public inbox for git@vger.kernel.org 
 help / color / mirror / Atom feed
From: Michael J Gruber <git@drmicha•warpmail.net>
To: "Uwe Kleine-König" <u.kleine-koenig@pengutronix•de>
Cc: Steven Rostedt <rostedt@goodmis•org>,
	Darren Hart <darren@dvhart•com>,
	git@vger•kernel.org
Subject: Re: [PATCH] Documentation/git-clone: describe --mirror more verbose
Date: Mon, 04 Oct 2010 11:28:35 +0200	[thread overview]
Message-ID: <4CA99E43.1000204@drmicha.warpmail.net> (raw)
In-Reply-To: <20101004085050.GQ28679@pengutronix.de>

Uwe Kleine-König venit, vidit, dixit 04.10.2010 10:50:
> Hello Michael,
> 
> On Mon, Oct 04, 2010 at 10:23:13AM +0200, Michael J Gruber wrote:
>> Uwe Kleine-König venit, vidit, dixit 04.10.2010 09:50:
>>> Hello Michael,
>>>
>>> On Mon, Oct 04, 2010 at 09:25:17AM +0200, Michael J Gruber wrote:
>>>> Steven Rostedt venit, vidit, dixit 01.10.2010 23:16:
>>>>> On Fri, 2010-10-01 at 13:18 -0700, Darren Hart wrote:
>>>>>> 2010/10/1 Uwe Kleine-König <u.kleine-koenig@pengutronix•de>:
>>>>>>> Some people in #linux-rt claimed that you cannot define "--mirror" with
>>>>>>> "mirror".
>>>>>>>
>>>>
>>>> I'd say "mirror" is a commonly known term for an exact copy. Moreover,
>>>> the text below doesn't explain what a mirror is either, only how
>>>> "update" behaves in it.
>>> hmm.  The --mirror option doesn't have any effect (apart from the
>>> changes in the config file) until you update.  So I think it's natural
>>> to talk about git update.  No?
>>
>> "git clone" (with or without --mirror) does a couple of things, and it
>> does them differently when "--mirror" is used. It mirrors each branch
>> from the source repo in the target repo under the same name, including
>> for example any remote branches in the source repo. This is completely
>> different without "--mirror", where clone does not look at the source's
>> remote branches at all. Also, it sets up a mirroring refspec, i.e.
>> +refs/*:refs/*
> Ah, didn't notice that.
> 
>> [...]
>> This makes me think that --mirror should be explained on top of --bare.
> OK.
> 
>> For example:
>>
>> In addition to the mapping of local branches to local branches which
>> --bare does, --mirror maps all refs which the source has under the same
>> name in the target (including remote branches, notes etc.) and sets up a
>> refspec configuration so that all these refs are updated by a `git
>> update` in the target repo.
> Hmm, I didn't understand this when I read it the first few times.  The
> special thing is that --mirror maps *all* refs, not *same name*.

Yep, my sentence was bad, it could be misunderstood.

> 
> So maybe:
> 
> 	Set up a mirror of the remote repository.  This implies `--bare`.
> 	Compared to `--bare`, `--mirror` doesn't only map local branches of

"does not"

> 	the remote to local branches of the target but all refs
> 	(including remote branches, notes etc.) and sets up a refspec

Maybe "of the source" for "of the remote"? Because remote comes up right
after in a different meaning.

> 	configuration such that all these refs are overwritten by a
> 	`git remote update` in the target repository.
> 

I like that a lot!

> I choosed to write "overwritten" instead of "updated" to make it clearer
> that it makes no sence to push into these branches from a different
> source.  Should this be noted more explicit?

"update" may be more Git lingo but I think either is fine. "overwritten"
makes it clearer this is not a repo to do development in...

Cheers,
Michael

  reply	other threads:[~2010-10-04  9:28 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-10-01 20:13 [PATCH] Documentation/git-clone: describe --mirror more verbose Uwe Kleine-König
2010-10-01 20:18 ` Darren Hart
2010-10-01 21:16   ` Steven Rostedt
2010-10-04  7:25     ` Michael J Gruber
2010-10-04  7:50       ` Uwe Kleine-König
2010-10-04  8:23         ` Michael J Gruber
2010-10-04  8:50           ` Uwe Kleine-König
2010-10-04  9:28             ` Michael J Gruber [this message]
2010-10-04  9:51               ` [PATCH v2] " Uwe Kleine-König
2010-10-04 12:47                 ` Steven Rostedt
2010-10-04 12:52                 ` Michael J Gruber
2010-10-04 14:42                 ` Darren Hart
2010-10-04 17:28                   ` [PATCH v3] Documentation/git-clone: describe --mirror more verbosely Uwe Kleine-König
2010-10-04 17:29               ` [PATCH] Documentation/git-clone: describe --mirror more verbose Junio C Hamano
2010-10-04 17:39                 ` Matthieu Moy
2010-10-04 19:05                   ` Darren Hart
2010-10-05  7:01                 ` Uwe Kleine-König
2010-10-05 15:41                   ` Junio C Hamano

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4CA99E43.1000204@drmicha.warpmail.net \
    --to=git@drmicha$(echo .)warpmail.net \
    --cc=darren@dvhart$(echo .)com \
    --cc=git@vger$(echo .)kernel.org \
    --cc=rostedt@goodmis$(echo .)org \
    --cc=u.kleine-koenig@pengutronix$(echo .)de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox