Incorrect orientation if when the patches are rotated by PI-radians about splinepy

[j042]

is this part of an existing test? a_s and b_s are flipped - jacobian determinants will be negative and you should at least use splines that aren't flipped

[deltaLuki]

is this part of an existing test? a_s and b_s are flipped - jacobian determinants will be negative and you should at least use splines that aren't flipped

No, this is a test from me. Does this mean that we don't need to consider the case where the patches are rotated 180 degrees?

[j042]

see Quick Start

...
The indexing of the control grid is defined by iterating
lower-indexed dimensions first. But if you prefer a
...

splines in your test case aren't doing that.

[jzwar]

Jacobians can be negative for multipatch geometries in these tests, as only the orientation is compared. However, I think the expected orientations vector might be wrong. From a quick drawing on a napkin I get

[0, 1, 1, 3, 1, 0, -1, -1]

[j042]

Nice. If you wanna add such test, please explicitly mark as negative jacobian multipatch test