Issue was in BLI's rotation_between_vecs_to_quat(), which did not handled correctly cases where both vectors are colinear.
Patch by Campbell Barton and me.
Issue originaly tracked down by Yan Shi, many thanks!
rather then getting the longest edge, get the edge which which is most different from the 2 others ends up giving more useful results: for an isosceles triangle it returns the base weather its longer or shorter then the other sides.
- Even preserves thickness but can give unsightly loops
- Smooth gives nicer shape but can give unsightly feather/spline mismatch for 'S' shapes created by beziers.
This is an example where smooth works much nicer.
http://www.graphicall.org/ftp/ideasman42/mask_compare.png
Notes:
*This implements a quite simple algorithm, which simply checks angles (actually, absolute cosines) of created tri and remaining face (which may be a tri, quad, or more NGon), so that both are "best" (ie avoid as much as possible too much narrow/wide corners), and also checks the new edge is OK (i.e. does not goes "out" of original face).
*Incidently, it fixes a typo in that bm_face_goodline() func!
*It's quite performant (a bit quicker than previous code, as far as I have tested it) and prevent creation of completely flat triangles as much as possible, but it's far from being a "best" solution (as it is still a "progressive" one)!
*It also introduces a new math func (in BLI_math_vector.h), cos_v3v3v3, which computes cosine (ie dot product of normalized vectors) and is roughly a quicker replacement for angle_v3v3v3, when real angles are not needed.
- makes wireframe from faces.
- options similar to inset (even offset, relative scale)
- copies face settings and loops (uvs, vcolors)
- optionally replaces the existing geometry.
- memset(..., 1.0); // isnt valid
- memset(pointer, sizeof(pointer)) // was using the sizeof the pointer, not the size of the array, since this was to fill in alpha values it was obviously wrong.
Problem was that area calculation of polygons was done relative to the xy plane, and with a very obscure (to me at least) algorithm. That meant that vertical ngons would get 0 area.
Commented initial code in case this is a strange optimization case that someone wants to use and used a cleaner algorithm: first project vertices to the ngon plane, defined by the normal of the ngon and the center (mean) of the ngon vertices. This will only be exact for convex and mostly planar ngons, still it is much better than the previous code.
Also fixed memory leak when stretch display was on.
Vector.angle_signed(other)
for 2D vectors to get the clockwise angle between them.
in BLI math its called - angle_signed_v2v2()
shorthand for...
atan2f((v1[1] * v2[0]) - (v1[0] * v2[1]), dot_v2v2(v1, v2))
also corrects compile error in last commit.
