https://bugs.documentfoundation.org/show_bug.cgi?id=173555
--- Comment #9 from [email protected] --- Created attachment 208530 --> https://bugs.documentfoundation.org/attachment.cgi?id=208530&action=edit A video of interesting behavior regarding the current warp method. So what it is is very interesting. With the current implementation we have some significantly unintuitive behavior, and here's an example of it. The first part showcases a simple curve that is actually invariant under movement of the top-left corner. It doesn't get warped at all, which is very interesting. This isn't really an edge case either, it could realistically happen, because the set of Bézier curves to which this applies is pretty large (those with control points that lie exclusively at corners or at invariant edges of the warp rectangle, even if the curve itself seems like it should be moved quite significantly). The second one is where we have another unintuitive behavior which is that two curves that look identical can get warped in completely different ways due to, potentially, disparities in control-point placement and segmenting invisible if one does not look at the underlying points. I don't think this is intentional, hence why I would regard this as a bug rather than a feature request. -- You are receiving this mail because: You are the assignee for the bug.
