AWK command

Hi,
I want to write a simple command at awk.
i have a file:

a=1
b=2
c=3

i want to print c=3.

i worte the following command:

awk '/c/ {print $1,$2,$3}' filename.(it works great)

i want to write the same command,but with an argumant:

x="c"
awk '/$x/ {print $1,$2,$3}' filename.(it's not working!!!).

Please help.

Thanks,

You need to tell awk that you want to pass a variable from the shell to awk, this can be done with awk's -v var=value option.

So your example will look like this:

X="c"
awk -v XYZ=$X '/c/ {print $1}' filename

You also use the $1, $2 and $3 incorrectly. There are no (standard) seperators between a=1, b=2 and c=3. This means that a=1 is $1 (same for b=2 and c=3).

Hope this helps.

I did not understand...

your example:

X="c"
awk -v XYZ=$X '/c/ {print $1}' filename

i want to use variable instead of value,i mean to replace the "c"

'/c/' to variable,so i can get change it according to the input i get,
"c" or "b" ot "a".

Thanks,

Sorry, my mistake!

But even with the intention I had it would not work. It seems that you can't (or better: I don't know how to) use a variable inside the /<pattern>/ construction in awk.

I do have another sollution, which works around the problem. Bit more complicated:

X="c"
awk -F= -v XYZ=$X '$1 ~ XYZ {print $0}' filename

First the fieldseperator is set to = (-F=). You will end up with not 1 field (a=1) but with 2 ( a and 1).

Next, pass variable to awk (-v XYZ=$X)

$1 ~ XYZ This construction does the following:

Compare a certain input field (in this case $1, which is c) and compare it (~) to a value/variable (in this case XYZ, which contains the value of $X, which contains c)

If the comparison matches awk prints the entire input record {print $0}, if it doesn't it checks the next record.

Hope this is not too complicated, if you have any questions just ask them.

Thanks,

It was very easy to understand,and it works.

10X again.

One last question....

I want to get an input from the user and to search the value by the command that you wrote'if there is such a value in the file,to print "OK".

i don't know what i'm doing wrong...

#!/bin/ksh
print -n "enter value you want to search"
read X
y=`awk -F= -v XYZ=$X '$1 ~ XYZ {print $1,$2}' myfile `
if [ $y = $X ]
then
print "OK"
fi

i get the following error message:
./test[10]: [: 3: unexpected operator/operand

Thanks,

This:
if [ $y = $X ]
should be:
if [[ $y = $X ]]

There is another thing that will not work:

Your awk statement prints one field ($2) to many. Now y is filled with c 3, there will always be a number on the end. If, like I understood from your previous posts, you only want to enter a,b,c etc, you have to remove the ,$2 part.

Here's your code, a bit rewritten and working.


One more piece of advise: Make you variables human readable. Use INPUT, or USER_INPUT instead of X and User_Input instead of XYZ.

In small scripts it might not be that important, but if you start writing longer scripts/programs it's very pleasent to know what a variable is used for. It will also help others (or yourself after a long time) to figure out what the script is doing. Maybe not that important if you only script at home, but it is if others are going to see/use your scripts.

Sorry again,
It's not working...

i enter "c" and in my file i have "c" and message "not ok"

the code:

#!/bin/ksh

print -n "enter value you want to search "

read X

y=`awk -F= -v XYZ=$X '$1 ~ XYZ {print $1}' myfile`
print $y
print $X

if [[ $y = $X ]]
then
print "OK"
else
print "Not OK"
fi

the message:

enter value you want to search c
c
c
Not OK

Can you please check what is the problem

If I copy/paste your example it works:

enter value you want to search c
c
c
OK

Only thing I can think of: Is your input file still the same (compared to the one in your first post). I'm using:

$ cat myfile
a=1
b=2
c=3

This script worked for me.

printf "Please enter the search string : "
read input_value
searched_value=`awk -F= ' $1 == X { print $1 } ' X="$input_value" logs1`
if [ "$input_value" = "$searched_value" ]
then
print "OK"
fi