delorie.com/archives/browse.cgi   search  
Mail Archives: geda-user/2012/11/20/16:31:00

X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Mime-Version: 1.0 (Apple Message framework v1085)
Subject: Re: [geda-user] git mirror of gedasymbols.org
From: John Doty <jpd AT noqsi DOT com>
In-Reply-To: <mnasn9x81v.ln2@skate.rswarbrick>
Date: Tue, 20 Nov 2012 14:29:47 -0700
Message-Id: <7C90D042-C0C8-4C9E-B639-ED3939468EE9@noqsi.com>
References: <20121116030224 DOT 5c7750ee AT akka> <201211160225 DOT qAG2PrVD005630 AT envy DOT delorie DOT com> <20121116035513 DOT 14519 DOT qmail AT stuge DOT se> <201211160357 DOT qAG3vUXo017504 AT envy DOT delorie DOT com> <20121116041216 DOT 16057 DOT qmail AT stuge DOT se> <201211160428 DOT qAG4Sh3h018631 AT envy DOT delorie DOT com> <llhhn9x25t DOT ln2 AT skate DOT rswarbrick> <k88cd4$3i2$1 AT ger DOT gmane DOT org> <20121117163351 DOT 14988 DOT qmail AT stuge DOT se> <5C6FC5CD-A358-476D-A8CA-C9AF2E72A914 AT noqsi DOT com> <20121117185520 DOT 25698 DOT qmail AT stuge DOT se> <E2A7C157-E0AB-437B-B8FD-1DA96C64AF39 AT noqsi DOT com> <k89b87$fip$1 AT ger DOT gmane DOT org> <B8C80ED4-E795-41D2-9D76-513A7D09666C AT noqsi DOT com> <CAN0Jx--dJGoQSMhsiRyycZ7xyPTX2dX1srZQ6SZV9N6s=ui-qg AT mail DOT gmail DOT com> <817F06A8-B916-4D3A-96EF-5542768A657F AT noqsi DOT com> <20121120031136 DOT 67cce538 AT akka> <mnasn9x81v DOT ln2 AT skate DOT rswarbrick>
To: geda-user AT delorie DOT com
X-Mailer: Apple Mail (2.1085)
X-MIME-Autoconverted: from quoted-printable to 8bit by delorie.com id qAKLTrDf029406
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

On Nov 20, 2012, at 5:47 AM, Rupert Swarbrick wrote:

> Kai-Martin Knaak <kmk AT familieknaak DOT de> writes:
>> John Doty schrieb am 18. November 2012:
>> 
>>>> My git cvs-import is      s s s s o o o o    s s s l l l o o o o w w
>>>> w w w  
>>> 
>>> That's not what I'm talking about, and I don't understand why you
>>> would want to import the whole gedasymbols website into your git
>>> repository.
>> 
>> Because other users do viable symbols and footprints, too. And because I
>> don't want to loose access to said symbols and footprints if a storm
>> hits some place on the other side of the globe.
> 
> I think you guys might be talking past each other. I'm a bit of a git
> fanboy, but I've been following this discussion quite closely and I
> think I understand DJ's point of view.
> 
> As far as I can see, Kai-Martin is concerned with the possibility of
> gedasymbols going down and also with integrating his local git
> development tree with the stuff uploaded to gedasymbols.

He's already got that, as he has set up a git mirror. But wants to be able to upload to the main repository via git. That collides with the simple permissions-based submodule model that gedasymbols uses.

> 
> John's point is that other people's symbols/footprints don't logically
> need to be in ones own development tree. If I understand him correctly,
> his model of how gedasymbols works is that you make symbols and
> footprints on your machine and then when you upload them to gedasymbols,
> you're basically publishing them. Then the versioning that CVS allows
> isn't particularly relevant, but doesn't hurt.

Yes.

But there's a piece you've missed: the current repository has, in git terms, 62 submodules, one for each author. Each author's access is restricted by file permissions. The submodule structure transparent to the user unless they attempt to trespass. While git has submodules, they have some subtle properties, and are not so transparent.

> 
> Please tell me if I've got your positions wrong!
> 
> Now, as far as I can see, it's not too difficult to make both of you
> happy :-)
> 
> CVS is indeed slow, but presumably no-one needs the development history
> for others' symbols. As such, DJ could stick a cron job on
> gedasymbols.org that makes a downloadable snapshot of the current
> repository (minus CVS directories) once a day. Then people using git
> could just download "the rest" in one go.
> 
> I'm not sure how big the entire lot is, but if it's enormous, maybe it
> would be worth setting up rsync. I wouldn't bother if the repository is
> under 10mb though.
> 
> With this easy change, we're still using CVS. Developers can set up
> their accounts to just push to the CVS subdirectory, which will be very
> quick. But it's easy to have a personal copy of the entire contents.
> (off-continent backups, yeah!)

The way I use it, I have that. "cvs update -d" takes a few minutes, but so what?

John Doty              Noqsi Aerospace, Ltd.
http://www.noqsi.com/
jpd AT noqsi DOT com



- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019