How let `sort | uniq -c` separate the number of occurrences by a tab charactor?

Question

uniq -c separate the occurrence number by spaces, which is hard for to cut or awk to separate it out later.

1000_A1\tB1\n
___1_A2\tB2\n

I can solve this problem using sed -r 's/^ *([0-9]+)/\1\t/' to change the delimiter to tab. Then cut -f1 could return:

1000\tA1\tB1\n
1\tA2\tB2\n

But it seems a common usage to have uniq -c separate the number by tab. Why is this feature missing? Is there any other easier way to do so?

Answer 1

Sounds like you've got a format like <number><space><field 1 name><tab><field 2 name>, and you want to check that the input is sorted by field 1 name. If that's what you want, simply remove the initial number part and check the sorting of the remaining part of the first column:

echo "$input" | sed -r 's/^ *[^ ]+ //' | sort -c -k1,1

Answer 2

You can easily use cut to separate them out. Just use cut -d " " -f 1 to get the number and cut -d " " -f 2- to get the rest. Awk will just process it as it is whitespace.

Answer 3

I'm not sure why you think it's hard for awk to separate the number. awk '{print $1}' prints it without a problem.

$ sort file | uniq -c
      5 x
     10 y
$ sort file | uniq -c | awk '{print $1}'
5
10
$