gnetlist can create an invalid netlist

Any luck on this one? I'm really dying here, since working around this would entail many changes to my schematics.

Thanks,
Matt

Still investigating this. I'm starting to wonder if I will need to extend some of the data structures of gnetlist in order to support this new behaviour and if I do, it won't be a simple fix.

Either way, attaching two netnames to a single net is asking for trouble, in essence you are allowing the netlister to make a naming decision for you. I will probably continue that trend (netlister will pick one of the nets to and rename all the others) if I find a fix for this.

-Ales

Verified that this bug is still present.

It seems the bug is triggered when one net is renamed multiple times, each time to another name.
This patch adds a check for this and works correctly for the attached testcase.

It might be worthwile to add a hook in gnetlist to resolve rename conflicts.
This would allow for deterministic choice of the resulting netname.

I'm not very familiar with these parts of gnetlist, but can only suggest that if it works (and still passes the gnetlist test-suite), please commit away.

Btw.. the test-suite is run by "make check" if you hadn't found that already.

You might find that it fails due to insignificant changes in output. If that is the case, the easiest way to proceed (for the "common" part of the suite) is to regenerate the golden files with:

cd gnetlist/tests/common/
./run_tests.sh --regen

git diff will then show you what changed.

The other parts of the test-suite might need manual attention if output has changed.

Test suite passess, fix commited to HEAD. Same fix works for #698395 as well.

Hi,

How are you enjoying developing gEDA so far?

I'm unfortunately very busy with other things, so am not really on as
top of gEDA coding as I'd like to be.

I was wondering if there is anything we other developers can do to help
you get into things? Are there any projects aside from the work you've
been doing already fixing bugs which you were wanting to take a crack
at?

Perhaps we should organise a (virtual) developer sprint at some point on
IRC at some point in the next couple of months. That should let us all
get together online, get to know each other a bit more and discuss any
plans we have for the future.

Best wishes,

--
Peter Clifton

Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA

Tel: +44 (0)7729 980173 - (No signal in the lab!)
Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me)

Oops.. meant to send last reply by private email, not to a bug reply address.