--On July 12, 2010 10:29:08 PM +0800 Aiza <aiz...@comclark.com> wrote:
> Sorry miss send, was not done yet.
>
> Have a .sh script that accepts an -s sparse file size.
> Only 2 suffix's are valid m and g.
>
> Been trying to get this line of code to strip out just the single letter. But
> it strips the letter and every thing to the right of it.
>
> Timagesize=`echo-n "${imagesize}" | sed 's/g.*$//'`
>
> I plan to strip just the m or g if its there and the result should be
> numeric. If not numeric know invalid suffix.
>
> Need help with the sed syntax. Or if there is better way I want to learn it.
>
> Thanks
Is this what you want?
sed -n 's/^\([0-9]\{1,\}[gm]\)$/\1/p'
Prints output only if the input begins with digits and ends with g or m.
Or this?
sed -n 's/^\([0-9]\{1,\}\)[gm]$/\1/p'
Prints numeric output only if the input begins with digits and ends with g
or m.
<http://www.grymoire.com/Unix/Sed.html>
<http://sed.sourceforge.net/sed1line.txt>
pgp0dDi9HpJYK.pgp
Description: PGP signature
