Hi,
I tested Robert's HEAD revision and it was OK for me.
But for 2.0, in my mind, we should at least support the reversion of polylines, what has to be implemented first. Of course one can explode, revert and rejoin the polyline. In my eyes this will only be a workaround and a large source for issues and questions.
Next are the circles. A circle is a common shape for milling jobs. LC has no feature to show the direction of a circle. For milling it depends on the spindle rotation, tool and material if the job is done clockwise or counter clockwise.
But one can draw a circle from arcs which can be reverted, so it isn't absolutely needed.
Another question is, if it makes sense to implement reversion for ellipses and splines. From sight of CAM is isn't, because many CNC machines doesn't support this entities. For these cases we need a approximation with tangentially lines and arcs to fit the contour and that are entities which can be reverted.
In general this feature is useful for mechanical drawings to be used with external postprocessors for generating CNC programs from DXF, that is out of the question.
My vision of CAM features track another approach. When getting drawings from friends or customers it shouldn't matter in which directions entities are drawn. Even when drawing yourself, arcs are always draw CCW, but milling in most cases is CW. I don't want to redraw everything to fit the postprocessors needs. I want to show the postprocessor where to start and in which direction the tool should go or where it should stop. To find the path is the job of the postprocessor. But this is definitely not for 2.0.
I'm afraid that this doesn't simplify the question to merge or not.
Why not start now a branch 2.1 with Robert's features and keep the master until we release 2.0. If there are bug fixes on the master, we can merge them to 2.1. After releasing 2.0 we merge 2.1 into master and start the 2.0 release branch for bug fixes.
This or something like that is the way we handle it in my company. It works very well without much conflicts on merging.
Armin
investing less than half an hour into Search function can save hours or days of waiting for a solution