From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 28558 invoked by alias); 18 Feb 2007 22:32:15 -0000 Received: (qmail 28550 invoked by uid 22791); 18 Feb 2007 22:32:15 -0000 X-Spam-Check-By: sourceware.org Received: from romy.inter.net.il (HELO romy.inter.net.il) (213.8.233.24) by sourceware.org (qpsmtpd/0.31) with ESMTP; Sun, 18 Feb 2007 22:32:09 +0000 Received: from HOME-C4E4A596F7 (IGLD-80-230-15-157.inter.net.il [80.230.15.157]) by romy.inter.net.il (MOS 3.7.3-GA) with ESMTP id HDN79928 (AUTH halo1); Mon, 19 Feb 2007 00:30:13 +0200 (IST) Date: Sun, 18 Feb 2007 22:32:00 -0000 Message-Id: From: Eli Zaretskii To: Daniel Jacobowitz CC: jan.kratochvil@redhat.com, gdb-patches@sourceware.org In-reply-to: <20070218203410.GA31724@caradoc.them.org> (message from Daniel Jacobowitz on Sun, 18 Feb 2007 15:34:10 -0500) Subject: Re: [RFC] print/x of a float/double should printf ("%a") Reply-to: Eli Zaretskii References: <20070218153705.GA25940@host0.dyn.jankratochvil.net> <20070218203410.GA31724@caradoc.them.org> X-IsSubscribed: yes Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2007-02/txt/msg00229.txt.bz2 > Date: Sun, 18 Feb 2007 15:34:10 -0500 > From: Daniel Jacobowitz > Cc: Jan Kratochvil , gdb-patches@sourceware.org > > On Sun, Feb 18, 2007 at 10:25:54PM +0200, Eli Zaretskii wrote: > > . I don't like this bit: > > > > > Systems not supporting "%a" will print the value as > > > print (float) a > > > which differs from the current way there: > > > print/x (long) a > > > > This makes this an incompatible change. For this reason, I'd prefer a > > new format letter, like Andreas suggested. > > Is that the only reason you prefer a new format letter, rather than > changing 'x'? We could make it work everywhere - it would just be a > bit of work, probably by adding appropriate bits to gnulib. I'll > even volunteer to do it. That'd be fine with me, but do we really want a deviation from past behavior? How do we know no one out there depends on it? > The current behavior is "print/x (long) val", which I consider to be > silly. I don't think it's so silly, but I cannot say I care for it too much, either. My main concern is to avoid gratuitous breakage of someone's .gdbinit. > I can't recall ever typing p/x on a float value and expecting > to get a result rounded to no decimal places. Me neither.