Hi Ben and all, On Wed, 2008-11-19 at 22:35 -0800, Ben Jackson wrote: > On Thu, Nov 20, 2008 at 12:55:51AM -0500, DJ Delorie wrote: > > > > > I can't repro the MM line garbage. > > > > I can. Set a 0.25mm grid and draw a bunch of lines that *look* like > > plain 45 lines (i.e. no obvious corner, like you'd often get with _/, > > Ah, ok, I drew a zillion lines in different modes and apparently never > made a 'V' with 45s. I can repro with your steps. >
There is an other issue: there are probably a zillion garbage lines in 'our" PCBs out there, the tiny lines nobody noticed. That is in for a cleanup solution (a plug-in or some code at load time of the pcb correcting the wrongful doings from the past), when "old" designs are revised. If I understood the solution below correctly, this only addresses the case to prevent any future wrongful doings at the creation time of traces. > I can think of two quick ways to fix this: > > 1) Change the _/ modes to not create those segments if they're not > electrically necessary. This is a pretty good fix and you'd only see > odd behavior if you were in rubberband mode and grabbed the corner to > move it. It wouldn't count as "connected" in that case. If you grabbed > one of the *lines* it *would* count (due to changes I made last year to > fix other problems with tiny connector segments that appear when you go > from grid to pad). > > 2) Change the line removal code to check both endpoints for tiny > segments and delete them. > > And a less quick way: > > 3) Add the requested 'remove net' feature which would presumably be > more popular than 'remove line' in cases where these tiny segments would > appear and it would handle this cause automatically. > Kind regards, Bert Timmerman. _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user