Question

I have a file structured as record list, where field separator is \t.
I want to extract only records where the second field is a number from 1 to 9, but my awk script doesn’t work.
The awk script is

cat file |awk -v FS="\t" '$2 ~ /[0-9]{1}/ {print $0;}'

or this

cat file |awk -v FS="\t" '$2 ~ /.{1}/ {print $0;}'  #because the second fields of my file have  all second fields as number

Why these sscript don’t work? Isn’t regex a good regex?

Answer 1

Update

Even with the interval {1}, you are still going to match a field like 23 because the 2 matches a single number. What you really want to use are anchors and forget about intervals:

awk '$2 ~ /^[0-9]$/{print}' FS="\t" file

---

The problem is the use of intervals {1}. awk less than version 4 doesn’t support intervals. gawk on the other hand will if you add the following flag: --re-interval

Try this:

awk --re-interval '$2 ~ /[0-9]{1}/{print}' FS="\t" file

Some other things to note:

1. Built in vars such as FS can be assigned at the end without the need for -v
2. You can use just print rather than print $0 as that is its default behavior
3. Useless use of cat. awk can take a file as an argument, use that instead