The active_users.sh script reads from /var/log/wtmp or a wtmp log file defined on the command line. The last -f command extracts the log file contents. The first column in the log file is the username. The cut command extracts the first column from the log file. The sort and uniq commands reduce this to a list of unique users.

The script's outer loop iterates through the users. For each user, grep is used to extract the log lines corresponding to a particular user.

The last column of each line is the duration of this login session. These values are summed in the inner while read t loop.

The session duration is formatted as (HOUR:SEC). This value is extracted with awk to report the last field and then piped to tr -d to remove the parentheses. A second awk command converts the HH::MM string to minutes and the minutes are totaled. When the loop is complete, the total minutes are converted to hours by dividing $minutes with 60.

The first login time for a user is the last line in the temporary file of user data. This is extracted with tail and awk. The number of login sessions is the number of lines in this file, calculated with wc.

The users are sorted by the total usage hours with sort's -nr option for the numeric and descending order and -k4 to specify the sort column (usage hour). Finally, the output of the sort is passed to awk, which prefixes each line with a line number representing the rank of each user.