Move file but keep directory tree

[Mike Ehrman]

I am writing a script that searches for files with specific extensions and moves them to another directory.

I'd like to be able to keep the directory tree of the file when the file is in it's new location.

Example: I'd like to move the file stuff.gz from /home/jerry to /var/tmp/home/jerry.

Any suggestions on how to do this within the structure of my already written script?

(Main part of my script is below)


find / -name \*.gz >$OUTPUT

for i in $(<$OUTPUT)
do
{insert command here}
done



Thank you for any suggestions you may have.
Mike

[Alan Meyer_4]

I would do a cp -rp and then remove the source file. The cp -rp would create the necessary directory structure and keep the same file protection as the source file. So the script would look like;

find / -name \*.gz >$OUTPUT

for i in $(<$OUTPUT)
do
cp -rp $i /var/tmp/$i
rm $1
done

[A. Clay Stephenson · Accepted solution]

Easy:

find /home/jerry -name '*.gz' | cpio -pudvm /var/tmp

[Rick Garland]

The cpio passthru option as listed by Clay.
As the root user;

# cd /home/jerry
# find . -depth | cpio -pmuldv /var/tmp/

This will keep perms and owners as well.

[Mike Ehrman]

Thank you Alan, Clay and Rick.

Alan - the cp command with the rp options did not create the directory structure on my HP-UX 11i box. The man page says p is for keeping permissions.

Clay & Rick - the cpio command works.

Here is how I incorporated it into my scripts...

find / -name \*.gz >$OUTPUT

cat $OUTPUT | cpio -dumpv /var/tmp



If anyone has a better way to incorporate it into my script I'd love to hear it.

Mike

[Rick Garland]

Not exactly sure what you are looking for but give it try:

Instead of the 2 lines you have, just have the single line as listed previously posted.

find /home/jerry -name '*.gz' | cpio -pmuldv /var/tmp


You can have just the 1 line doing the copies.

[Alan Meyer_4]

Hmm it's the -r that's suppose to create the subtrees though... In anycase, what's important is that you got it solved... :)

[Vibhor Kumar Agarwal]

Lets's try this way out,

for filename in `find / -name \*.gz`;
do
newdir=dirname $filename
mkdir -p /var/tmp/$newdir
mv $filename /var/tmp/$newdir
done

[Arunvijai_4]

Put your $OUTPUT in a loop and pass through it, It would be efficient and good way of doing.

-Arun

[Muthukumar_5]

You can do it simply as,

# find / -name "*.gz" | awk '{ print "cp -R "$0" /var/tmp" }' | ksh

It will copy directory sturcture to /var/tmp. You can change location also with awk statement.

hth.