098cf90956
use strict flags for scanfill, also replace shorts with unsigned shorts and ints/bools in some cases.
2013-09-01 03:37:45 +00:00
9e561a641e
remove callback BLI_localErrorCallBack from scanfill, was here for years and only ever wrapped printf.
2013-08-28 02:14:24 +00:00
27df6a3b54
scanfill curves, ngons, masks had their own memarena code and would allocate a new one for every fill.
...
now use BLI_memarena and support passing the arena into the fill function, so the arena is re-used, when scanfill is called in a loop.
2013-08-28 02:07:54 +00:00
f97a4bd254
code cleanup: more zero as NULL pointers.
2013-08-07 03:36:05 +00:00
a050ddf279
remove unused callback
2013-06-26 03:42:45 +00:00
f8c37b083c
remove own optimization in scanfill, in rare cases it caused problems, reported as [ #35861 ]
2013-06-24 18:22:59 +00:00
99b55ef6eb
add asserts when scanfilling or triangulating with zero length normal.
2013-06-02 22:54:00 +00:00
5018ea5e29
real fix for [ #35097 ], (curve cap flipping).
...
previous commit was incorrect, the face flipping depended on the orientation of the curve.
fix by passing the bevel direction to the fill function so we can have a reliable front/back.
This also gives some speedup for all curve filling since we can avoid calculating the normal since its already known.
2013-04-26 21:04:12 +00:00
1143b658a0
fix [ #34947 ] Alt-F Fill not working
...
Regression in r54579.
2013-04-10 23:59:37 +00:00
ece766ee7e
code cleanup: unused defines, remove unused scanfill success value from BLI_scanfill_begin().
2013-04-10 23:52:07 +00:00
25fe5c20aa
code cleanup: rename ScanFillVert.h --> edge_tot, SF_VERT_UNKNOWN --> SF_VERT_AVAILABLE
2013-02-21 17:15:55 +00:00
7fa256815f
skip fix from r54579 when holes aren't used (keeps bmesh ngon filling fast)
2013-02-15 12:57:11 +00:00
964c35771c
Bug fix #34177
...
Blender's triangulator has been rescued :)
This commit fixes errors with concave holes inside polygons.
Simple explanation:
Blender "ScanFill" works by sorting vertices from top-left to bottom-right, and connecting
these vertices with a sorted list of edges they have.
The inner loop then goes over every vertex, its edges, and tries to make triangles by
checking vertices that are next in the list.
- if the triangle has points inside: it creates an edge to this vertex, and continues
- else: add new triangle.
Very simple, fast and efficient. But it needed one more check for the first step: it should
check every vertex inside the triangle, and pick the best vertex for an edge based on forming
the sharpest angle with the tested edge. That solves the case for concave holes.
Blender ScanFill was coded 20 years ago, and is an own invention. I wanted a triangulator that
just fills any collection of polygons, including with holes.
No idea if this was ever published in a paper!
2013-02-15 12:26:47 +00:00
942ad6d9cd
style cleanup
2013-02-10 17:06:05 +00:00
4a427d8e0d
style cleanup
2012-12-29 01:54:58 +00:00
f0c1bc830c
add option to BLI_scanfill_calc() - BLI_SCANFILL_CALC_HOLES, gives some speedup for BMesh ngons which never have holes and ensures predictable triangle count (totvert - 2), which is needed for pre-calculating array size.
2012-12-27 06:39:27 +00:00
f9e339ef00
fix/workaround [ #33281 ] script goes into not responding
...
scanfill remove-doubles pass assumes ordered edges (as with curves), otherwise it can hang.
workaround this problem by skipping removing-doubles for mesh ngons, since this isnt such a common case as it is with curves and we can just not support it.
2012-11-26 23:18:04 +00:00
b867f9f17e
style cleanup: comments & spelling
2012-11-18 01:22:31 +00:00
aeba4950c3
style cleanup
2012-10-27 10:42:28 +00:00
f70d2c65d8
rename api functions...
...
- minf, maxf, mini, maxi --> min_ff, max_ff, min_ii, max_ii
2012-10-23 13:28:22 +00:00
1767b65846
style cleanup: also rename bmesh_decimate.c --> bmesh_decimate_collapse.c
2012-10-23 03:38:26 +00:00
2fb8292005
style cleanup
2012-09-16 04:58:18 +00:00
a0ae47f06c
add some missing NULL checks, a few parts of the code used a pointer then checked it for NULL after.
...
also made it more clear that some areas assume the pointer isnt null (remove redundant NULL checks).
2012-09-11 02:18:27 +00:00
b96c622015
style cleanup
2012-08-11 22:12:32 +00:00
c41e1e434a
code cleanup: replace MIN2/MAX2 with minf/maxf
2012-07-29 16:59:51 +00:00
1597ad9377
style cleanup
2012-07-01 09:54:44 +00:00
c8ebfe1d12
code cleanup:
...
- use bmesh iterator macros in more places
- rename scanfill variables (were using same names as mesh faces/verts which was confusing)
2012-05-13 14:47:53 +00:00
bddc7dfc20
style cleanup: remaining BLI files.
2012-05-12 15:13:06 +00:00
4c5502bfd6
code cleanup: function naming for BLI functions.
2012-05-05 00:23:55 +00:00
8765dfccf7
style cleanup: correct typos
2012-04-21 14:14:58 +00:00
d7542be6ac
code cleanup: more minor changes to scanfill
2012-04-18 14:36:56 +00:00
abcaec0edc
code cleanup: replace magic numbers in scanfill with defines, also use some vector functions where possible
2012-04-18 14:30:57 +00:00
ca913254aa
fix for eternal loop in scanfill,
...
The cause for this is bmesh faces which have zero area have their normals set to a fake value (Z-Up), this would break scanfill, possible fixes are to calculate the faces normal each time or tag as invalid but its easiest to avoid the eternal loop with an extra test.
2012-04-18 14:06:59 +00:00
6389301eb5
fix for own error in recent scanfill updates, scanfill can hang if run on a polygon with no area.
2012-04-18 05:52:18 +00:00
195d6c1b1a
minor speedup for scanfill, dont calculate the normal if its already known - use for editmode ngon filling.
2012-04-16 18:24:49 +00:00
67f8e3a3a7
inline function for "Newell's Method" used for normal calc.
2012-04-16 16:49:37 +00:00
e889fa4678
improve scanfill for uneven ngons, previously scanfill would use the most angular corner, but this made non planer ngons rip frequently (often reported problem).
...
now calculate the normal as with ngons.
2012-04-16 16:24:55 +00:00
0635f8101c
make scanfill threadsafe (wasnt threadsafe before BMesh merge but before the merge it didn't need to be) - now rendering uses its better if its threadsafe.
2012-04-16 06:48:57 +00:00
fb1e60762f
style cleanup: scanfill, spelling and use NULL rather then 0
2012-04-16 05:23:40 +00:00
79693e4543
code cleanup: avoid confusion with incorrectly named argument to BLI_edgefill(), was 'mat_nr', now 'do_quad_tri_speedup'
2012-04-16 05:03:04 +00:00
b8a71efeba
style cleanup: follow style guide for/with/if spacing
2012-03-24 07:52:14 +00:00
69e6894b15
style cleanup: follow style guide for formatting of if/for/while loops, and else if's
2012-03-24 06:18:31 +00:00
577bef62eb
Fix #30543 : tessellating ngons with small angles was failing. There was a quite
...
arbitrary limit of pi/24 for picking 3 vertices to use for computing a normal,
now it uses vertices with the maximum angle between them.
2012-03-20 19:32:31 +00:00
dfdfa3d51b
code cleanup: replace macros VECCOPY, VECADD, VECSUB, INPR - with BLI_math funcs.
...
added copy float/double funcs: copy_v3fl_v3db(), copy_v3db_v3fl(). 2d & 4d too.
2012-03-09 06:04:17 +00:00
95670e03a0
style cleanup / comment formatting for bli/bke/bmesh
2012-03-03 20:19:11 +00:00
ed04c21374
code cleanup: use float vector size in function definitions, and const's where the values are unchanged.
2012-02-28 14:05:00 +00:00
1cad189e32
fix harmless but annoying memory leak prints, "newmem", now free all scanfill memory on exit.
2012-02-19 22:36:24 +00:00
a368e6771a
- remove some unused editmesh functions.
...
- copy & rename EditMesh stricts for use with scanfill (remove unused members)
2012-02-19 22:17:30 +00:00
d74c564235
svn merge ^/trunk/blender -r42333:42361
2011-12-02 23:02:21 +00:00
06c3d5bd09
de-duplicate dominant axis calculation, exact same checks were in 6 different places.
...
added function: axis_dominant_v3(...)
2011-12-02 22:14:20 +00:00
