Using shell instead of awk for printing columns

There are two typical ways to get say the first column of a file. Most people currently use awk, like so:

awk '{print $1}' webc-6.0.txt

See my [[awk_tip|01085]] for more.

However did you know you can use bash like so?

while read -r -a fields; do echo -e "${fields[1]}"; done < webc-6.1.txt > webc-6.1-1col.txt

To get the last field using bash, use:

while read -r -a fields; do echo "${fields[@]:(-1)}"; done < webc-6.0.txt

However awk is still much faster than shell for this task.

If your columns are delimited with say a full stop, the recipe may help you:

while IFS=. read -r -a fields; do echo "${fields[@]:(-2)}" | tr ' '  ','; done < $1 | grep .