>_ IzzyLab

Topic breakdown

count N counts the number of lines (values) in a column, while countunique N counts the number of distinct values, ignoring repeats (like COUNT(DISTINCT ...) in SQL). unique and collapse are useful alongside it: the first collects the group's unique values into a list, the second all values in a row, repeats included. countunique is often combined with grouping: datamash -s -g 1 countunique 2 gives the number of unique values of the second column within each group — for example, how many different users logged in from each IP.

What's checked

  1. unique count is correct

What it looks like