From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=58219 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1PevWp-0003Zj-M9 for qemu-devel@nongnu.org; Mon, 17 Jan 2011 15:18:16 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1PevWo-0008Ls-AB for qemu-devel@nongnu.org; Mon, 17 Jan 2011 15:18:15 -0500 Received: from mail-px0-f173.google.com ([209.85.212.173]:38906) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1PevWo-0008Lm-3O for qemu-devel@nongnu.org; Mon, 17 Jan 2011 15:18:14 -0500 Received: by pxi16 with SMTP id 16so1268326pxi.4 for ; Mon, 17 Jan 2011 12:18:13 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <4D345A90.3030409@codemonkey.ws> References: <4D345A90.3030409@codemonkey.ws> From: Blue Swirl Date: Mon, 17 Jan 2011 20:17:52 +0000 Message-ID: Subject: Re: [Qemu-devel] RFC: mass comment conversion from C99 to C89 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Anthony Liguori Cc: Peter Maydell , qemu-devel On Mon, Jan 17, 2011 at 3:04 PM, Anthony Liguori wr= ote: > On 01/16/2011 06:12 PM, Peter Maydell wrote: >> >> On 15 January 2011 16:02, Blue Swirl =C2=A0wrote: >> >>> >>> With the sed script below (my first I think), I'm able to convert most >>> files in QEMU from C99 comment style to C89. When successive line with >>> C99 comments are converted, the comments are merged. Two files >>> (hw/rtl8139.c and microblaze-dis.c) still fail. >>> >>> Is this a good idea? >>> >> >> I'm a bit wary of this kind of wide-scale no-semantic-change commit >> (and more so for things like indent, brace and whitespace fixes >> which can affect large chunks of actual code), because it makes >> it harder to deal with qemu forks (especially if you were hoping to >> be able to periodically rebase with an eye to eventually getting >> changes back into mainline qemu). >> > > Yeah, I'm equally wary of such changes unless they have a very clear valu= e > (e.g. changing an API definition to accommodate an extra parameter). In this case the value is improved consistency. If we only mass converted target-sparc and cirrus (~1300 lines), the rest could be converted gradually.