X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Date: Thu, 8 Jul 2021 15:11:14 +0200 (CEST)
From: Roland Lutz <rlutz AT hedmen DOT org>
To: "Vladimir Zhbanov (vzhbanov AT gmail DOT com) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
Subject: Re: [geda-user] Thank you - some requests and an offer
In-Reply-To: <YOayUYFK9b1GiBoB@lepton>
Message-ID: <alpine.DEB.2.21.2107081429220.1847@nimbus>
References: <20210707101437 DOT 2c7c63a6 AT queeg> <YOayUYFK9b1GiBoB AT lepton>
User-Agent: Alpine 2.21 (DEB 202 2017-01-01)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
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 Thu, 8 Jul 2021, Vladimir Zhbanov (vzhbanov AT gmail DOT com) [via 
geda-user AT delorie DOT com] wrote:
> For me, the biggest issue was introducing xorn (and its dependency 
> python2) into the project, which effectively hid my work on legacy 
> gnetlist of that time.

*sigh*

I haven't made any secret out of the fact that I was working on 
refactoring gnetlist, a task which I believe you agree was direly 
necessary.  So, you can't really complain about the fact that you were 
duplicating work.

In contrast to you, though, I believe the result of this refactoring 
process should behave as faithfully like the previous version as possible.

Yes, that means going the extra mile to understand and preserve various 
less-known aspects of the netlisting process.  I was willing to put in 
that extra work; you weren't.  Which is kind of okay.  Just be honest 
about the fact that your netlister probably won't give the expected result 
when run on an existing schematic.

For the record, here is an exhaustive list of the differences in the 
output of gnetlist before and after I refactored it:

     http://git.geda-project.org/geda-gaf/tree/xorn/tests/netlist/README

The fact that no-one even tried to assemble such a list for lepton-netlist 
speaks volumes.


> The work had been already done in legacy gnetlist just before replacing 
> it with gnetlist2/xorn so nobody might notice it.

Calling this `gnetlist2' is really obfuscating the truth.  The gnetlist 
sources are in the xorn/ directory because they are using the new gEDA/gaf 
libraries; they are much more faithful to the previous version than the 
netlister you re-wrote in Guile.

All existing backends continue to be able to be used with gnetlist.  In 
addition, I ported all backends, including the backends partslist1, 
partslist2, and partslist3 you mentioned, to the new Python format:

     http://git.geda-project.org/geda-gaf/tree/xorn/src/backend

For anyone even a bit familiar with Python, this lowers the entry barrier 
for writing custom backends tremendously.

Roland