Revision ec470d7 by jens verwiebe November 6, 2013, 23:19 (GMT) |
OSX/scons: for completeness, added WITH_BF_OPENMP switch |
Revision f07ed3c by Lukas Toenne November 6, 2013, 21:21 (GMT) |
Removed the automatic "link swapping" feature from the node link operator: When a link was being dragged to an already connected input, the existing links were shifted to the next free socket. This was originally intended as a way to speed up workflow for math and mix nodes, but more often than not it just gets in the way. Most binary (or n-ary) functions are not even commutative, i.e. changing the order of sockets does not usually produce the correct result. Also this includes the more common case where one actually wants to replace a socket, which then requires a second click to remove the shifted connection. All in all this is not a helpful feature. |
Revision 232c97f by Lukas Toenne November 6, 2013, 21:08 (GMT) |
Fix for error in r61159: the new gpencil_new_layer_col in UserDef is supposed to be a 4 float RGBA color, but has only 3 floats. This was overwriting tweak_threshold in UserDef and breaking transform operators ... |
Revision f392ebd by jens verwiebe November 6, 2013, 20:59 (GMT) |
OSX/scons: huge change, now you can set a user-config.py just with the desired variables changed |
Revision ad34a5c by Bastien Montagne November 6, 2013, 20:56 (GMT) |
Fix [#34675] *AFTER 2.69* Info view shows duplicate operators with incorrect values of args Refactored a bit WM api to generate operator's pystring, now it can also handle correctly macro operators. Thanks to Campbell for the review! |
Revision 5cd28bb by Mitchell Stokes November 6, 2013, 19:40 (GMT) |
BGE: Fix for #37335 "Moving the camera with a key (after the recent BGE cleanup commits) now crashes the game" reported by Ace Dragon. CcdPhysicsEnvironment->GetCharacterController(); was missing a NULL check. |
Revision 61c4110 by Lukas Toenne November 6, 2013, 19:21 (GMT) |
Patch #37274: Circle select for node editor, by Henrik Aarnio (hjaarnio). Circle select was missing from node editor, and C key was assigned to now defunct "show cyclic dependencies". This patch remaps the key and adds circle select operator. Functions to check intersection between rctf/rcti and a circle were also added to rct.c for code cleanliness and consistency. |
Revision 549ed3d by Lukas Toenne November 6, 2013, 18:56 (GMT) |
Removed the "Show Cyclic Dependencies" operator for nodes. This operator is an old relic implemented for showing cyclic node connections as red links. This is not necessary any more, the operator simply called the generic node tree update function, which is happening anyway after all relevant node operators (if it doesn't that has to be considered a bug). It has been suggested to better use the C key for circle select, this remains to be discussed. |
Revision 7398600 by Lukas Toenne November 6, 2013, 17:46 (GMT) |
True grid snapping for nodes: This snaps nodes to the actual background grid instead of using incremental offset (which is not useful for nodes). Increment snapping has been disabled for nodes to avoid confusion, grid snap is now the default as it seems to be the most wanted and easy to use mode. Absolute grid snapping happens in a somewhat generic function 'applyGridAbsolute', which could also be used for objects and other transforms later on. It is conceptually similar to the 'project' snapping option, in that it calculates a delta vector for each element on top of the overall transform, which places each node on the grid. Node transform now uses the top-left node corner for TransformData->loc. The transform center is still the average of node centers, so that scaling and rotation works nicely. snapGrid*** functions have been renamed to snapGridIncrement*** to distinguish better between incremental and absolute grid snapping. |
Revision 480d431 by jens verwiebe November 6, 2013, 15:46 (GMT) |
OSX/scons: remove obsolete compile flag |
Revision c81c085 by jens verwiebe November 6, 2013, 15:42 (GMT) |
OSX/scons: remove obsolete compile flags and silence warnings same time |
Revision 5557332 by jens verwiebe November 6, 2013, 15:37 (GMT) |
OSX/scons: remove another outdated conditional |
Revision 3143536 by Howard Trickey November 6, 2013, 15:27 (GMT) |
Fix knife bug exposed by valgrind. Was reading cage coordinate from those of existing BMVerts even for newly created verts that don't have cage coordinates there. |
Revision 3e6c369 by jens verwiebe November 6, 2013, 14:03 (GMT) |
OSX/scons: remove local MAC_MIN_VERS var and use the always identical MACOSX_DEPLOYMENT_TARGET instead |
Revision 96aaea4 by jens verwiebe November 6, 2013, 13:56 (GMT) |
OSX/scons: more cleanup of redundant vars, preparation to remove local vars to not need to cp whole config as user-config later |
Revision 5f084ac by jens verwiebe November 6, 2013, 12:52 (GMT) |
OSX/scons: simplification for getting the system version ( major, minor ), use sw_vers -productVersion instead of uname -r, we must not redine it then from darwin version -> osx version |
Revision b8f22a0 by Lukas Toenne November 6, 2013, 12:44 (GMT) |
Syncing methods for Color Balance node LGG and ASC-CDL modes. The settings for either mode are converted into equivalent settings of the other. This keeps the result of both modes roughly the same and mimics the previous behavior when settings were shared by both modes (but not equivalent). NOTE: Due to the use of additional sRGB conversion in the LGG mode the result is not entirely accurate, this should perhaps be fixed. Settings for each mode are kept in their own color values nevertheless, this avoids potential problems with float precision. |
Revision b91e841 by Lukas Toenne November 6, 2013, 12:44 (GMT) |
Fix #37333: Bad default value in Color Balance. Use independent offset/power/slope variables for the CDL mode in color balance node. This avoids stupid default values in particular for offset, which would be 1 when just using the lift value for it. |
Revision b9aa637 by Lukas Toenne November 6, 2013, 12:44 (GMT) |
Removed the DNA storage for LGG lift and inverse gamma in the color balance node. These values were always calculated at execution time, so there is no need to keep them around in DNA data and no forward compatibility break either. Only reason they were stored in DNA before is that the old compositor had no other means of keeping precomputed values around for every pixel than storing the DNA node data, with new compositor this is no longer necessary (values are stored in operations). |
Revision 50ac2ee by Joshua Leung November 6, 2013, 10:59 (GMT) |
Grease Pencil: User-Pref for setting the default colour of newly created layers |
|
|
|


Master Commits
MiikaH:n Sivut a.k.a. MiikaHweb | 2003-2021