X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Message-ID: <51CC7D11.5060403@sonic.net>
Date: Thu, 27 Jun 2013 10:57:37 -0700
From: Dave Curtis <davecurtis AT sonic DOT net>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:16.0) Gecko/20121028 Thunderbird/16.0.2
MIME-Version: 1.0
To: geda-user AT delorie DOT com
Subject: Re: [geda-user] gschem off-page signal cross-referencing script pushed
 to github
References: <51CA333A DOT 2020008 AT sonic DOT net> <CACNnPR=2pCaUX2Me5U9XLwY+E7+S4O04JpWz_+L07Fr5vHAhGw AT mail DOT gmail DOT com>,<CAGRhJMaFibo61=3X=zbZk7kAy3vkhWQjxCMPsAHA0pemdQZ5vg AT mail DOT gmail DOT com> <51CB04E5 DOT 8020700 AT sonic DOT net>,<CAGRhJMbtifzJg1JUy0NRqkdROcoZ36-XwKoG4t-7Cqrh2sRitg AT mail DOT gmail DOT com> <COL126-W1FD8299BD3FE3DA80460AC6740 AT phx DOT gbl>
In-Reply-To: <COL126-W1FD8299BD3FE3DA80460AC6740@phx.gbl>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Reply-To: geda-user AT delorie DOT com

On 06/26/2013 03:01 PM, Hannu Vuolasaho wrote:
> I tried both of them with the test cases and liked.
I'm glad you find it useful.
>
> My 2 cents and feature request.
>
> Now that there are tools collect off page nets, output them to djboxsym file from the each page and create outline/contents page.
Well, but page boundaries don't necessarily correspond to block 
boundaries, and off-page I/O doesn't correspond to block I/O.   What I 
could see, though, is making a different component for "block I/O" and 
construct the block from that.
>   Unfortunately it would require some autorouting stuff to make the full schematics
That too.... many years ago I was involved in a project to do an 
auto-router for schematic nets.  It turns out that it is a different 
problem from PCB auto-route -- a PCB route is mainly a connectivity 
issue.  A schematic auto-route is a readability and aesthetic problem... 
very different strategies required.

>   and might confuse netlister
So... don't send the block diagram sheets to the netlister.  You don't 
have to netlist *.sch
>   but for documentation purposes it would be highly valuable tool for block diagram generation or even hierarchy handling.
I'm going to leave hierarchy to somebody else.

A related tool that I've though about, though, is a "signal 
dictionary".  Given an agreed-upon 'sheetnum' attribute, it would be 
easy to scan sheets for 'netname' attributes and build a dictionary of 
all named nets, not just those that go off-page.  A sorted list of 
signals and sheets where they appear could be printed as a report or 
turned into a schematic sheet full of text.

-dave
>
> Best regards,
> Hannu Vuolasaho 		 	   		
>