From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff King Subject: Re: [PATCH v4 15/16] branch: use ref-filter printing APIs Date: Fri, 15 Apr 2016 17:09:47 -0400 Message-ID: <20160415210947.GA21678@sigill.intra.peff.net> References: <20160413220559.GC8712@sigill.intra.peff.net> <20160414200530.GA26513@sigill.intra.peff.net> <20160414203615.GA31504@sigill.intra.peff.net> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Cc: Junio C Hamano , Git To: Karthik Nayak X-From: git-owner@vger.kernel.org Fri Apr 15 23:09:56 2016 Return-path: Envelope-to: gcvg-git-2@plane.gmane.org Received: from vger.kernel.org ([209.132.180.67]) by plane.gmane.org with esmtp (Exim 4.69) (envelope-from ) id 1arB00-00033z-38 for gcvg-git-2@plane.gmane.org; Fri, 15 Apr 2016 23:09:56 +0200 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751096AbcDOVJw (ORCPT ); Fri, 15 Apr 2016 17:09:52 -0400 Received: from cloud.peff.net ([50.56.180.127]:50436 "HELO cloud.peff.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1750801AbcDOVJv (ORCPT ); Fri, 15 Apr 2016 17:09:51 -0400 Received: (qmail 10375 invoked by uid 102); 15 Apr 2016 21:09:50 -0000 Received: from Unknown (HELO peff.net) (10.0.1.2) by cloud.peff.net (qpsmtpd/0.84) with SMTP; Fri, 15 Apr 2016 17:09:50 -0400 Received: (qmail 14998 invoked by uid 107); 15 Apr 2016 21:09:55 -0000 Received: from sigill.intra.peff.net (HELO sigill.intra.peff.net) (10.0.0.7) by peff.net (qpsmtpd/0.84) with SMTP; Fri, 15 Apr 2016 17:09:55 -0400 Received: by sigill.intra.peff.net (sSMTP sendmail emulation); Fri, 15 Apr 2016 17:09:47 -0400 Content-Disposition: inline In-Reply-To: Sender: git-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: git@vger.kernel.org Archived-At: On Sat, Apr 16, 2016 at 01:57:23AM +0530, Karthik Nayak wrote: > I had a look at your patch and even tested it, seems solid, I like how you > integrated all these atoms together under refname_atom_parser_internal(). > I'm squashing this in, for my re-roll. Thanks. Great, thanks for picking it up. > > So actually, we _do_ accept "%(upstream:short,track)" with your > > patch, which is somewhat nonsensical. It just ignores "short" and > > takes whatever option came last. Which is reasonable, though > > flagging an error would also be reasonable (and I think is what > > existing git does). I don't think it matters much either way. > > > > I think it was decided a while ago that it'd be best to ignore all and > accept the last argument without barfing, I couldn't find the exact > link. But I'm open to both. But if you look at the %(align) atom, > even that accepts mutually exclusive arguments and accepts the last > argument without reporting an error. Makes sense, and I'm fine with how you have it (and my patch tried to retain that property). I just wasn't sure if it was intentional, as I did a bad job of paying attention to earlier rounds of the series. Thank you for keeping at it. -Peff