Message-ID: <42067988.30802@panoramicfeedback.com>
Date: Sun, 06 Feb 2005 15:09:44 -0500
From: Aaron Bentley <abentley@panoramicfeedback.com>
User-Agent: Mozilla Thunderbird 0.9 (X11/20041124)
MIME-Version: 1.0
To: Stefan Monnier <monnier@iro.umontreal.ca>
Subject: Re: [Gnu-arch-users] Re: release goals for Bazaar 1.1
References: <1102095916.6942.43.camel@localhost>	<pan.2004.12.11.14.43.23.535008@progsoc.org>	<20041211154916.GA1832@ukato>
	<20041211183809.GC3388@wisq.net>	<873byasd73.fsf@tleepslib.sk.tsukuba.ac.jp>	<cpk7vo$gu8$1@sea.gmane.org>
	<41BDADF5.5070507@arbash-meinel.com>	<buo1xbwsxhi.fsf@mctpc71.ucom.lsi.nec.co.jp>	<20050205024050.GB26365@ukato>	<fc339e4a05020420383dab9f7d@mail.gmail.com>	<20050205070604.GB18959@ukato>	<877jllr4jl.fsf-monnier+gmane.comp.version-control.arch.user@gnu.org>	<42065DB7.3090606@arbash-meinel.com>
	<87vf95pks4.fsf-monnier+gmane.comp.version-control.arch.user@gnu.org>
In-Reply-To: <87vf95pks4.fsf-monnier+gmane.comp.version-control.arch.user@gnu.org>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: gnu-arch-users@gnu.org
Precedence: list
Sender: gnu-arch-users-bounces+migo=homemail.com@gnu.org
Errors-To: gnu-arch-users-bounces+migo=homemail.com@gnu.org
Status: RO
Content-Length: 1536
Lines: 40
Stefan Monnier wrote:
>>You might seriously want to look into fai, since it seems to have some
>>of the features that you are asking for.
> 
> 
> Indeed, there are several third party packages offering functionality along
> the lines of what I'd ideally want.  But I'd like to see something like that
> merged into the main application.
> 
> Note also that the feature I described can be implemented without any need
> for side-info: all it needs is the {arch} patch-logs and the diffs from
> the archive.  No need to cache any kind of annotation.  You don't even need
> a revlib.  That's why I think it'd make sense to see this in Arch before
> "tla annotate".

Fai annotate just uses the {arch} patch-logs and diffs from the archive. 
  On local archives, it can be pretty quick.  On remote archives, it is 
usually damn slow.

The reason is that assembling this information requires retrieving all 
changesets for all the revisions that modified a particular file.  So 
the more changesets that modified the file, (and the higher the latency, 
and the lower the bandwidth) the slower the process.

The motivation for caching is to reduce the pain of repeatedly 
re-calculating this data every time you want to view the annotation of a 
given file.

I believe the same would be true of your "tla trace-code".

Aaron


_______________________________________________
Gnu-arch-users mailing list
Gnu-arch-users@gnu.org
http://lists.gnu.org/mailman/listinfo/gnu-arch-users

GNU arch home page:
http://savannah.gnu.org/projects/gnu-arch/