Get second part of line with SED 1

[blarneyme]

cat /etc/*release
Red Hat Enterprise Linux Server release 5.8 (Tikanga)

cat /etc/*release 
Red Hat Enterprise Linux ES release 4 (Nahant Update 9)

I want to extract just 5.8 or 4

I have used

cat /etc/*release | sed 's/\(.*\)\([0-9]\{1\}.?[0-9]\?\)\((.*)\)/\2/'

but it returns the entire line.

cat /etc/*release | sed 's/\(.*\)\([0-9]\{1\}.\?[0-9]\?\)\(.*\)/\2/'

returns just 8 from 5.8 and 9 from the RHEL 4 line.

 

[blarneyme]

cat /etc/*release | sed 's/\(.*\)\([0-9]\{1\}.\?[0-9]\{1,\}\)\(.*\)/\2/'

Correctly returns 5.8 but still returns the entire line for RHEL 4

 

[PHV]

What about this ?

sed 's/.*\([0-9]\{1\}\.\?[0-9]*\).*/\1/' /etc/*release

Hope This Helps, PH.
FAQ219-2884
FAQ181-2886

 

[blarneyme]

# cat /etc/*release
Red Hat Enterprise Linux Server release 5.8 (Tikanga)
# sed 's/.*\([0-9]\{1\}\.\?[0-9]*\).*/\1/' /etc/*release
8
#

$ cat /etc/*release
Red Hat Enterprise Linux ES release 4 (Nahant Update 9)
$ sed 's/.*\([0-9]\{1\}\.\?[0-9]*\).*/\1/' /etc/*release
9
$

 

[PHV]

And this ?

sed 's/.* \([0-9]\{1\}\.\?[0-9]*\) .*/\1/' /etc/*release

Hope This Helps, PH.
FAQ219-2884
FAQ181-2886

 

[blarneyme]

That did it! Thank you!!!

 

[FlorianAwk]

First, the suggested solutions won't work with:
release 10.1
Or
release release 4 (Nahant Update 9 extended)

What about

sed 's/.*release \([^ ]*\) .*/\1/' /etc/*release

?

 

[AnotherAlan]

An AWK solution....dont bite, i'm just bored

awk '{for(i=1;i<NF;i++)if($i=="release") print $(i+1)}' /etc/*release