kicad-developers team mailing list archive
-
kicad-developers team
-
Mailing list archive
-
Message #37649
Re: Handling SEGZONEs
-
To:
kicad-developers@xxxxxxxxxxxxxxxxxxx
-
From:
Wayne Stambaugh <stambaughw@xxxxxxxxx>
-
Date:
Wed, 26 Sep 2018 14:55:47 -0400
-
In-reply-to:
<CALHbTmagW4xokxwKZYBd=aXHkpsrqgnxrn6+i1Ajgq2GazGmxw@mail.gmail.com>
-
User-agent:
Mozilla/5.0 (Windows NT 10.0; WOW64; rv:60.0) Gecko/20100101 Thunderbird/60.0
Seth,
I think this is due to an unhandled exception in the file loading code.
If I open the board with a debug build of kicad, I get a warning about
the board being an old board file version and the board opens just fine.
With release builds, I don't see the board file version warning and it
immediately crashes.
Cheers,
Wayne
On 9/26/2018 2:22 PM, Seth Hillbrand wrote:
> So we have an archaeological (and very valid) bug report over at lp:1794571
>
> The board in question (thermo.brd attached) has a SEGZONE fill without
> an associated area. As far as I can tell, the "zone" doesn't have
> boundaries so much as just segments. This fails when we load into our
> connectivity as a zone because we assume a bounded polygon even for the
> old segment-based fills.
>
> I loaded it up in 4.0.5 and it looks like the attached image. I can't
> see that there was any way to edit the segment fill other than just
> deleting individual segments.
>
> Converting into 5.0, we have (I think) three options:
>
> 1) Drop all segments that are in unbounded zones
> 2) Convert the unbounded zone segments to plain segments (this looks
> like the 5.0 image attached)
> 3) Work out a bounding box when loading and set that to segment-based fill
>
> All three options will result in changing the board. Option 2 results
> in the most fidelity to the original design but is hardest to edit.
> Option 3 is easiest to edit but has the biggest chance for breaking
> something. Option 1 is the furthest from the original board design but
> easier to edit than Option 2 and safer than Option 3.
>
> If there was never a good way to edit these type of fills, then I think
> we should go with option 2. But I'm open to suggestions here.
>
> -Seth
>
> _______________________________________________
> Mailing list: https://launchpad.net/~kicad-developers
> Post to : kicad-developers@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~kicad-developers
> More help : https://help.launchpad.net/ListHelp
>
Follow ups
References