Search Results

None of the solutions so far covers your other condition "if the 3rd column is empty it shouldn't delete that line". Adapt your own solution like

awk -F "^" '$3=="|" || $4!="|"' fileNote that...

Hello Tuxidow,

Could you please try following.


awk -F'\\|\\^' '$3==""{next} $4' Input_file


NOTE: After seeing Rudi sir's comment edited this solution now.

Thanks,
R. Singh

One way to do it would be to keep appending and "building" two different strings - one for all users and the other for all user role assignments inside the while loop.
Once you're done with the...

awk '{if (!b[$1,$2]++) {if (a) print a; a=$0} else {a=a "," $NF}} END { print a } ' infile

A bit altered coding:
awk '{key=$1FS$2; a[key]=((key in a) ? a[key]","$3 : $3)} END {for (key in a) {print key, a[key]}}' testfile

try below
awk '{if (a[$1FS$2FS]) {a[$1FS$2FS] =a[$1FS$2FS]","$3} else {a[$1FS$2FS]=$3}} END {for (key in a) { print key a[key]}}' testfile

This is not Perl. This is 99% shell code in a transparently thin wrapping of perl. You could dispense with the Perl and hardly see a difference:

string="test.binary.value"...

Instead of
my $returnCode = system ( "ssh $host 'cp -p /tmp/testfile /tmp/testfile.bkup; sed -i 's/\$string.*/\$toggle/g' /tmp/testfile'" );
try
my $returnCode = qx( ssh $host "cp -p /tmp/testfile...

Heya

You are reading fname, and compare name later.
Also highly recomend, use quotes around variables when comparing (if, case, ...)
Otherwise an empty variable results in a failing script,...

sed -n '/10\.28\.[78]6\.15/{
s/\("[^"]*\)\{3\}"/ /
s/,.*//
p
}' file

like this.....?


awk -F '[" ]' '{for (i=1;i<=NF;i++){ if ((! s) && $i ~ /[0-9][0-9].[0-9][0-9].[0-9][0-9][0-9].[0-9][0-9][0-9]/) {print $1,$2,$i}}}' file

Or:
# awk '/10.28.76.15|10.28.86.15/{gsub(/\"|,/,"");print $1,$2,$(NF-1)}' infile

Must it be sed?


$ awk -F "[ ,\"]" '/(10.28.76.15|10.28.86.15)/ {print $1,$2,$(NF-3)}' infile
10.28.76.15 SSLC=Z42 71.19.175.130
10.28.76.15 SSLC=Z42 71.19.175.130
10.28.76.15 SSLC=Z42...

Hi

$ sed 's/\([^ ]*\).*\(HPSAC=[^\&]*\).*/\1 \2/' file
66.150.161.195 HPSAC=Z05
66.150.161.196 HPSAC=A05

Guru.

The error is pretty descriptive. You have 2 columns too many in those additional /proc mount entries. Try: #device device mount FS fsck mount mount
#to mount ...

tar -C /library/utility/apache-tomcat/tbase-6001/repositories -cvf - . | ( mkdir -p /library/utility/apache-tomcat/tbase2008-6001/repositories ; cd...

In some other country whose timezone differs from GMT, you should take into account the offset a reliable calculation could be something like


# -- calculate time using GMT --
Uday=`TZ=GMT date...

unzip -l AMT1-NA-20101007-1.zip | awk '/[0-9]:[0-9]/ {print $4,$1}'

You can try this: nohup tail -f file | awk '/logout/{system("echo message | write root")}' &It will run in the background until someone kills it or reboots the server (it will wun after you log out...

Something like this (untested, run with nohup in background):

tail -f <your_file> |
while IFS= read -r; do
case $REPLY in
( *your pattern* ) run_your_commands ;;
esac
done

#!/usr/bin/awk -f
BEGIN {
FS = "[\" :]+";
}

{
sub(FS "$", "");
DateTime = $1 ":" $2 ":" $3 ":" $4;
Action = $5;
ActionLower = tolower($5);

if (ActionLower ==...