Closed spuder closed 8 years ago
Before
After
Notice that I moved the 1 pad and the 2 vias so they aren't so close together.
Still pretty new at KiCad but it passes the Diagnostic Rules check. I don't know if it matters, but this was done with a 10mils grid instead of a 50 mils grid
Thanks for the fix! I'll have to set up visual diffing for KiCad to be able to check contributions because upon regenerating the netlist thousands of lines get changed, just like in this pull request. Luckily, there are some tutorials on the subject. Please let me get back to this a little bit later.
From my research, you can run the following script.
https://github.com/madworm/KiCad-Stuff/tree/master/scripts/visual-diffs_on_gerber-files
https://www.youtube.com/watch?v=diMIU0YYB9w
I can't quite get it to work on OSX yet. It may be possible to use travis-ci to automatically generate a diff file and save it as an artifact.
Here is a script that generates a gerber in kicad
http://www.evilmadscientist.com/2011/improving-open-source-hardware-visual-diffs/
If this script could be modified to output to a .pdf instead of a gerber file, then this blog has instructions on how to generate a diff file.
http://www.evilmadscientist.com/2011/improving-open-source-hardware-visual-diffs/
I spent a few hours on this without progress. I'll pick this back up soon.
Closing. In writing the scripts to create the visual diffs, I noticed a problem. The pad takes away from the thickness of the path. Will open a new pull request when fixed.
Fixes issue #8