Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1NmNXr-0004To-AP for qemu-devel@nongnu.org; Tue, 02 Mar 2010 03:33:35 -0500 Received: from [199.232.76.173] (port=42697 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1NmNXq-0004TU-Pf for qemu-devel@nongnu.org; Tue, 02 Mar 2010 03:33:34 -0500 Received: from Debian-exim by monty-python.gnu.org with spam-scanned (Exim 4.60) (envelope-from ) id 1NmNXn-0002Pp-Ug for qemu-devel@nongnu.org; Tue, 02 Mar 2010 03:33:34 -0500 Received: from mx1.redhat.com ([209.132.183.28]:51612) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1NmNXn-0002P9-Fp for qemu-devel@nongnu.org; Tue, 02 Mar 2010 03:33:31 -0500 Received: from int-mx04.intmail.prod.int.phx2.redhat.com (int-mx04.intmail.prod.int.phx2.redhat.com [10.5.11.17]) by mx1.redhat.com (8.13.8/8.13.8) with ESMTP id o228XTJl004489 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Tue, 2 Mar 2010 03:33:30 -0500 From: Markus Armbruster Subject: Re: [Qemu-devel] [PATCH RFC 12/48] error: New error_printf() and error_vprintf() References: <1267034160-3517-1-git-send-email-armbru@redhat.com> <1267034160-3517-13-git-send-email-armbru@redhat.com> <20100226164354.6be8466d@redhat.com> <20100301130941.34e28a8c@redhat.com> Date: Tue, 02 Mar 2010 09:33:27 +0100 In-Reply-To: <20100301130941.34e28a8c@redhat.com> (Luiz Capitulino's message of "Mon, 1 Mar 2010 13:09:41 -0300") Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Luiz Capitulino Cc: qemu-devel@nongnu.org Luiz Capitulino writes: > On Mon, 01 Mar 2010 09:54:32 +0100 > Markus Armbruster wrote: > >> Luiz Capitulino writes: >> >> > On Wed, 24 Feb 2010 18:55:24 +0100 >> > Markus Armbruster wrote: >> > >> >> FIXME They should return int, so callers can calculate width. >> >> >> >> Signed-off-by: Markus Armbruster >> >> --- >> >> qemu-error.c | 49 ++++++++++++++++++++++++++++++++++++++++++------- >> >> qemu-error.h | 14 ++++++++++++++ >> >> 2 files changed, 56 insertions(+), 7 deletions(-) >> >> >> >> diff --git a/qemu-error.c b/qemu-error.c >> >> index 63bcdcf..76c660a 100644 >> >> --- a/qemu-error.c >> >> +++ b/qemu-error.c >> >> @@ -1,18 +1,53 @@ [...] >> >> +/* >> >> + * Print to current monitor if we have one, else to stderr. >> >> + * FIXME should return int, so callers can calculate width, but that >> >> + * requires surgery to monitor_printf(). Left for another day. >> >> + */ >> >> +void error_vprintf(const char *fmt, va_list ap) >> >> { >> >> - va_list args; >> >> - >> >> - va_start(args, fmt); >> >> if (cur_mon) { >> >> - monitor_vprintf(cur_mon, fmt, args); >> >> + monitor_vprintf(cur_mon, fmt, ap); >> >> } else { >> >> - vfprintf(stderr, fmt, args); >> >> + vfprintf(stderr, fmt, ap); >> >> } >> >> - va_end(args); >> >> +} >> > >> > This can be static. >> >> Yes. But why would that be useful? It's neither a name space pollution >> nor does it poke a hole into an abstraction. > > Well, IMHO unused public symbols serve only one purpose: to pollute the > global namespace :) > > So, I think the question is: if it doesn't have any user and if you > don't expect it to be used anytime soon: why make it public? It's a basic building block. Uses will come. >> >> + >> >> +/* >> >> + * Print to current monitor if we have one, else to stderr. >> >> + * FIXME just like error_vprintf() >> >> + */ >> >> +void error_printf(const char *fmt, ...) >> >> +{ >> >> + va_list ap; >> >> + >> >> + va_start(ap, fmt); >> >> + error_vprintf(fmt, ap); >> >> + va_end(ap); >> >> +} >> > >> > This function's name is inconsistent with qemu_error() and >> > qemu_error_new(). >> >> I'm fond of prepending qemu_ to random symbols left and right. Yes, I >> know I'm reading QEMU source code, thank you :) >> >> If the names here are really important: What about stripping qemu_ from >> qemu_error() & friends? > > I'm ok with that (and Paolo gave some suggestions), but I hope you > submit a patch soon. It's ok to criticize/improve bad consistency policies, > but it's not ok to break them. Paolo's error_raise() works for me, although I like error_report() a bit better. But we need two names, one for simple, direct error reporting (now qemu_error()), and one for QMP-compatible error reporting (now qemu_error_new()). Call them error_report() and qerror_report()? Or is that too similar?