From mboxrd@z Thu Jan 1 00:00:00 1970 From: Johannes Schindelin Subject: Re: [BUG?] git-archive ignores remote .gitattributes (was: .git/info/attributes not cloned) Date: Fri, 28 Mar 2008 14:22:53 +0100 (CET) Message-ID: References: <47EB0FAE.5000102@rea-group.com> <20080327033341.GB5417@coredump.intra.peff.net> <47EB213F.1020503@rea-group.com> <20080327042925.GA6426@coredump.intra.peff.net> <47EB271F.1050307@rea-group.com> <20080327045342.GC6426@coredump.intra.peff.net> <47EC7DD1.3060102@rea-group.com> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Cc: Toby Corkindale , git@vger.kernel.org To: Jakub Narebski X-From: git-owner@vger.kernel.org Fri Mar 28 14:24:06 2008 Return-path: Envelope-to: gcvg-git-2@gmane.org Received: from vger.kernel.org ([209.132.176.167]) by lo.gmane.org with esmtp (Exim 4.50) id 1JfEYV-0006Cl-Ni for gcvg-git-2@gmane.org; Fri, 28 Mar 2008 14:23:40 +0100 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751212AbYC1NWz (ORCPT ); Fri, 28 Mar 2008 09:22:55 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751419AbYC1NWz (ORCPT ); Fri, 28 Mar 2008 09:22:55 -0400 Received: from mail.gmx.net ([213.165.64.20]:37969 "HELO mail.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1750969AbYC1NWz (ORCPT ); Fri, 28 Mar 2008 09:22:55 -0400 Received: (qmail invoked by alias); 28 Mar 2008 13:22:53 -0000 Received: from unknown (EHLO [138.251.11.74]) [138.251.11.74] by mail.gmx.net (mp029) with SMTP; 28 Mar 2008 14:22:53 +0100 X-Authenticated: #1490710 X-Provags-ID: V01U2FsdGVkX1+gacZysLigxffB/+xDa9mi+Rw49o2sut0jTGDHTX vkAhLI9ny0hHY7 X-X-Sender: gene099@racer.site In-Reply-To: User-Agent: Alpine 1.00 (LSU 882 2007-12-20) X-Y-GMX-Trusted: 0 Sender: git-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org Archived-At: Hi, On Fri, 28 Mar 2008, Jakub Narebski wrote: > Johannes Schindelin writes: > > > On Fri, 28 Mar 2008, Toby Corkindale wrote: > > > > > I submit that this is a bug, or at least undesirable behaviour: > > > > > > "git-archive --remote=/some/repo" will ignore > > > /some/repo/.gitattributes, but check /some/repo/info/attributes. > > > > > > I think the problem is in the loop that looks for .gitattributes, > > > which seems to do so by taking the current path and iterating down > > > through it? > > > > The problem is that "git archive --remote" operates on the remote > > repository as if it were bare. Which in many cases is true. > > > > So I'd submit that this is not the usage .gitattributes is meant for, > > and that you should clone the thing if you want to generate archives > > heeding the .gitattributes. > > This is simply caused by lacking implementation of .gitattributes (which > is quite new feature, so it is somewhat understandable). > > As I see it nothing prevents git to take and use .gitattributes from a > given tree (from a top tree of a given commit)... well, nothing except > the fact that git-check-attr, and probably also API used by attributes > code in builtins, doesn't have place to provide blob to be used as > .gitattributes (or tree to take .gitattributes from). Of course, you can ask for people to patch it. Or even provide a patch yourself. Personally, I do not think it is worth it. Ciao, Dscho