detection of changes to dependencies. This should resolve issues with
Cygwin builds using makefiles which have bizarre behavior if 'make clean'
was not used before compile.
from brian hayward (bthayward)
Detailed description:
Currently, when an armature deformed object's mesh is replaced by the ReplaceMesh actuator, the new mesh fails to deform to the armature's movement.
My patch fixes this by properly replacing the deform controller along with the mesh (when appropriete).
For instance, if one had an animated character using any of the standard deformation techniques (armature, ipo, RVK, or AVK), that character's mesh would currently be prevented from changing mid-game. It could be replaced, but the new mesh would lack the controller which tells it how to deform. If one wanted to dynamiclly add a hat on top of the character's head, it would require storing a secondary prebuilt character (mesh, armature, logic, ect...) on another layer FOR EACH HAT the character could possibly wear, then swapping out the whole character when the hat change was desired. So if you had 4 possible hat/character combos, you would have 4 character meshes, 4 armatures, 4 sets of logic, and so on. I find this lack of flexibility to be unresonable.
With my patch, one could accomplish the same thing mearly by making one version of the character in the main layer, and adding an invisible object atop the character's head (which is parented to the head bone). Then whenever it becomes desirable, one can replace the invisible object's mesh with the desirable hat's mesh, then make it visible. With my patch, the hat object would then continue to deform to the character's head regardless of which hat was currently being worn.
*note 1*
for armature/mesh deformations, the new mesh must have properly assigned vertex groups which match one or more of the bones of the target armature before the replaceMesh call is made. Otherwise the vertices won't react to the armature because they won't know how. (not sure if vertices can be scripted to change groups after the game has started)
*note 2*
The added processing time involved with replacing the object's deform controller is negligible.
This will add Minneart diffuse and WardIso specular to our shader menu.
Minneart gives nice control over darkness/brightness areas, the wardIso
over 'plastic' style sharp or fuzzy specular.
Webpage is being made with nice samples. Will be in release log.
Jorge: one change is in the do_versions, you inserted it on wrong location.
With a few tests I have discovered that when quitting Blender the filelist
in SpaceFile doesn't get free'd.
storage.c:
I have replaced strdup for the relname member in BLI_builddir with BLI_strdup.
and malloc with MEM_mallocN for the string member in BLI_addstrings().
filesel.c:
Of course also had to replace free with MEM_freeN in freefilelist().
In freespacelist (space.c) I added call to freefilelist for the SPACE_FILE
space type.
It seemed trivial enough, so I updated all systems (makefiles, xcode, scons, scons win installer), please complain if something went wrong. Mostly it was just removing release/bpydata references, since the release/scripts dir is already worked on recursevely, handling dirs inside it.
For the scons nsi file writer script I had to write code for each new dir, but it can be recoded recursively, too (in fact joining all release stuff in a single dir tree with nothing else would be a good idea, making installation code simpler). Since it's just python and I have a little more time now, I can help scons managers if they still need.
Thanks Campbell Barton for reporting.
When curve objects have no front/back faces, the bevels were only drawn
for the outer part. Now it also makes interior bevels, so yo can
have a curve become a 'solid' ring easily.
Note: only works when you disable filling option "Front" and "Back".
- after grab/duplicate the softbody didn't get a reset signal
- added 'copy properties' for softbody settings
- duplicate object didn't copy softbody yet
- Added browsing for vertex group for "Goal" in SoftBody buttons.
- Means the default name "SOFTGOAL" isn't needed anymore
- temporally, on file read, the "SOFTGOAL" vertex group is set, if exists
- removing vertex group possible too
- changed softbody.c code to use this
Scripts:
- Thanks Jean-Michel Soler (jms) for updated versions of dispaint, fixfromarmature and unweld (also renamed to remove version part).
- Thanks Bart for the upgraded VRML exporter (great doc webpage!). It is available as VRML 97 and the original VRML 2 is for now still there, to help users testing the new version. For the next release the old one should be removed, of course.
- New script: Scripts Config Editor (Scripts win -> Scripts -> System). Scripts with config options (simple data that is to be set according to user needs or preferences) can use this facility instead of providing a gui and writing config files to disk themselves.
- Added new menu: System, available in the Scripts win.
- Updated sys_info.py, help_browse.py and the AC3D importer and exporter.
- Removed use of the Scrollbar and added arrow keys and mouse wheel support instead in Daniel Dunbar's old doc_browser.py. The scrollbar events handling doesn't exist, Ton suggested removing the scrollbar from the API months ago. For now its ref doc is gone and no bundled script uses it, until we get time to implement it properly.
- Added module BPyRegistry.py with functions to handle reading / writing config files automatically to the scripts/bpydata/config dir.
- Removing dir release/bpydata and its contents (moved earlier to release/scripts/bpydata/)
- Bug #2379: made small changes to bevel_center's ui to fix a problem reported by Alexander Ewering (intrr):
http://projects.blender.org/tracker/?func=detail&atid=125&aid=2379&group_id=9
BPython:
- Thanks Campbell Barton for new functionality: Blender.Get() now can also return all the paths from the user prefs -> file paths win and there is a new function: Blender.sys.expandpath() to transform Blender paths (those starting with '//' and ending with '#') to absolute paths.
- Added function Blender.ShowHelp(), to open the Scripts Help Browser with a given help page -- just a time saver for scripts.
- Improved function Blender.Run() to also work with gui and file select scripts.
- Found a (new?) crash related to NMesh.PutRaw when creating a new object while in edit mode. Leaving / entering edit mode fixes the problem, so a check for obj created, edit mode and leaving / re-entering it were added to the code for now (gdb didn't help much, no backtrace)
- doc updates, including splitting intro page in two, with bpython related stuff (registering / documenting / configuring scripts and command line mode (thanks Chris Want for "use system variables to pass parameters to scripts" idea).
- Registry: functions have been updated to support writing to / reading from disk, for the config editor -- only simple config data supported, for large amounts coders should write to a file themselves. This is done with a new parameter: Registry.GetKey(keyname, True) will also search for the key on the config dir, if not already loaded; equiv. for Registry.SetKey(keyname, dict, True). Data is only written to / read from disk when needed and only scripts already used (assuming they support this functionality) will have config data saved.
Also adds docs for above API.
My testing cases seem to work fine. Please test to make sure I have not
missed anything!
Stivs, I got ascotan's signoff on thses. But please make any adjustments that
you deem needed.
Johnny
I can now say that I declared it fixed erroneously before, and
I hope that when I call it fixed now it can really, truly be laid
to rest.
Sometimes when you escape a transform on an object that is the
target of an IK constraint on an armature bone, the rotations were
not restored correctly on the bones in the IK chain. This commit
fixes this problem in a rather brute force manner.
so we can decide later what proper timing is meant to be.
i've prepared something in static float sb_time_scale(Object *ob) [softbody.c] (ton :) )
'hacked in' particle collision system to match softbodies needs
naa .. i don't realy like the hack to int pdDoDeflection(..) in kernel .. effect.c (but it works :) )
so .. until we'll have a 'nice' collision detector this is what 'softbodies simulator can live with'
did not remove intentionally test function static int sb_deflect_test(..) for further discussions
http://mitglied.lycos.de/mosebjorn/hidden/ sbcol6.blend should work for a while
The addition in 2.36 to correct bumpmapping for View and Object rotation,
made many nice bumpmap setups in older files less nice. Especially apparent
when you apply "flipped" mappings in "Map Input" Panel, like map the X to Z
and the Z to X. A correction then gives wrong results.
Since the "correct bumpmapping" is actually a new option, causing extra
rendertime too, I've made it an option now. With the "Map To" Panel being
fully used, I've put the new button option (temporally I hope) in the first
"Texture" Panel in material context.
Will be updated in release log page!
While in camera view, going to local axis rotation (R, X, X) the view
became instable. Was still using the view3d view matrix, instead of
initialized TransInfo->viewmat.
Martin; you could check on the mouse motion in this case, for some reason
it only uses horizontal motion?
ALso; renamed button in InfoWindow "Disable sound" to "Disable Game Sound".
Rotation acted weird in perspective mode. Works much better using viewinv
instead of persinv to get the rotation axis. What would be even better is
using real view vector depending on transformation center (with
getViewVector), but that can wait a bit.
Meta Element resize needed some code to reset radius changes or axis size
changes when switching from one to the other (my fault, I didn't think of this
when telling Jiri how to implement it).
(Looks like big commit, but is mostly just change of API call!)
Particle emittors now can be parented to an armature Bone, and give the
correct path for each particle. Note that this doesn't work for deform!
And, for each particle the entire armature is evaluated, all actions and
NLA strips.
It used to work a little while BTW, but the code just called ALL armatures
and made ALL deforms again. Was quite slow... thats why the API call
change: do_all_actions() now accepts Object * to only do that one. With
a NULL it now does all still. Will disapppear in recode of course...
