ST_AddEdge* fails to use index on edge table on face split
|Reported by:||strk||Owned by:||strk|
The edge table update on face split fails to make use of the edge table index by passing the return of ST_LineInterpolatePoint(geom) and a literal to ST_Contains.
As a result, topology population has poor scaling performance. See http://lists.osgeo.org/pipermail/postgis-devel/2014-January/024076.html
… I wonder how could such a defect be put under regression testing … (or we'll need a performance testsuite)