I normally don't like to rant but OMFG! I just looked at the code path that lead to this wonderful piece of code and I'm now regretting that. You just cannot unsee that. See if you can follow the logic (assuming logic was actually used to write this code) backwards from Triangle::NeighborAcross(). I'm not sure removing the invalid null reference checks is a good idea even though I've seen the compiler complain about it and agree with it. In what universe do you blindly convert a potentially NULL pointer into a C++ reference, pass it to the caller, and go merrily on you way? This really needs to be rewritten using pointers with the appropriate NULL pointer checks along the way. It's hard to imagine this ever worked properly.
On 9/27/2016 1:33 PM, Simon Richter wrote: > > Triangle::NeighborAcross returns a reference, which must refer to a valid > object whose address cannot be 0. Thus, this test is nonsensical. > --- > polygon/poly2tri/sweep/sweep.cc | 16 ---------------- > 1 file changed, 16 deletions(-) > > > > _______________________________________________ > Mailing list: https://launchpad.net/~kicad-developers > Post to : [email protected] > Unsubscribe : https://launchpad.net/~kicad-developers > More help : https://help.launchpad.net/ListHelp > _______________________________________________ Mailing list: https://launchpad.net/~kicad-developers Post to : [email protected] Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
