From mboxrd@z Thu Jan  1 00:00:00 1970
MIME-Version: 1.0
In-Reply-To: <6f3428794c36600244db89414a1ccd19@kw.quanstro.net>
References: <4be4dd37.NDxcevbsQ+CfYFvN%yard-ape@telus.net>
	<104169d77ce3a909d635845ac86efc30@9netics.com>
	<o2ydd6fe68a1005081035n3256da29w44c2cf0fce911ba1@mail.gmail.com>
	<g2ndf49a7371005100057u3e4788d7m1bc8c7ed40561f85@mail.gmail.com>
	<6f3428794c36600244db89414a1ccd19@kw.quanstro.net>
Date: Mon, 10 May 2010 13:17:05 +0100
Message-ID: <AANLkTimUrcZVSbSZLJuQiDLK9T5sk5j8QOdAQh53fgMz@mail.gmail.com>
From: roger peppe <rogpeppe@gmail.com>
To: Fans of the OS Plan 9 from Bell Labs <9fans@9fans.net>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Subject: Re: [9fans] Binary File split
Topicbox-Message-UUID: 1eb067d6-ead6-11e9-9d60-3106f5b1d025

On 10 May 2010 12:40, erik quanstrom <quanstro@quanstro.net> wrote:
> this isn't awk's fault. =C2=A0awk gets the right result. =C2=A0you've ill=
ustrated
> the dismalness of seq.

good point. i should have worked that out!

it's not the first time i've been caught out by %g - perhaps
the default precision of %g should be 12 or more. at any
rate, seq's default format should be %.12g or better.

here's a version that works a bit better (also using seq -w so
that you can do "cat $file.*" to reassemble it)

{for(i in `{seq -w 0 `{ls -l $file | awk '{print int((int($6) - 1) /
'$bs')}'}}) {
     dd -bs $bs -count 1 -quiet 1 -of $file.$i }
} < $file