Hi, I would like to delete the files in the list below. The file list is over 5000 lines long. It looks something like this: test.12001 test.12002 test.12003 ... I would like to remove all these testfiles with a script. How can I do this? BR Mats

Using sed, you could do something like: #! /bin/sh sed '/^test\.[0-9]\{5\}$/d' file > file.new mv file.new file This removes a string that begins with 'test', followed by a literal period, followed by 5 numbers at the end of a line. -jim

Or if you meant you want to feed the 5000 filenames into an 'rm' command (i.e. delete the 5000 files listed), try: rm $( deletethese.dat) Make sure that is what you want to do first though - rm can be a weapon of mass destruction. A less efficient bit more flexible approach would be to read each line: for f in $(deletethese.dat) do print rm $f done (remove the 'print' above to actually delete the files)

Hi, I tried your little script above on my file but i didn't remove the files that were in the list. Do I need to run it in any shell (ksh, bash)? What permissions should have on the file that contains the list of files? How can I see if there is a problem with the script?

Apologies, the less-than symbols got lost in posting. That should have been: Mass-destruction version: rm $(Line-by-line version: for f in $(do print rm $f done "$(

...nope, that's even worse. Trying again... Mass-destruction version: rm $(Line-by-line version: for f in $(do print rm $f done "$(

You know, some days I get this to work without any problems (using HTML '<')... That was supposed to show rm $(less-than deletethese.dat) and for f in $(less-than deletethese.dat) do print rm $f done