* Part/Toponaming: tests for flushElementMap and resetElementMap
* Added test for resetElementMap()
* Added test for flushElementMap()
* restored ComplexGeoData::resetElementMap to RT fork
* modified TopoShape::resetElementMap to match the restored signature of ComplexGeoData::resetElementMap
* added checks in TopoShape::makeShapeWithElementMap to avoid segfaults
* added check in TopoShape::setElementComboName to avoid segfaults
* restored testFindSubShapesForSourceWithTypeAndIndex
* modified test saveDocFileWithNoElementMap to match the restored signature of ComplexGeoData::resetElementMap
* Part/Toponaming: tests for flushElementMap and resetElementMap
* Removed comments in ComplexGeoData::resetElementMap()
* Changed TopoShape::resetElementMap():: from virtual to override
* Removed comment in TopoShape::resetElementMap()
* Added comments about differences between main and RT fork
* Removed comment in test saveDocFileWithNoElementMap
* Updated expected test result in makeElementCompoundTwoShapesGeneratesMap
* Updated expected test result in resetElementMapTest
* Fixed typo in test flushElementMapTest
* Part/Toponaming: tests for flushElementMap and resetElementMap
* updated expected tests results after previous methods modifications
---------
Signed-off-by: CalligaroV <vincenzo.calligaro@gmail.com>
Remove adjacent coincident and nearly-coincident points from Path.
This is based on PR#5276 by sundtek, but only collapses adjacent points
on the Path. It also addresses (IMO) most of the feedback on the original
PR.
see also discussion on the forum:
https://forum.freecadweb.org/viewtopic.php?f=15&t=42755&start=10#p555566
The descriptions were sometimes not accurate (major/minor radius for ellipses). I have also tried to improve the constistency a little.
Forum topic:
https://forum.freecad.org/viewtopic.php?t=85396
This adds G43 tool height compensation to the centroid CAM post-processor using
a similar approach to the G43 support in the MACH3 post.
Specifically, G43 is emitted at every tool change by default and a new
`--no-tlo` post argument can restore the old behaviour of no height
compensation being emitted. These both match how the MACH3 post does it.
* Change the calculation of the b parameter of the hyperbolic arc
so that it does not give 0/0 when cos(phi) = 0
This gave problems if the first two points defined in the sketcher had
the same x-coordinate.
If endpoint-to-endpoint tangent constraint exists between the connecting points
of the curves to be joined, also apply C1 continuity (i.e. a knot multiplicity
of degree-1).
