Create a symbolic link tree that shadows a directory structure

find /home/user/doc/ -type d -printf "mkdir -vp '/home/user/Dropbox%p'\n" -o -type f -printf "ln -vs '%p' '/home/user/Dropbox%p'\n" | sh
Extremely useful to maintain backups if you're using Dropbox. This mirrors the entire directory structure and places symlinks in each to the original file. Instead of copying over the data again to the ~/Dropbox folder creating a symbolic link tree is much more sensible in terms of space usage. This has to be supplemented by another script that removes dead symlinks in the Dropbox folder which point to files that have been moved/removed. find -L ./ -type l -delete And then removing empty directories find ./ -type d -exec rmdir 2>/dev/null {} \; **Actually after some finding I found lndir which creates symbolic trees but it wasn't in the Arch repos so.. ;)

By: jnash
2009-03-29 09:25:12

What Others Think

Holy crap, that's compicated. If the src and dest are on the same filesystem (and you're on a *nix box), you may as well use hard links. Your command is reduced to: cp -al src dest Rather than dangling symlinks, which can also be found using the "symlinks" command, you will end up with files with a hard-link count of 1. find src -type f -links 1 -delete Finally you can tidy the empty directory command: find . -depth -type d ! -name . -exec rmdir --ignore-fail-on-non-empty {} \; This version will remove deep, but empty trees, and it won't complain about non-empty dirs.
flatcap · 691 weeks and 2 days ago
Er, .. well I'll take that as a compliment :) But, in all truthfullness, I think it is simple. find just branches into two (if/else) for dirs and files and either makes a directory or makes a symlink to the file. Relatively, however, I guess yours wins in terms of complexity. But hard links might be a tad little more unsafe IMO considering I'm backing up stuff..
jnash · 691 weeks and 1 day ago
If you're trying to back up data, this is very much the wrong way to do it. Symlinks are NOT a backup. All they do is point to a file. If that file is changed, the symlink reflects that. Try this: echo foo>file;ln -s file symlink;cat file;echo whoops>file;cat symlink This creates a file, symlinks it, edits the file, and changed your "backup". Hard links aren't safe either. do the same thing, only, you have to look harder to find out it's a link. echo foo>file;ln file hardlink;cat file;echo whoops>file;cat hardlink The only way to back up the data is to actually copy it elsewhere. If you want to back up and save space, then rsync is your friend. It has some parameters that will compare to previous backups, and copy any changed files since the last backup, and hard link the rest. This way, you can delete any backup you want, and all others are unaffected. I can't tell you what they are, but google will help out: You may also be interested in Flyback, , which is a gui wrapper over rsync. Good luck!
clockworkavian · 691 weeks and 1 day ago
OP mentioned "dropbox", I think that copies the stuff to the server. I am interested in expanding this idea to re-point the symlinks to other locations if the information has moved. it won't be foolproof but something that "finds" and repoints the link. Only ways I can think of doing this involve long scripts, so oneliners would be apprecaited if anyone can think of one.
unixmonkey3054 · 691 weeks and 1 day ago
@clockworkavian I think this is useful in the dropbox scenario, maybe you have a folder full of massive files and it doesn't make sense for you to keep it within your dropbox folder, by symlinking you are not using up local HDD space but it will still get uploaded to dropbox. I've done this by just symlinking the root folder. I'm interested in this mirroring approach however as there are certain files in my source folder that are massive and would fill up my dropbox however I am wanting to backup the smaller files in there so I am currently attempting to modify this script to accept a filetype filter.
stubar · 386 weeks and 5 days ago
test (wiki)[] [[ wiki]] [[|admission essay service]] [wiki]( [ wiki] wiki: "link title": formatting
Latricia31 · 59 weeks and 5 days ago
An mvp software design is the best way to start any project because it allows for agile development cycles. This means that your product gets to users faster, who then will provide crucial feedback that will then help your product become even better in the next development cycle.
Latricia31 · 59 weeks and 5 days ago
seofox · 31 weeks and 3 days ago ????? ??? ????? ????? SEO - ????? ???? ????? ????? ????? ??? ????? ????? SEO - ????? ???? ????? ????? ????? ??? ????? ????? SEO - ????? ???? ????? ????? ????? ??? ????? ????? SEO - ????? ???? ????? ????? ????? ??? ????? ????? SEO - ????? ???? ????? ????? ????? ??? ????? ????? SEO - ????? ???? ????? ????? ????? ??? ????? ???? SEO ????? ????? ????? ???? ????? ????? ??? ???? ???? ??????? Search Engine Optimization ??? ????? ??????????? ?? ????? ?????????? ????? ????? ?? ???? ???SEO
seofox · 20 weeks and 6 days ago
seofox · 20 weeks ago
tehrandasht · 3 weeks and 5 days 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? 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.


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: