topic Re: grep only one column in Operating System - HP-UX

grep only one column

Maurice Skubski — Mon, 18 Mar 2002 17:04:52 GMT

Hi, i want to grep only one column out of the bdf command. is there one command (bdf | grep ...) to do this. i think with awk it is possible, but i don't know anything about awk :( is there a good documentation? thanx for help.

Re: grep only one column

Ron Kinner — Mon, 18 Mar 2002 17:15:24 GMT

Not clear exactly what you want. If you just want the one column x you can do
bdf|cut -fx

Ron

Re: grep only one column

John Carr_2 — Mon, 18 Mar 2002 17:16:08 GMT

Hi

try this I think this is what you mean

bdf | awk '{ print $1 }' | grep xyz

where $1 represents the column number in bdf

cheers
John

Re: grep only one column

S.K. Chan — Mon, 18 Mar 2002 17:17:58 GMT

If you want to use awk, you can do this ..
something simple ..

# bdf|awk '{print $1}'
==> will print 1st column
# bdf|awk '{print $2}'
==> will print 2nd column .. and so on.

A good book on awk and sed is the O'Reilly book.

Re: grep only one column

Maurice Skubski — Mon, 18 Mar 2002 17:22:53 GMT

sorry, both things does not work. i need a command where i can get only the available memory output of a bdf command (column 4):

avail
89711
44471
230522
1847582

and only this part. any idea?

Re: grep only one column

S.K. Chan — Mon, 18 Mar 2002 17:24:25 GMT

Have you tried ..

# bdf|awk '{print $4}'

Re: grep only one column

Maurice Skubski — Mon, 18 Mar 2002 17:24:31 GMT

oh, i replaced the $1 :(
now it is working. thak you for help!!!!

Re: grep only one column

Darrell Allen — Mon, 18 Mar 2002 17:25:07 GMT

Hi,

You can use:
bdf | awk '/some_text/ {print $1}'

The result would be as if you only printed the first column from:
bdf | grep some_text

Darrell

Re: grep only one column

Bill Hassell — Mon, 18 Mar 2002 17:30:27 GMT

Simple answer:

bdf | awk '{print $4}'

No grep needed since you are looking just for the 4th field where a field is separated by whitespace. However, this will FAIL if the bdf is split scross 2 lines (as in a long NFS mountpoint). In that case, you'll need something a bit more sophisticated:

bdf | while read FS TOT USED AVAIL PERCENT MNT
do
if [ "$TOT" = "" ]
then
read TOT USED AVAIL PERCENT MNT
fi
echo $AVAIL
done

The advantage of the above code is that it will handle a 2-line bdf and also extracts all the rest of the columns, not just $AVAIL.

Re: grep only one column

John Carr_2 — Mon, 18 Mar 2002 17:51:56 GMT

Hi again

I bet now you wish to omit the column heading and you can do that by using grep -v

bdf | awk '{ print $4 }' | grep -v avail

Ron's answer was also correct and simpler.

bdf | cut -f4 | grep -v avail

cheers
John

Re: grep only one column

TrustNo1 — Tue, 19 Mar 2002 13:51:28 GMT

Bill,
I wasn't in on this thread but... I wanted to thank you for the valuable learning experience you provided in your "Sophisticated Code" example.
GREAT STUFF! :)
~jdk