Q: I have a comma separated file that I want to loop through and put each column on it’s own line. How can I loop through the csv file and echo each field?  The catch is I don’t know how many fields are going to be in the csv file.

A: I thought about this for a while and came up with this solution.  I am far from a programmer and not great with scripting, but it get’s the job done.  Using an incrementing variable and a count of the fields to create a while loop.  I commented the code with my logic (as inefficient as it may be) and highlighted the comments below.

#!/bin/bash
# Name of the csv file to loop through.
INPUTFILE="pat.csv"
# Set a FIELD variable at 1 to test when to stop looping
FIELD=1
# Find how many columns are in csv and tick by 1
COUNT=`sed 's/[^,]//g' $INPUTFILE | wc -c`; let "COUNT+=1"
# Keep looping until the field is less than the count+1 (until all fields are caught in the loop)
while [ "$FIELD" -lt "$COUNT" ]; do
# Read the file and pull the current field number and print it to terminal
        cat $INPUTFILE | cut -d, -f$FIELD
# Increment the FIELD variable
        let "FIELD+=1"
done

Script without the comments:

#!/bin/bash
INPUTFILE="pat.csv"
FIELD=1
COUNT=`sed 's/[^,]//g' $INPUTFILE | wc -c`; let "COUNT+=1"

while [ "$FIELD" -lt "$COUNT" ]; do
        cat $INPUTFILE | cut -d, -f$FIELD
        let "FIELD+=1"
done

Let’s try it on a test file. For this test I created a csv file with the following contents:

Name,Address,Phone,Cell,Email,Nickname

Here is the output of the script:

Name
Address
Phone
Cell
Email
Nickname

I am sure some people are reading this and think they have a much better way, and they probably do.  I hope they will share them in the comments.

Also see "Column Command Usage and Examples" if you work with csv files on the command line often.

Links and Resources