Bonus: Substitute: Reference Match Groups

BONUS

This entire section is a bonus. It probably won’t be covered, but you can run the examples to see what is happening, and learn even more about RegEx and sed.

Bonus: Reference Groups

Enter a Nickname

sed -E 's/(Phillip),(Holmes),/\1 "Phil",\2,/' user-data.corrected.csv | grep 'Phillip "Phil"'

The replacement text is:

• \1: text from first match group
• "Phil",: the exact string of one space, quotes around Phil and a comma
• \2: text from second match group
• ,: an exact comma

Switch first_name & last_name columns

sed -E 's/^([^,]+,)([^,]+,)/\2\1/' user-data.corrected.csv

The first_name and last_name columns were switched!

Switch all the columns around

sed -E 's/^([^,]+),([^,]+),([^,]+),(.*)/\4,\2,\1,\3/' user-data.corrected.csv

Reorders all the records in the file to be:

1. Company
2. last_name
3. first_name
4. email