Commit Graph

960 Commits

Author SHA1 Message Date
Campbell Barton
4982f200fa move alloca define into its own header since its not related to BLI_array 2013-07-28 10:38:25 +00:00
Campbell Barton
0cb9a83113 code cleanup: remove unused code 2013-07-27 18:17:19 +00:00
Campbell Barton
7c58ec9337 use math functions for copying matrix/vector types, also cast const pointers for freeing (clang error's out otherwise) 2013-07-26 10:43:23 +00:00
Campbell Barton
f77405356a remove use_relative option from simple deform,
all modifiers should be using object transformations relatively.
2013-07-23 19:31:49 +00:00
Campbell Barton
e50c37d90f code cleanup: remove deprecated bevel code (unused since 2.64) 2013-07-23 14:28:19 +00:00
Campbell Barton
6f8913b498 replave isfinite with finite for msvc 2013-07-23 07:40:07 +00:00
Campbell Barton
0682ade250 fix/workaround for crash in ocean modifier when size is zero (causes invalid array index/lookups). 2013-07-22 21:02:45 +00:00
Campbell Barton
003df3d2de fix [#36247] Modifier Ocean (Geometry in Displace)- Subdivision Surface
regression in 2.68
2013-07-22 20:53:52 +00:00
Campbell Barton
f4609acedc optimization: avoid memcpy with shake key vertex array. 2013-07-19 10:40:20 +00:00
Campbell Barton
922320f86e fix own regression [#36154] Simple deform modifier doesn't recalculate normals
get_cddm and get_dm are called within modifiers so they wont ensure normals are valid, added an arg to optionally ensure valid normals.
2013-07-16 08:24:53 +00:00
Sergey Sharybin
1dd7156c4c Fix #36058: Displace Modifier errors using a baked Image and displace baking inconsistency between 2.67/2.68RC and previous versions
This was in fact really nasty bug, caused by multitex_nodes
function using global variable R (which is a copy of current
renderer). this variable is not initialized to anything
meaningful for until first rendering (preview or final)
happened.

Since multitex_nodes might be used outside of render pipeline,
made it so whether CM is on or off as an argument to functions
multitex_ext_safe and multitex_ext. Now multitex_nodes() is
only shall be used for stuff happening from render pipeline!

Also needed to make some changes to other places, so all the
usages of texture sampling knows for the fact whether CM is
on or off.

And one more change is related on behavior of dispalcement,
wave, warp, weightvg modifiers and smoke. They'll be always
using CM off since texture is used for influence, not for
color.

It's rather bigger patch, but it's mostly straightforward
changes, which we really need to be done.

Reviewed by Brecht, thanks!
2013-07-15 14:47:58 +00:00
Campbell Barton
3c4c2478b6 fix for own regression, face index ranges still need checking in some places. 2013-07-14 23:41:33 +00:00
Campbell Barton
83a172e40c fix for case where negative index was used when rendering particles with UV's, also removed redundant array check. 2013-07-14 13:18:05 +00:00
Campbell Barton
1c15beb6b2 remove NULL checks on fixed size arrays, also was calling BLI_testextensie_glob every time in the file selector with a blank string. 2013-07-13 14:16:59 +00:00
Campbell Barton
8a40444d6b fix bad uses of sizeof() with memory allocation. 2013-07-13 05:43:35 +00:00
Campbell Barton
a7858767f4 fix [#35858] Weight Paint: Hiding faces isnt flushing the flag to the vertices. 2013-06-26 02:47:56 +00:00
Campbell Barton
fed8f76098 fix [#35795] Edge split modifier do not show result in viewport with curves.
caused by recent changes in normal calculation, however curves were not being very smart about calculating modifiers (calling unneeded re-tessellation for every modifier)
2013-06-19 08:19:36 +00:00
Campbell Barton
d4cbd78a79 fix for recent changes in solidify, normals were OK but customdata for loops wasnt. 2013-06-15 08:59:54 +00:00
Campbell Barton
9161a4daa5 fix for own error in recent solitify refactor (r57402), face flip check was incorrect. 2013-06-14 16:10:32 +00:00
Campbell Barton
69153c7089 ifdef bugfix since its a little strange. 2013-06-12 10:03:13 +00:00
Campbell Barton
31e667c10e solidify: dont add poly-normal layer to the derived mesh, since this is no longer a convention. 2013-06-12 09:35:02 +00:00
Campbell Barton
8e2e590484 fix [#35710] Mesh explodes while using solidify modifier with tubular non-manifold base mesh. 2013-06-12 08:24:31 +00:00
Campbell Barton
0b316d72ca change to solidify modifiers own normal calculation function, dont attempt to calculate normals for edges with 3+ face users. 2013-06-12 07:53:54 +00:00
Campbell Barton
293b3f0bdf optimize solidify, no need to do edgehash lookups, the edges are known. 2013-06-12 07:16:13 +00:00
Campbell Barton
d00ca6eb2c solidify: reduce sign conversions. 2013-06-12 06:51:02 +00:00
Campbell Barton
3994ad8ced solidify: remove BLI_array realloc's. 2013-06-12 06:20:24 +00:00
Campbell Barton
da733a25a5 correct solidify normal calculation logic
- always calculate vertex normals since they are used as fallbacks.
- only calculate rim normals if the normals are not already flagged as dirty.
2013-06-12 06:06:19 +00:00
Bastien Montagne
67a5d1b660 Follow up to r57354: WeightVG and Wave modifiers too need to handle tex id refcount ;) 2013-06-10 18:12:07 +00:00
Sergey Sharybin
bfbac9b448 Fix #35678: Duplicating an object with Displace modifier does not increase the reference count of the texture 2013-06-10 14:04:03 +00:00
Campbell Barton
c423e3ed8f code cleanup: warn undefined compiler defines. 2013-06-08 05:24:55 +00:00
Campbell Barton
8a5e3a8425 fix [#35628] Dynamic paint waves over ocean modifier doesn't work
Caused by my recent normal calculation changes, added dependsOnNormals callback which was missing for ocean modifier (it assumed input normals were set).
2013-06-05 19:22:12 +00:00
Campbell Barton
e03bbcec65 fix [#35453] "copy mirrored uv coords" doesn't work
- made precision configurable.
- report a warning when doubles are found since they cause problems.

added Polygon.center attribute to avoid calculating in python.
2013-06-05 03:10:29 +00:00
Campbell Barton
790e9d9fa0 fix [#35311] Planar Decimate / Limited Dissolve fails to merge some adjacent faces
optionally limit by face flipping, also added support to delimit by material and edge crease.
2013-06-03 05:07:16 +00:00
Campbell Barton
dfad9b0c09 fix [#35555] Collada: export destroys mesh in some cases
add arguments to calculate normals when converting to bmesh:
  BM_mesh_bm_from_me, DM_to_bmesh

This gives some speedup to undo (which didnt need to re-calculate vertex normals), and array modifier which doesnt need to calculate face normals at all
2013-06-02 23:20:49 +00:00
Campbell Barton
ad0305281b fix [#35545] Weight paint with mirror modifier does not accurate represent bone influence 2013-06-02 22:40:45 +00:00
Campbell Barton
ec30e3f00e corrections to modifiers from recent normal handling changes
- solidify didn't define a dependsOnNormals callback (which it should have)
- build wasn't passing on dirty normals.
- decimate wasnt setting dirty normals.
2013-06-02 04:09:29 +00:00
Campbell Barton
e648ca8862 use booleans for modifiers and api callbacks. 2013-06-02 03:59:19 +00:00
Campbell Barton
075a910d72 fix regression in reducing normal recalculation [#35595],
pass the dirty-normal-state from the input dm to the output.
2013-06-02 00:20:26 +00:00
Campbell Barton
d7f51f83f6 remove CD_MASK_NORMAL from CD_MASK_DERIVEDMESH, bmesh merge included this but its not needed.
Now add asserts to make sure this layer is only added once the modifier stack has been calculated.
this saves normal layer being calculated whenver vertex normals need updating.
2013-05-30 18:09:19 +00:00
Campbell Barton
09e11ad6ef modifier stack: lazy initialize normals
many modifiers were calculating normals, when those normals were ignored by the next modifier.
now flag normals as dirty and recalculate for modifiers that set use `dependsOnNormals()` callback.

Quick test on mesh with 12 modifiers (mostly build type), calculated normals 6 times, now it only runs once - so this will give some speedup too.
2013-05-30 17:36:43 +00:00
Sergey Sharybin
9e05f6571f Make sure bool will always have the same size in C and C++
There were an issues with data structures defined in headers
and being used by both C and C++ on systems with stdbool
unavailable.

This happened because bool in this case will be defined as
unsigned int, which is 4 bytes. But C++'s bool is only 1
byte and this lead to alignment issues.

Now bool is always 1 byte, also made sure there's no situation
like bool foo = BitField & BitFlag, which could give overflow
issues. Use (BitField & BitFlag) != 0 instead.

Fixes #35553: Compositor broken (Backdrop & Preview)
2013-05-29 11:49:39 +00:00
Campbell Barton
225c5fee6b move BLO_sys_types.h -> BLI_sys_types.h (it had nothing todo with loading)
remove MEM_sys_types.h which was a duplicate.
2013-05-28 19:35:26 +00:00
Campbell Barton
a70fa65592 optimize mirror merging, remove array reallocation, replace with fixed size arrays. 2013-05-27 20:11:12 +00:00
Campbell Barton
ec8d277c64 BLI_math rename functions:
- mult_m4_m4m4 -> mul_m4_m4m4
- mult_m3_m3m4 -> mul_m3_m3m4

these temporary names were used to avoid problems when argument order was switched.
2013-05-26 18:36:25 +00:00
Campbell Barton
28937676f3 fix for out of bounds memory access in solidify (harmless with guardedalloc padding but bad practice) 2013-05-25 00:22:04 +00:00
Brecht Van Lommel
c4ee1f2658 Fluid simulation manual number of threads option now doesn't set the number
of threads to the number of cores when the fluid is created. Rather it is now
set to 0 which means "use the number of threads specified for the scene".
2013-05-21 11:37:42 +00:00
Brecht Van Lommel
698f1b5603 Fix #35317: crash with boolean modifier using an object whose library linked
mesh datablock went missing.
2013-05-12 09:14:07 +00:00
Campbell Barton
0ee45c9301 more optimal method of calculating the normal for the solidify modifier.
When adding 2 unit length vectors, the length can be used to calculate the angle.
2013-05-08 14:33:02 +00:00
Campbell Barton
f74201190d code cleanup: remove references to BLI_rand.h 2013-05-08 12:59:35 +00:00
Campbell Barton
a9fb183901 rename BLI_ghashIterator_notDone() -> BLI_ghashIterator_done()
was renamed fairly recently but other similar iterators not negated
	like this, would prefer to keep it as it was
2013-05-08 12:58:11 +00:00