How do I separate columns using sed

Basse1 · 07-01-2010, 11:37 AM

I have a file that contains a couple of email addresses and I want to extract the usernames ( Letters before @ symbol ). How can I do that using sed/awk.

I know cut will work, but the current environment doesn't allow me to use cut command. I can use either awk or sed.

Eg: in the email address something@example.com, I want to extract "something".

Suggestions?

anon237 · 07-01-2010, 11:45 AM

Hi

It depends a bit on where the email address is located. If it is the first and leftmost part:

awk -F@ '{ print $1 }' infile
or
sed 's/@.*//' infile

If it is not the leftmost part: Please provide an example.

Hope this helps.

Basse1 · 07-01-2010, 11:51 AM

Quote:

Originally Posted by druuna

Hi

It depends a bit on where the email address is located. If it is the first and leftmost part:

awk -F@ '{ print $1 }' infile
or
sed 's/@.*//' infile

If it is not the leftmost part: Please provide an example.

Hope this helps.

Ah.. Cool! Thank you!

Andrew Benton · 07-01-2010, 12:02 PM

Maybe you should split the file into separate lines for each word and then delete all the lines which don't contain @
It feels quite naff to me to pipe one sed into another. There must be a way to do it all with one sed

Code:

sed 's/ /\n/g' infile | sed '/@/!d;s/@.*//'

Basse1 · 07-01-2010, 12:54 PM

Hi Andrew,

It is nice indeed..