| ID: | 2945 |
| From: | ge...@igor2.repo.hu |
| Date: | Fri, 19 Apr 2019 17:55:57 +0200 (CEST) |
| Subject: | Re: [pcb-rnd] bug: fixed short always remained orange |
| in-reply-to: | 2942 from Peter Stuge <pe...@stuge.se> |
| replies: | 2949 from Peter Stuge <pe...@stuge.se> |
On Fri, 19 Apr 2019, Peter Stuge wrote: >I don't seem to be able to create a test case for this with current >svn, so maybe it has been fixed. > >Towards the end of my board layout (with an older svn rev) I moved a >via so that it touched a trace of another net, and when I optimized >rats they were both highlighted in orange. > >I quickly fixed it and then re-ran both rats optimization and DRC, >but both the trace and the via still remain orange in my layout file, >even though both rats optimization and DRC pass. > >Attached is my attempt at recreating the problem, but here the orange >highlight disappears if I start moving either via or trace. Great! > >With my board .lht this isn't the case, however. I'm sorry, I don't >want to share it. This is "only" cosmetic so no big deal, I'll just fix >it up in my .lht. Some background that may help you recreating the problem: The orange thing is called the WARN flag. Since it's a flag, it's saved in the file (inherited idea...). And since they had so many flag bits in one big flag word, they had to reuse the bits, so rubberend and warn share the same bit, so you will see this bit saved as "rubberend" in the lht (inherited hack). So if you save a file while it has WARN objects, the state is preserved and you should see the same WARNs after load. Normally if you left-click anywhere in the drawing area, the WARN flag is cleared on all objects on the board. This wouldn't depend on anything else (whether your nets are still shorted). If this doesn't happen, that's clearly a bug. If that's the case, I'd try to re-create the problem not from scratch on a new board, but by removing any unnecessary object and net from the board where it is reproducible. There may be some unusual combination of object types and flags that triggers the bug. HTH, Igor2
Reply subtree:
2945 Re: [pcb-rnd] bug: fixed short always remained orange from ge...@igor2.repo.hu
2949 Re: [pcb-rnd] bug: fixed short always remained orange from Peter Stuge <pe...@stuge.se>
2952 Re: [pcb-rnd] bug: fixed short always remained orange from ge...@igor2.repo.hu
2964 Re: [pcb-rnd] bug: fixed short always remained orange from Peter Stuge <pe...@stuge.se>
2971 Re: [pcb-rnd] bug: fixed short always remained orange from ge...@igor2.repo.hu