCGSubsurf has already a function to disable calculation of normals,
but seems it wasn't used. This patch changes subsurf UV and
subsurf_calculate_limit_positions() to not calculate normals, and also
not allocate space for them.
Should be no functional changes, just a small speedup/less memory use
during subdivision for these cases.
Code review link:
http://codereview.appspot.com/5558058/
The structs stored in the anonymous void *default_value in bNodeSocket are now handled completely inside node_socket.c. All allocation/freeing/duplicating for this has been replaced by the appropriate calls to generic API functions (declared in NOD_socket.h).
This will make the default value handling more reliable for future node socket code. Group socket copying and value conversion has also been moved into the generic socket API file.
Detect feedback loop and do not bake to images detected in this loop and show
nice warning message in such cases.
It's a way which wouldn't overcomplicate code trying to duplicate images and so
without real benefit.
=============
* refactoring, separate common functionality to functions.
* enable each uv to check independently for other stitchable uvs. This allows to limit stitch uvs belonging to the same vertex, while previously this was impossible.
* made changes to island rotation when calculated from sole uvs when using midpoint stitch. It should now be evenly distributed across islands.
There are still some glitches with vertices that have multiple uvs per island for the snap island + midpoint stitch combination. The solution is to precalculate the static island rotation/translation in this case. If I haven't managed to make this work correctly I may turn this combination off before release. It's more like an extra feature than really necessary.
The paint cursor code is fairly muddled still and needs futher cleanup
(commented in the new file.) Over half the paint_stroke code was just
called from the cursor draw function.
There should be no functional changes from this.
through Object.dm_info('SOURCE/DEFORM/FINAL')
this is to help tracking down issues with modifiers where loosing data
layers between modifiers can cause bugs, also to helo with comparing
bmesh/trunk's modifier stack.
it exposes sensor.controllers and controller.actuators
this is how the data is exposed in blender.
to have controller.sensors or actuator.controllers is not that straightforward
This script was defining an operator within the panels draw function, while its possible to support this its really asking for trouble.
the fix is to raise an error when this happens.
also fix crash passing non classes to register_class/unregister_class
Curve points of tracks curves now can be selected for X and Y channels
separately and can be moved along Y axis of curve viewer, points currently
can't change frame they belong to. This allows to smooth spikes caused by
unwanted marker jump.
Also fixed some mistakes in cancel transform in cases when locked tracks
were being trying to be moved; fixed incorrect calculation of marker speed
for curve drawing.
- make red outline more obvious (was hard to see when the camera was selected)
- remove text, red-alert text IMHO should only be used when something is wrong (example - its used when FPS cant keep up), but this is just information.
realize out UI isnt dealing well with added view3d options but dont think its good to solve by displaying options as text overlay.
red outline for the camera is just a hint which users notice when enabling the camera, think its acceptable.
=========================
Documentation: http://wiki.blender.org/index.php/User:Psy-Fi/UV_Tools
Major features include:
*16 bit image support in viewport
*Subsurf aware unwrapping
*Smart Stitch(snap/rotate islands, preview, middlepoint/endpoint stitching)
*Seams from islands tool (marks seams and sharp, depending on settings)
*Uv Sculpting(Grab/Pinch/Rotate)
All tools are complete apart from stitching that is considered stable but with an extra edge mode under development(will be in soc-2011-onion-uv-tools).
and length of the meta strip properly, only its contents. Also overlap check was
done too soon giving false positives, should be after meta strip update.
