X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Date: Wed, 18 Jan 2012 11:59:42 +0100
From: Gabriel Paubert <paubert AT iram DOT es>
To: geda-user AT delorie DOT com
Cc: Krzysztof =?utf-8?Q?Ko=C5=9Bciuszkiewicz?= <k DOT kosciuszkiewicz AT gmail DOT com>
Subject: Re: [geda-user] BUG: disappearing dots in gschem
Message-ID: <20120118105942.GA10816@visitor2.iram.es>
References: <20120118090450 DOT GB28925 AT visitor2 DOT iram DOT es>
 <20120118093353 DOT 10720 DOT qmail AT stuge DOT se>
 <CAJisMU2b9O5pq3KTTbAMYdy8ZnkQh6m0kC3rwU1AjfrR7uL-Kw AT mail DOT gmail DOT com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <CAJisMU2b9O5pq3KTTbAMYdy8ZnkQh6m0kC3rwU1AjfrR7uL-Kw@mail.gmail.com>
User-Agent: Mutt/1.5.20 (2009-06-14)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by delorie.com id q0IAxkaZ015504
Reply-To: geda-user AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: geda-user AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

On Wed, Jan 18, 2012 at 10:50:26AM +0100, Krzysztof Kościuszkiewicz wrote:
> W dniu 18 stycznia 2012 10:33 użytkownik Peter Stuge <peter AT stuge DOT se> napisał:
> > Gabriel Paubert wrote:
> >> gschem: skip drawing endpoint cues on fully connected nets.
> >>
> >> Affects-bug: lp-707064
> >>
> >> causes some dots to disappear.
> >
> > I too think this is a problem.
> >
> > Keep in mind that there are different types of dots.
> >
> > Recently there was a report about an unconnected net, which also
> > lost it's endpoint cue.
> >
> > What gives? Krzysztof, maybe you can talk a little about the theory
> > behind the change?
> 
> Probably I have not tested all the corner cases appropriately.
> Dumping all the cue types in one function does not help either.

Indeed, I've given up trying to understand the code.

However after sending the mail, I reverted the commit 
(actually applied the patch with -R) and the dot is back. 

> In the bug report it seems the lower pin of the diode connects with
> two net segments, while the top pin connects to net in a midpoint.

Correct. The dots only seem to appear in the middle of segments, but 
not when you have a mix of pins and net segments. And, as far as I remember
there was never a dot when three pins coincide, which may happen fairly
frequently in passive circuits as shown in the new attachments; these do
not cover all the cases, since up to 4 pins/segments may coincide easily, 
and more if you use non horizontal/vertical net segments or, shudder, pins. 

> Possibly this is the missed corner case - I'll investigate this later on.
> Analysis of John Doty's report on the issues with implementation is
> also on my TODO list.

Good.

	Regards,
	Gabriel