View Full Version : [SOLVED] Strip first line from the results of the ls command

July 25th, 2011, 08:30 PM
Hi all , I have a directory which will be updated at each startup and I will need to access the most recent file to compare against the new one. I can access the most recent file using

ls -t

The problem is there are several files in this directory, and ls returns the list of the whole lot. I cant use regex's as all the files will use the date of creation as their names.

I need to be able to take the first result from the ls command and store it in a variable.

Any suggestions on what would be the best way to approach this

cut ? or sed or awk maybe ?



July 25th, 2011, 08:43 PM
Try this...

VARIABLE=$(ls -1t | head -n 1)
I added the '1'
(a number one, NOT a lowercase 'L')

July 25th, 2011, 09:00 PM
Cheers, that did the trick. I'll be posting the result when I'm done . I'm sure some people will find it useful.

July 25th, 2011, 09:15 PM
Please check out:

ls lists the files in human readable format, so you shouldn't use it for anything else, but listing files... ;)

If the modification time is in the filenames, then the glob order is also the mtime order. You can read the filenames in an array. The first element will be the oldest and the last the newest:

shopt -s nullglob
echo "oldest file: ${files[0]}"
echo "most recent: ${files[${#files[@]}-1]}"