Shell Programming and Scripting

Hello,

I would like to pipe two variables into awk, but I don't know how to do.

Each variable, "a" and "b", are in fact a list of data. They are not files.
So to get awk to work with it I am using:

Code:

 
echo $a | awk 'FNR==NR{print $1}FNR!=NR{print $4}'

The above works, but when I am trying with two inputs:

Code:

 
( echo $a; echo $b ) | awk 'FNR==NR{print $1}FNR!=NR{print $4}'

echo $a; echo $b are concatenated as a single input. Thus, my awk script will display $1 for both $a and $b.

Of course I could use:

Code:

 
echo $a > filea ; echo $b > fileb
awk 'FNR==NR{print $1}FNR!=NR{print $4}' filea fileb
\rm file[ab]

But that is very unconvenient.

Any suggestion?

Thanks

Edit: alternately, I am looking to solve the same problem, but using one variable and one file:

Code:

echo $a | awk 'FNR==NR{print $1}FNR!=NR{print $4}' fileb

but of course the above does not work!

How about

Code:

echo $a $b

I already tried it and it does not give the expected result!

use the -v option of awk to pass in variables

Unfortunately I cannot. If I do that, I don't read any file with awk, so it won't run.

I know I could use:
echo "blah" | awk -v a=$a -v b=$b '{...}'

but in that case it does not do at all what I want.

For now my only solution is to "echo" my variable content in a temporary file and read from the file.

Thanks anyway.

Place the code in BEGIN section, here's an example how to place the variables in arrays with the split function, supposing the lines are separated by a newline:

Code:

awk -v v1="$a -v v2="$b" 'BEGIN{
  n1=split(v1,arr1,"\n")    # n1 is the number of elements of the array
  n2=split(v2,arr2,"\n")
  print arr1[1]            # print the 1st element of arr1
}'

Regards