X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Message-ID: <1437505227.1322.13.camel@ssalewski.de>
Subject: Re: [geda-user] My gschem 1.9.1 is breaking rubber bands
From: Stefan Salewski <mail AT ssalewski DOT de>
To: geda-user AT delorie DOT com
Date: Tue, 21 Jul 2015 21:00:27 +0200
In-Reply-To: <201507211821.t6LILegK016798@envy.delorie.com>
References: <1437430095 DOT 1321 DOT 17 DOT camel AT ssalewski DOT de>
	 <alpine DOT DEB DOT 2 DOT 11 DOT 1507211410260 DOT 2035 AT nimbus>
	 <3B92001A-596F-4595-9DD3-15AAD3C3B1FF AT noqsi DOT com>
	 <1437494996 DOT 729 DOT 7 DOT camel AT ssalewski DOT de>
	 <0C548460-0355-4EFF-9495-75DE7AD13A16 AT noqsi DOT com>
	 <CAOFvGD4ijprW7hoaJNYjo55A1=mnQuWCiK7Sku_WqC3NeUzgQw AT mail DOT gmail DOT com>
	 <201507211821 DOT t6LILegK016798 AT envy DOT delorie DOT com>
Content-Type: text/plain; charset="UTF-8"
X-Mailer: Evolution 3.12.11 
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
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 Tue, 2015-07-21 at 14:21 -0400, DJ Delorie wrote:
> > How about a popup dialog ("Netlist change detected. Are you sure you want
> > to make these changes. Yes/No") that could be enabled in gSchem's settings
> > file? Every time a drag is completed, if it was enabled, it would compare
> > the old netlist to the new one and ask if things change.
> 
> Connect a pin.  Are you sure?  Connect the next pin.  Are you sure?
> Connect another pin.  Are you sure?
> 
> Let's make sure that is NOT the default ;-)
> 
> If we want an option that prevents drags from changing the netlist,
> let's fix drag so that it doesn't change the netlist in the first
> place.  "Hi, I made a mistake during that drag.  Would you like me to
> get away with it? [Yes/No]"

No, of course no popup. More like gvim does it when a text buffer is
modified -- it adds a + char after the filename.

I still think that this feature is useful, it generates no clutter, and
it is easy to code for gschem, because gschem can already generate
netlists. Useful, because I can remember that I draw schematics, and
later cleaned it up, taking care not changing wiring.

Fixing rubberband drag: I have not already started thinking about coding
rubberband, so I am not sure how difficult it is to keep all
connections. But I can imagine that when dragging nets, it may occur
that nets segments are dragged onto other nets ends, indicating new
connections. So it is not really easy. (Maybe we skip that critical
position when dragging?)

For extern tools, as John suggested: Of course, we can use gnetlist to
create a temporary list before cleanup, and one more after cleanup, and
use diff to compare them.

Extern tools may have benefits -- I consider removing the front lamps
from my car and use a torch instead :-)