Copy but keep dupes

[jamesdcarroll]

I would like to find all the pictures on my system and copy them to a single folder. I found a little script that will see them out and do that:


find . -name '*.jpg' -type f -exec sh -c 'exec cp -f "$@" /home/jim/AllPictures' find-copy {}

but when it tries to copy files that have the same name as one that's already in the destination folder it won't overwrite which is fine, but I would rather that it make another copy(ie image.jpg, image1.jpg, image2.jpg... etc)

Would it be possible to do that? I know I'm gonna have dupes and will deal with that later. Right now I just want to get everything in one place.

Thanks!

[geirha]

The cp command can make numbered backups, which means that if the destination file allready exists, the destination file is renamed with a .~1~, .~2~, etc... extension. That would be the easiest way to get all copied to the same directory at least. Unless there are alot of dupes, renaming should be easy enough to do manually.


find . -iname "*.jpg" -type f -print0 |
xargs -0 -- cp -v --backup=numbered --target-directory="/home/jim/AllPictures" --

[geirha]

In case there are alot of dupes, this rename command will rename "something.jpg.~1~" into "something.1.jpg"


cd /home/jim/AllPictures/
rename -v 's/\.jpg\.~(\d+)~$/.$1.jpg/i' *

[jamesdcarroll]

Thanks! I saw the '--backup=numbered' thing in the man page but I didn't quite understand how to make it work or what its action was.