Duplicate a directory tree using tar and pipes

(cd /source/dir ; tar cv .)|(cd /dest/dir ; tar xv)
the f is for file and - stdout, This way little shorter. I Like copy-directory function It does the job but looks like SH**, and this doesn't understand folders with whitespaces and can only handle full path, but otherwise fine, function copy-directory () { ; FrDir="$(echo $1 | sed 's:/: :g' | awk '/ / {print $NF}')" ; SiZe="$(du -sb $1 | awk '{print $1}')" ; (cd $1 ; cd .. ; tar c $FrDir/ )|pv -s $SiZe|(cd $2 ; tar x ) ; }
Sample Output
copy-directory /media/cdrom /mnt/backup/
 104MB 0:00:06 [7.43MB/s] [============>                                                      ]  9% ETA 0:00:58

2 Alternatives + Submit Alt

What Others Think

how about a good old cp, since you're on the same machine? Also, I think you're missing the - from your tar commands because tar doesn't use stdin/stdout by default.
bwoodacre · 482 weeks and 4 days ago
From man pages of tar. -f, --file [HOSTNAME:]F use archive file or device F (otherwise value of TAPE environment variable; if unset, "-", meaning stdin/stdout) And I use it mostly on large folder copies on a mounted network drive. i.e my copy-directory function
marssi · 482 weeks and 4 days ago
tar cvpf - . | tar -C /where/you/want/the/copy xvpf -
unixmonkey4704 · 482 weeks and 4 days ago
BITD you need -f - because the default was a tape drive device, not stdout :-) However, on a modern system, this doesn't beat cp -av (or if you want things like --exclude, rsync -av)
eichin · 482 weeks and 4 days ago
You can also use find+cpio: cd /src; find . | cpio -dump /dst A nice variant is the -l option: It doesn't really copy the files, but creates a tree of hardlinks. Only works within the same file system, of course: cd /src; find . | cpio -dumpl /dst If some files contain white space characters, then you should use the -print0 option with find, and the -0 (zero) option with cpio. On BSD systems, the cpdup command is most convenient and very efficient: cpdup /src /dst By the way, the cp command (with -r or -R) should be avoided. It's not portable, the exact semantics are not well-defined. Notably it doesn't copy all kinds of files correctly, on some platforms it doesn't handle hardlinks, FIFOs or other special files correctly.
inof · 482 weeks and 2 days ago
The function is really overcomplicated. Rather than 'echo | sed | awk' you can use parameter expansion, ${1##*/} ... which will return the last item of the path. Also with the use 'sed |awk', awk understands fields, and you can tell what the FS is, so the above can abreviated to: awk -F/ '{print $NF}' ... though again your better off with parameter expansion.
khayyam · 289 weeks and 1 day ago

What do you think?

Any thoughts on this command? Does it work on your machine? Can you do the same thing with only 14 characters?

You must be signed in to comment.

What's this?

commandlinefu.com is the place to record those command-line gems that you return to again and again. That way others can gain from your CLI wisdom and you from theirs too. All commands can be commented on, discussed and voted up or down.

Share Your Commands



Stay in the loop…

Follow the Tweets.

Every new command is wrapped in a tweet and posted to Twitter. Following the stream is a great way of staying abreast of the latest commands. For the more discerning, there are Twitter accounts for commands that get a minimum of 3 and 10 votes - that way only the great commands get tweeted.

» http://twitter.com/commandlinefu
» http://twitter.com/commandlinefu3
» http://twitter.com/commandlinefu10

Subscribe to the feeds.

Use your favourite RSS aggregator to stay in touch with the latest commands. There are feeds mirroring the 3 Twitter streams as well as for virtually every other subset (users, tags, functions,…):

Subscribe to the feed for: