From: Petr Baudis <pasky@ucw.cz>
Date: Fri, 13 May 2005 05:26:33 GMT
Subject: Re: [PATCH] Stop git-rev-list at sha1 match
Message-ID: <20050513052633.GA16464@pasky.ji.cz>
In-Reply-To: <1115857873.22180.253.camel@tglx>

Dear diary, on Thu, May 12, 2005 at 02:31:13AM CEST, I got a letter
where Thomas Gleixner <tglx@linutronix.de> told me that...
> On Thu, 2005-05-12 at 01:44 +0200, Petr Baudis wrote:
> > for extensive discussion on how (it is impossible or very hard) to do
> > better.
> 
> :)
> 
> > So how would you order the list of commits?
> 
> Rn
>   merged Mn
>   merged Mn-1
> Rn-1
> ....
> 
> That's the relevant information in repository R. Looking at it from
> repository M after M updated to Rn
> 
> (Mn+1) == Rn	; Mn+1 is not created due to head forward
>   merged Rn
>   .. 
>   merged Rn-3
> Mn
> Mn-1
> 
> Thats the historical correct ordering from a repository point of view.
> Thats the only relevant information IMNSHO.

But it is impossible to reconstruct without the repoid or something. So
my point that it makes no sense and is actually dangerous with the
current rev-list output order holds.

-- 
				Petr "Pasky" Baudis
Stuff: http://pasky.or.cz/
C++: an octopus made by nailing extra legs onto a dog. -- Steve Taylor