These values change because they are computed on the *linearization*
of those curves, and refactoring linearization codes results in
slighly different values.
NOTE: adding or not adding the control point would also affect these
results
- prevent creation of a QgsPointV2 with a non-point WKB type
- fix bounding box was not invalidated for some modification routines
- if WKT type is PointZ/M/ZM and not enough coordinates specified,
initialise extra coordinates to 0
- when creating geometry from WKT, upgrade dimensionality of geometry
if coordinates are 3/4 dimensional
- match dimensionality of collections to child dimensionality
- fix area of curves was non-zero if curve is closed
- don't consider m values when testing for curve closedness
- add unit tests for closedness
- add unit tests for CircularStrings, CompoundCurves, CurvePolygon,
tests with geometries with Z/M values
(inc geometry count, node count, ring count, area, centroid,
x/y min/max)
TODO:
- invesigate whether Z/M is required for GeometryCollection WKT
data type if children have Z/M dimension
- fix QGIS handling of mixing length and perimeter for collections
resulting in incorrect length calculation for mixed Collections
