extracting files from tar

Question

tar -xvf company_raw_2012-03-16.tgz --directory=/root/test --strip-components=4

I am using the following tar option to remove the leading directories and it is working as expected.

--strip-components NUMBER
strip NUMBER of leading components from file names before extraction

It works only when I know that there are going to be 4 sub-directories. I have tar files and I do not know if there will be 2, 3 or 4 folders inside. How do I strip the entire path and extract files in the given "directory" path.

Answer 1

You have two options: read the tar file first ( tar tf ) and count the / in the output, then extract; or extract first, then move back all files up and delete empty directories.

Here is one way to achieve the first approach:

tar tf archive.tgz | tail -n1 | perl -ne '@l=split(/\//);print "$#l\n"'

Other methods are left as an exercise for the reader :)

Answer 2

You can just extract the tar file as normal without adding the switch --strip-components. Then, you can use the following find command to move all extracted files to some destination directory:

$ find <your_extracted_dir> -type f -exec mv -t /path/to/dest/dir {} \;

Answer 3

tar --transform 's#.*/\([^/]*\)$#\1#' -xvf <filename>

this will use the regex on the file name and strip all the path from the filename so all files are extracted to the current dir.

Answer 4

Inspect the tarball before issuing the extract with --strip-components by using:

tar -tzf filename.tar.gz | less