AWK Delimiter

Since awk field separator seems to be a rather popular search term on this blog, I’d like to expand on the topic of using awk delimiters (field separators).

Two ways of separating fields in awk

There’s actually more than one way of separating awk fields: the commonly used -F option (specified as a parameter of the awk command) and the field separator variable FS (specified inside the awk script code).

awk -F field separator

As you may have seen from the awk field separator post, the easiest and quicked way to use one is by specifying -F command like option for awk (in the example we’re extracting the last octet of the IPv4 address):

greys@maverick:/ $ ifconfig en0 | grep "inet " | awk '{print $2}'
192.168.1.220
greys@maverick:/ $ echo 192.168.1.220 | awk -F. '{print $4}'
220

Field Separator (FS) variable in awk

As your awk scripting gets better and more complex, you’ll probably recognise that it’s best to put such options inside the awk script instead of passing them as command line options. The benefit is, of course, that you don’t risk getting different (wrong!) results just because you forgot to specify a command line option – everything is contained in your script, so you run it with minimal set of parameters and always get the same result.

So, the second way of separating fields in awk script is by using the FS (field separator) variable, like this:

greys@maverick:~ $ echo 192.168.1.200 | awk 'BEGIN { FS = "." } ; {print $4}'
200

See Also

• awk
• awk field separator
• top 5 awk field separators
• Review logs with tail and awk
• Unix commands
• Advanced Unix Commands