> I submitted my first PCB bug report to SF last month (#2946254), and > shortly after added a patch that fixed the problem. I must admit that > the lack of response was discouraging - but I fully appreciate that > the developers are time poor (I am also!). > > BTW thank you to Rikster, for taking the time to try the patch & > confirm that it fixes the bug, and posting the result back to SF :-) ... The patch looks good to me (although I've only skimmed it). It might warrant a definition of what a "freckle" is, if that term isn't use elsewhere. The optimisation is probably fine to add. A complete fix would address the issue in the auto-router as well. I'm happy to apply the patch, but I'm heading home now, as its getting late. Someone bug me to apply the patch!
It looks to me like the SQ() macro risks integer overflow when squaring the lengths. I too only glanced at the patch so maybe I'm wrong. This is one of the difficulties in getting patches on a fast track. The internals of pcb are ugly and hard to understand in a lot of ways, and even many of the developers don't fully understand them (myself included these days!!). pcb has a huge amount of cruft from its 20 year life. It is extremely easy to create a patch that on its face looks good, appears to fix the problem at hand and passes certain tests, BUT, introduces ugly lurking bugs that can be a nightmare to find and resolve. harry
_______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user