[Crystal-develop] Bug in tritri.cpp

[Ralph Campbell]

Coverity found an obvious bug in HandlePointsInTriangle()
in libs/csgeom/tritri.cpp but I'm not sure what the fix is.

It accesses tri2[2] but clearly declares tri2[2].

This code hasn't changed in a long time so it's an old bug.

[Christian Van Brussel]

Post by Ralph Campbell
Coverity found an obvious bug in HandlePointsInTriangle()
in libs/csgeom/tritri.cpp but I'm not sure what the fix is.
It accesses tri2[2] but clearly declares tri2[2].

Hi,

The methods involved are manipulating triangles, that is arrays of 3
vectors, not 2.

Accessing tri2[2] is therefore OK but it is a priori designed to be
defined as tri2[3].

I fixed the problem in r39976, and looking at that I found another
problem in the coplanar case, it was missing an iteration on the second
triangle, see r39977.