Using awk to grab first number series

buee · 02-03-2010, 04:05 PM

Hello, just wondering if this can be done as I'd like to use it in a script.

I want to take this:

Code:

root@ubox:~$ du /etc
80	/etc/laptop-mode/conf.d
4	/etc/laptop-mode/lm-ac-start
4	/etc/laptop-mode/lm-ac-stop
4	/etc/laptop-mode/nolm-ac-start
4	/etc/laptop-mode/batt-start
4	/etc/laptop-mode/nolm-ac-stop
124	/etc/laptop-mode
16	/etc/defoma/config
160	/etc/defoma/hints
204	/etc/defoma
8	/etc/firefox-3.5/pref
12	/etc/firefox-3.5/profile/chrome
48	/etc/firefox-3.5/profile
60	/etc/firefox-3.5
8	/etc/byobu
4	/etc/opt
68	/etc/speech-dispatcher/modules
12	/etc/speech-dispatcher/clients
96	/etc/speech-dispatcher
8	/etc/xulrunner-1.9.1

And just keep the file sizes at the very beginning, no spaces, no letters, nothing after the first series of numbers. Then I want to output those to a file to be added together later. I can handle the addition part, but awk confuses me

If it has any bearing, the endgame here is to create a script that will count files and file sizes, add them together on a daily basis to come up with a total. We have a NAS for computer backups and I would like to be able to look at one line of text and see how much data and file count we've backed up from today.

rweaver · 02-03-2010, 04:08 PM

If you want a total number for a specific directory like /etc...

Code:

du -s /etc

I would also suggest that you use rsync to do your backup and you can get a summary from it along with any problems it encountered during the operation...

You could also use cut to grab just the numbers...

Code:

du /etc | cut -f1

But totaling that up is going to yield a lie because of the way du displays data.