awk to convert CSV into colums

UNIX for Beginners Q & A
1

My file (FILE1) looks like this:

HiringManager_RHMC, DirectSupervisor_RHMC, ProcessServerReadAccess_ST

I'd like to turn into:

HiringManager_RHMC
DirectSupervisor_RHMC
ProcessServerReadAccess_ST

I tried awk -F, '{print $1 $3}' FILE1
and result as this:

HiringManager_RHMC ProcessServerReadAccess_ST

Thank you in advance!

2

Hi,

If you're wanting to print the separate fields of a CSV out one per line, then something like this would work:

$ cat test.csv
HiringManager_RHMC,DirectSupervisor_RHMC,ProcessServerReadAccess_ST
$ awk -F, '{print $1"\n"$2"\n"$3}' < test.csv      
HiringManager_RHMC
DirectSupervisor_RHMC
ProcessServerReadAccess_ST
$

Hope this helps - if it doesn't quite work for you then if you can let us know what doesn't work I'm sure we can help further.

1 Like
3 
echo 'HiringManager_RHMC, DirectSupervisor_RHMC, ProcessServerReadAccess_ST' | awk -F, '$1=$1' OFS='\n'
1 Like
4

Not sure this works with all tr versions:

echo "HiringManager_RHMC, DirectSupervisor_RHMC, ProcessServerReadAccess_ST" | tr -s ', ' '\n'
HiringManager_RHMC
DirectSupervisor_RHMC
ProcessServerReadAccess_ST

tr --version
tr (GNU coreutils) 8.30
2 Likes
6

Does not work with all tr versions, because each character,range,class in arg1 should match a character,range,class in arg2.
The following is portable:

echo "HiringManager_RHMC, DirectSupervisor_RHMC, ProcessServerReadAccess_ST" | tr -s ', ' '\n\n'

The tr substitutes each space or comma, regardless where they occur.
The following sed script takes a " , " (two character) delimiter; a single comma or space remains.

echo "HiringManager RHMC, DirectSupervisor RHMC, Process,ServerReadAccess_ST" | sed 's/, /\
/g'
HiringManager RHMC
DirectSupervisor RHMC
Process,ServerReadAccess_ST

GNU sed takes \n in the substitution string, so you can do

echo "HiringManager RHMC, DirectSupervisor RHMC, Process,ServerReadAccess_ST" | sed 's/, /\n/g'