Sort output by column

The following command line uses awk to prepend the last field of each line of file.txt, does a reverse numerical sort, then uses cut to remove the added field:

awk '{print $NF,$0}' file.txt | sort -nr | cut -f2- -d' '

You can use -k to specify where in the input line the sort key starts and ends.

You use the -n option or add n to the sort key specification to trigger a numeric sort as opposed to a lexical sort by default. And -r to reverse the sort.

Here, for a numerical sort, you don't need to limit the sort key to the sort command. If you do:

sort -rn file

It will treat the whole line as a number. The 14 hjoeow; line for instance will be understood as the 14 number.

You could do:

sort -bk 1,1rn file

To sort on the first field only, but that would make no difference.

The command to join lines is paste with the -s option. You can specify the delimiter with -d. \0 is a special value for the delimiter that means join with nothing in between. So, to get your expected output:

sort -rn file | paste -sd '\0' -

POSIX paste require a filename be passed. - is again a special value that means standard input. You can omit it with some paste implementations.