From mboxrd@z Thu Jan  1 00:00:00 1970
MIME-Version: 1.0
In-Reply-To: <o2ydd6fe68a1005081035n3256da29w44c2cf0fce911ba1@mail.gmail.com>
References: <4be4dd37.NDxcevbsQ+CfYFvN%yard-ape@telus.net>
	<104169d77ce3a909d635845ac86efc30@9netics.com>
	<o2ydd6fe68a1005081035n3256da29w44c2cf0fce911ba1@mail.gmail.com>
Date: Mon, 10 May 2010 08:57:55 +0100
Message-ID: <g2ndf49a7371005100057u3e4788d7m1bc8c7ed40561f85@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
Subject: Re: [9fans] Binary File split
Topicbox-Message-UUID: 1e8f70bc-ead6-11e9-9d60-3106f5b1d025

On 8 May 2010 18:35, Russ Cox <rsc@swtch.com> wrote:
> bs=1474560
> cat $file | for(i in `{seq 0 `{ls -l $file | awk '{print
> int($6)/'$bs'}'}}) { dd -bs $bs -count 1 -of $file.$i }

that looks very plausible, but it doesn't actually work,
as awk doesn't coelesce short reads (it gets short
reads from the pipe)

this works a little better:

bs=1474560
{for(i in `{seq 0 `{ls -l $file | awk '{print
int($6)/'$bs'}'}}) { dd -bs $bs -count 1 -of $file.$i -quiet 1 }} < $file

but there's still a little annoyance - if the file size is an exact
multiple of the block size, it will generate an unnecessary zero-length
file at the end.

i tried to fix it to get rid of this, but ran hard up against awk's
dismalness. to illustrate:

% seq 106908504 106908509 | awk '{print $1 % 4}'
0
0
0
0
0
0
%

oh dear.