Blender Git Loki

BLI_polyfill2d_test: script to generate test data

Correct test ifdef in polyfill

Cleanup: use stubs for eigen gtest

BLI_polyfill2d_test: Try flipped x/y axis

In T52834 this makes a difference.

BLI_polyfill2d_test: add test for T52834

Commented since it currently fails.

Cleanup: redundant casts

Missing from last commit

Math Lib: clamped rounding utility functions

Polyfill Beautify: option to rotate out of degenerate state

Needed for 3D iterative edge-rotation to avoid flipping when projected
from different angles,
but could keep zero area faces in 2D polygons.

Fix error in recent boolean changes w/ quad split

Avoid bias when calculating quad split direction

Some error checks weren't being done in both directions
when calculating the best split direction for a quad.

BMesh: use less involved check for edge rotation

Was using function for edge rotation which was doing unnecessary checks
Use the face normal and BLI_polyfill_beautify_quad_rotate_calc directly.

Correct recent error in boolean quad split check

Fix T52291: Boolean fails w/ co-linear edged ngons

This means boolean tessellation wont match viewport tessellation
however it's needed to avoid zero area triangles causing problems.

Boolean Modifier: add debug options

Only show & use when running in debug mode.

BLI_utildefines: sync w/ 2.8

Revert "Fix T52871: `BLI_polyfill_beautify_quad_rotate_calc_ex` was mistakenly considering the state as degenerated"

This reverts commit a8f11f5422afa01608817755eea310e987f896be.

Result is no longer symmetrical, will investigate a different fix.

Fix T52871: beauty fill error

Only lock tri's facing different directions.
Needed because scanfill creates zero area faces.

Beauty fill was skipping small faces

Polyfill Beautify: half-edge optimization

Was using an edge hash for triangle -> edge lookups,
updating triangle indices for each edge-rotation.

Replace this with half-edge which can rotate edges much more simply,
writing triangles back once the solution has been calculated.

Gives ~33% speedup in own tests.