Regex

We can also use regular expressions to filter the data before we process it.

#Matching

We continue to use phones.txt. Let’s say that we only want to print the lines where the string pp is included:

#!/usr/bin/env awk

{
    if ($0 ~ /pp/) {
        print $0
    }
}

The condition $0 ~ /pp/ compares the whole line with the letters pp. We could have flipped it around with $0! ~ /pp/.

From the commandline the expression would have looked like this: $ awk '/pp/' phones.txt.

#match()

There is a built-in function that helps us with regex, match(). We have a look at an example:

#!/usr/bin/env awk

BEGIN {
    FS=","
}
{
    ans = match($2, /[a-z][[:digit:]]/)

    if (ans) {
        print $2
    }
}

Here we select the phones where the model name has a small letter followed by a number. The answer is:

$ awk -f phones.awk phones.txt
Galaxy Note20

#Revision history