How could I pipe a list of files to delete to the rm command? Yes, I know, very basic, but I just can't find how to do it!

rm < list.txt

Didn't work at all. Would be great if you could help me. Final problem to be solved since installing the myth2ipod script (it should really include a automatic "remove-old-files-to-save-diskspace" function).

Use with "find", like this:

find $HOME/dr/DAILY_*.zip -mtime +7 -exec rm {} \; >/dev/null 2>&1

This is from another sytem, but it's going to delete all DAILY_*.zip files over a week old. The "-exec rm" does the removal of whatever "find" found.
Use caution with this!
Notice that I qualify the full path. Less robust solutions will perform a "cd" first, to hop into a directory, then clean it out.
I have seen such scripts do horrible things. If the "cd" fails, the find/remove will chew on whatever directory you happened to be in. Probably recursively. Until it deletes something important, and the system crashes.

That's some extreme cabbage. I'll try it out. Carefully... Thanks!

backquotes ' are your friend. if you use them in a command line, the command within a set of backquotes is executed, the output of the command is inserted into the original command.



another technique is to turn your list.txt file into a shell script:



while in vi do this:


then


The vi step replaces the beginning of everyline with 'rm ' basically creatig a script that deletes one file per line.

I can explain more if you like.

the find command is very powerful and well worth learning. probably not required here since you have a file that contains the list of files you want to delete. find is a tool that will locate files (using lots of different criteria) and use the list of 'found files' in a command.

I use it daily at work. For example "find any log file that contains the string 'foo'"

_________________
R5F1 - Dell P4 2.4Ghz 500MB - PVR250 x 2 - GeForce FX 5200 - Onboard sound/NIC 80GB ATA/250GB ATA/400GB SATA

Nobody is going to mention xargs or the newer $() version of backtick expansion?

I love unix, it is like legos. Lots of small pieces that you put together any way you like.

I often use awk to build command lines and then I send it to sh for evaluation.

cat file | awk '{print "rm "$0}' |sh

I like that. I've added to my "tips" file, where I keep everything I know about Linux/Unix. It's a fairly short file...

I opted to use this command in crontab -e:

20 14 * * 0,3 find /myth/ipodfeed/*.ipod.* -mtime +10 -exec rm {} \; >/dev/null 2>&1

Does everything I want (I think ). Thanks for all the suggestions.

The above won't work if the file has more than 2000 (or so) entries. So, another method:

cat file | xargs /bin/rm

Putting the hard-coded path to "rm" prevents and funky aliases that are defined to take effect.
________
When Rover Met Bmw