From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnaldo Carvalho de Melo Subject: Re: diff vs codiff against empty Date: Thu, 31 Jan 2008 10:08:01 -0200 Message-ID: <20080131120801.GI1819@ghostprotocols.net> References: Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: Sender: dwarves-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Ilpo =?iso-8859-1?Q?J=E4rvinen?= Cc: dwarves-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: dwarves@vger.kernel.org Em Wed, Jan 30, 2008 at 09:42:18PM +0200, Ilpo J=E4rvinen escreveu: > Hi, >=20 > Not a bad problem but makes some tricks I do harder than they would n= eed=20 > to be: >=20 > ijjarvin@kivilampi-30:~/work/src/submit$ diff /dev/null lib/jhash.o > Binary files /dev/null and lib/jhash.o differ > ijjarvin@kivilampi-30:~/work/src/submit$ codiff /dev/null lib/jhash.o > codiff: /dev/null: not a valid ELF file [acme@doppio examples]$ codiff /dev/null hello hello.c: main | +47 1 function changed, 47 bytes added, diff: +47 [acme@doppio examples]$ codiff hello /dev/null hello.c: main | -47 1 function changed, 47 bytes removed, diff: -47 [acme@doppio examples]$ =20 > I.e., I created new file jhash and just want to diff it against empty= =2E I=20 > also tried codiff against echo "" | gcc -g -c -o lib/jhash.o.base -x = c -=20 > but that wasn't any better. I can fix that with sed trick by using pf= unct=20 > -s for it first though. >=20 > Btw, codiffs help/usage printout is incorrect or uses unusual notatio= n, it=20 > uses [] brackets around old_file and new_file even though they are no= t=20 > optional arguments. =46ixed in all the dwarves, thanks. - Arnaldo