Script question (Remove certain lines from a text file)

[boym1]

Hi,

Can anyone tell me how in a script to make the original input below look like the desired output also shown below. Want to get rid of the lines that only have one [number] and keep the lines that have two lines with the same [number]? Thanks.....

Original :
05:20:13 [1159] OracleH3
05:22:16 [1159] 178741 Kbytes at 1619.574 Kbytes/sec
14:49:35 [11630] 497376 Kbytes at 4048.645 Kbytes/sec
04:08:15 [11635] OracleH3
04:11:17 [11635] 468679 Kbytes at 2718.051 Kbytes/sec
05:20:13 [1160] SDP1_exp
05:22:16 [1160] 1782741 Kbytes at 1619.574 Kbytes/sec
14:49:35 [19830] 497376 Kbytes at 4048.645 Kbytes/sec
04:08:15 [11636] PRD3
04:11:17 [11636] 468679 Kbytes at 2718.051 Kbytes/sec


Desired Output: Lines with [11630] and [19830] removed - numbers will change.

05:20:13 [1159] OracleH3
05:22:16 [1159] 178741 Kbytes at 1619.574 Kbytes/sec
04:08:15 [11635] OracleH3
04:11:17 [11635] 468679 Kbytes at 2718.051 Kbytes/sec
05:20:13 [1160] SDP1_exp
05:22:16 [1160] 1782741 Kbytes at 1619.574 Kbytes/sec
04:08:15 [11636] PRD3
04:11:17 [11636] 468679 Kbytes at 2718.051 Kbytes/sec

 

[RobJordan]

egrep -v "\[11635\]|\[19830\]" some_file

OR

egrep -v "11635|19830" some_file

Robert G. Jordan
Unix Sys Admin
Sleepy Hollow, Illinois U.S.A.

http://www.InYourMind.com

FREE Unix Scripts

 

[RobJordan]

Ahh.. first i need to come up with something
to find out which lines have duplicate numbers... Robert G. Jordan
Unix Sys Admin
Sleepy Hollow, Illinois U.S.A.

http://www.InYourMind.com

FREE Unix Scripts

 

[RobJordan]

#!/usr/bin/ksh

SOURCE_FILE=data1

cat $SOURCE_FILE|while read LINE
do
CODE=`echo $LINE|awk -F"[" '{print $2}'|awk -F"]" '{print $1}'`
LINE_COUNT=`grep $CODE $SOURCE_FILE|wc -l`
if [ $LINE_COUNT -gt 1 ]
then
echo $LINE
fi
done Robert G. Jordan
Unix Sys Admin
Sleepy Hollow, Illinois U.S.A.

http://www.InYourMind.com

FREE Unix Scripts

 

[boym1]

Maybe my description was not clear but for this one the forms of grep will not work as the numbers in the []'s change.

 

[boym1]

Sorry sent that before your second posting arrived Thank You I'll give it a try

 

[RobJordan]

I don't think that should matter.
Here's an example of what the awk statements will do
even if there are more than one set of [] per line

LINE="05:22:16 [1159] 178741 [Kbytes] at 1619.574 Kbytes/sec"

echo $LINE|awk "]" '{print $2}'
will print to the right of the first [ bracket
giving the result: 1159] 178741 [Kbytes] at 1619.574

LINE="1159] 178741 [Kbytes] at 1619.574"

echo $LINE|awk -F"]" '{print $1}'
will print to the left of the first ] braket
giving the result: 1159

If you are talking about another scenario,
please give an example.







Robert G. Jordan
Unix Sys Admin
Sleepy Hollow, Illinois U.S.A.

http://www.InYourMind.com

FREE Unix Scripts

 

[boym1]

Robert,

THANK YOU!!!!!!!!!!!!!!!!!!!!!!!!! I am trying to learn awk and sed more but it is going slow.. Thanks so very much for your help.....

Netbackup bptm log.050102 for Log file log.050102
**********************************************************
09:34:01 [14692] vincetst
09:44:52 [14692] 1286683 Kbytes at 2603.615 Kbytes/sec
14:06:29 [19989] xerox
14:32:00 [19989] 810348 Kbytes at 598.643 Kbytes/sec
ush4: