awk split parameter by char

Question

file is:

BASH.NIR
SH.ABI

I want the awk script to show:

User is NIR, SHELL is BASH
User is ABI, SHELL is SH

I don't know how to split a parameter by char. The idea is:

cat file.txt | awk '{print "User is " afterDot($1) ", SHELL is " beforeDot($1)}'

Answer 1

You can just use . as the field separator:

awk -F. '{ print "User is " $2 ", SHELL is " $1 }' file.txt

Answer 2

You can use the string functions in awk.

$ (echo BASH.NIR; echo SH.ABI FOOBAR) | awk '{p=index($1,".");print "User is", substr($1,p+1) ", SHELL IS", substr($1,0,p-1)}'
User is NIR, SHELL IS BASH
User is ABI, SHELL IS SH

The index function returns the position of the character to be found (in this case a dot) and substr will return a substring. We use p+1 and p-1 to not include the dot.

For more information look in the "String Functions" section of the awk manpage.

Answer 3

not as elegant as I hoped but it got the job done :

echo "BASH.NIR\nSH.ABI" \
 \
 | mawk 'sub( /$/,  ", SHELL is " ($!_=$!_))+\
         sub(/^[^ ]* /,"User is ")' FS='[.]'

User is NIR, SHELL is BASH
User is ABI, SHELL is SH

Only can confirm it works for gawk, mawk-1, and mawk-2. Can't get nawk onboard.