topic Re: Awk scripting in Operating System - HP-UX

Awk scripting

Oktay Tasdemir — Tue, 24 Jun 2003 23:03:35 GMT

Hi,

I need to extract the first occurrence of the field ABNA which is located on line 13 of the attached file and rename the file using this field. i.e test.ABNA.file

Could you please assist.

Many thanks
Oktay

Re: Awk scripting

curt larson_1 — Tue, 24 Jun 2003 23:38:03 GMT

cat test |
awk '/ABNA/ {print $0;exit;}' > test.ABNA.file

Re: Awk scripting

Oktay Tasdemir — Wed, 25 Jun 2003 00:05:30 GMT

Thanks for the reply,
The field where ABNA is located is variable and it denotes a company name which changes all the time.

Sorry I should have been more explicit.

Re: Awk scripting

curt larson_1 — Wed, 25 Jun 2003 00:29:21 GMT

yes some additional information would be helpful

Re: Awk scripting

Fragon — Wed, 25 Jun 2003 00:38:55 GMT

Hi Oktay,
From the sample file you posted,I found the the first occurrence of the filed "ABNA" is always exists in the line begins with "BHP".
So, this may help:
fnname=`more test|sed -n '/BHP/p'|awk '{print $8}'`
mv test test.${fnname}.file

Hope it can help you!

-ux

Re: Awk scripting

Fragon — Wed, 25 Jun 2003 01:06:23 GMT

If you prefer that line 13, the script should be:
fnname=`more test|sed -n '13p`|awk '{print $1}'`

Good luck!
-ux

Re: Awk scripting

Sridhar Bhaskarla — Wed, 25 Jun 2003 02:33:07 GMT

Hi Oktay,

Line 13th is fixed. Is the field 8th in 13th line fixed too?.

In that case you can simply use these two lines.

FIELD=$(sed -n '13p' your_file|awk '{print $8}')
cp your_file test.${FIELD}.file

-Sri

Re: Awk scripting

Oktay Tasdemir — Wed, 25 Jun 2003 04:09:13 GMT

Thanks for your replies,

Line 13 is fixed however the 8th variable is not and depending on the file it changes?

Re: Awk scripting

Francisco J. Soler — Wed, 25 Jun 2003 06:16:44 GMT

Hi Oktay,
If the 8th field in the 13th line is not fixed, how can we isolated the company name?

You must supply some condition to detect this name in order to extract it.

Frank.

Re: Awk scripting

Ionut Grigorescu_2 — Wed, 25 Jun 2003 08:31:01 GMT

Hi Oktay,

try this script:

#!/bin/sh
export file_name=`awk '{if (NR==13) {for (i=1;i<=NF;i++) {if($i != ""){print $i; next}}}}' $1`
mv $1 "$file_name"

than you can use it like
#<script_name>

it will rename your file to ABNA
awk will search record number 13 for the first non-empty field(in this case ABNA), then will print this field and exit the for loop.
you can change the last line to
mv $1 test. "$file_name".txt if you like.
Not so short but it works.

Have fun, ionut

Re: Awk scripting

Michael Kelly_5 — Wed, 25 Jun 2003 09:19:16 GMT

Oktay,
in the file you attached there is no ABNA string in line 13 but it does appear twice in line 11.
I'm assuming that the string you want to extract is the last non-blank field in line 11.
If this is the case, you can do this:

#!/usr/bin/sh

if [ ! -e $1 ]; then
echo "File $1 does not exist. Exiting"
exit 1
fi

newname=`awk 'NR == 11 { sub("[\t ]+\015$", ""); print $NF}' $1`
[ ! -z $newname ] && mv $1 test.${newname}.file

The awk bit strips the trailing whitespace and CR from line 11 and then prints the last field.
If you want the second last field, change $NF to $(NF - 1)

HTH,
Michael.