SKIP grep, use AWK

Over the years, I’ve seen many people use this pattern (filter-map):

$ [data is generated] | grep something | awk '{print $2}'

but it can be shortened to:

$ [data is generated] | awk '/something/ {print $2}'

You (probably) don’t need grep

Following this logic, you can replace a simple grep with:

$ [data is generated] | awk '/something/'

This will implicitly print lines that match the regular expression.

If feel lost, I’ve got an illuminating series of posts about awk for you.

Why would I do this?

I can think of 4 reasons:

• it’s shorter to type
• it spawns one less process
• awk uses modern (read “Perl”) regular expressions, by default – like grep -E
• it’s ready to “augment” with more awk

But “grep -v” is OK…

It’s possible to emulate grep -v with awk, but it’s not a good idea:

$ [data is generated] | awk '/something/ {next} 1'

• it’s uglier
• it’s longer than grep -v
• “what’s does that even do?!” – it requires a deeper understanding of awk

UPDATE:

Many people have pointed out that “grep -v” can be done more consicely with:

$ [data is generated] | awk '! /something/'

which isn’t too bad at all.