Blender Git Commits

This commit makes it so openal finally works on solaris.

Kent

Essential cleanup for mess involved with reading files, initializing UI and
patching versions for UI settings.

Currently four different levels of routines for .blend file reading exist;

/* interface level */
1) BIF_init() -> calls 3
2) BIF_read_file() -> calls 11, optional 4
3) BIF_read_homefile() -> calls 11 or 12, and then 4
4) init_userdef_file()

/* kernel level */
11) BKE_read_file() -> calls 21, and then 14
12) BKE_read_file_from_memory() -> calls 22, and then 14
13) BKE_read_file_from_memfile() -> calls 23, and then 14
14) setup_app_data()

/* loader module level */
21) BLO_read_from_file() -> calls 24
22) BLO_read_from_memory() -> calls 24
23) BLO_read_from_memfile() -> calls 24

/* loader module, internal */
24) blo_read_file_internal()


Note:
- BIF_read_homefile() has additional UI initialize calls, like windows fullscreen and executing commandline options

- Reading from memory (12) only happens for the compiled-in .B.blend

- The "memfile" here is a name I gave to the undo "file" structure. Which is constructed out of memory chunks with basic compression features.

- the kernel function setup_app_data() sets globals like "current screen" and "current scene".

So far, so good. The levels as mentioned here clearly distinguish UI from kernel, and should enable for example game loading (runtime) or background (no UI) loading. In the past years however, 'bad level' dependencies were added, and especially the patches for 'file versions' were added in too many places. The latter is evidently a result of the problem that the "UserDef" struct cannot be initialized/patched if there's not a need for a UI.

Here's how the flow goes in four different cases:

----- Starting up Blender, in foreground with UI --------------------

- creator/creator.c, main() -> calls 1
- If the commandline contains a filename, it calls 11

----- Starting up Blender, in background without UI --------------------

- creator/creator.c, main() -> calls 11 if the commandline has a filename

Note: no Userdef is read, nor initialized. Please note that this was already an existing problem for using Yafray, not setting proper file paths in background mode. The Yafray paths don't belong in the User menu.

----- Starting up Blender as a runtime executable --------------------

This only has calls to 22

----- Loading a file from within the UI (with F1, CTRL+O, using pulldowns) -----

Only calls allowed to 2. It detects if a UserDef has been read too, and in that case the init_userdef_file() will be executed.

Hope this is understandable :)

-Ton-

Bug fix 1982

Three buttons, showing float values, had only precision of 2 digits.
Made it three.

Bugfix #1981

Pad-minus key didn't work in typemode transform().
Very easy, just flipping an && with || :)

Sneak in little feature for texture editing pleasure;

- New channel for "Map to" added, "Warp"
- Use the slider next to the option to set amount of influence the texture
will have on the coordinates of the next texture.
- Warp uses for this the same values as for Normal or Displacement mapping
- Warp remains active for all channels, until replaced (or zeroed).

More tweaks related to bump mapping quality;

While going over the code, I found out the "nabla", the size of offset
vectors for calculating derivatives of a texture, is a built in constant.
Even worse, the value was different for new noise types (musgrave etc).
So I've added a new slider for it in the procedural texture panels, which
by default is set to 0.025, the value of the old constant. Also made sure
it works with equal effect in all procedurals.

NOTE: a small Nabla will give sharper, detailed bump, but the effect also
becomes smaller, correct that in the Mapping Panel of materials.

For better & compliant control over the bumpmapping, I've also included
the Colorband output in derivatives calculus, so the bump output then
matches the color created. It's also a nice tool to finetune output of
textures for bumpmapping in general.

Bug fix; clicking on the rightmose 'item' in ColorBand didn't activate it.
Found out the ColorBand was slightly drawn off (2 pixels).

Added support for normalmaps.

Bug fix #1978

Stencil in World didn't work with RGB textures (like what returns from
colorbands).

Tweaked support for normalmaps to be OK.

http://www.blender3d.org/cms/Normal_Mapping.491.0.html

main changes:
- The "Map Input" in Blender for normals was negated
- Range for Z in normal map is Z (0-1) to Blue (0-255)

Bug #1970

Objects being vertex-parented to a mesh, which in turn was being deformed
by a curve or lattice, were never update correctly.
Whilst this is a typical issue for our beloved new Dependency Graph, I
consider it worth at least solving for now. An issue since Blender 1.0
you know. :)

Solution is only working for this specific case; vertex-parent to deformed
object.

Fix the case when the physics system starves the logic & renderer of CPU time. If this happens, we reduce physics precision to maintain some framerate.

Fix numerical precision issue in physics. Dividing by a number too close to zero would make the impulse response normal large hence the jittering.

Set the default physics tic rate to 90 Hz

Don't try to interpolate from prop ipo actuators

Bug fix #1972

AKEY didn't toggle open/close outliner. It'll be back temporal, until
definitely defined what to do with selections in it.

Outliner crash reported bu K-rich;

- with outliner visible
- alt+d (linked dupli) an armature

Bugfix: normal mapping for textures wasn't rotated back to view space,
causing weird results on camera moves or object rotations.

Added: support for "normal maps". See for example:
http://members.shaw.ca/jimht03/normal.html

The Image panel in Texture buttons has new option "Normal Map" for it.
When this is used, normals are read straight from RGB values, and blended
with the current normal.

Bugfix/feat request... in textured view mode the Build Effect didnt show.

Reported on IRC;

Converting subsurf to mesh didn't display new Mesh object in correct wire,
when susurf was optimal.

BPython, two bug fixes:
- #1911: http://projects.blender.org/tracker/?func=detail&atid=125&aid=1911&group_id=9
BOOL properties were returning True always, small mistake in logic.c's getData method.
- #1944: http://projects.blender.org/tracker/?func=detail&atid=125&aid=1944&group_id=9
G.vd was not being checked against NULL in Window.ViewLayer, causing a crash when this function was called from a command line script. Now it returns an error in such cases.
- small doc updates, tiny minor change in Object.c.