Comment 4 for bug 799157

Hi Sergio - Thanks for the contribution.

Inode numbers are a great way to correlate upper and lower inodes. However, they're just a quick and dirty way to transform a plaintext filename to an encrypted filename (or vice-versa). The reason is that multiple dentries may point to a single inode.

Therefore, this script could give inaccurate results and I'm afraid users could shoot themselves in the foot by not understanding that subtle detail.

It is useful (I use this same technique at times), so maybe we can ship it in the source tree, but not install it into /usr/bin.

Also, the script is much too complicated for what it does. All you need is a stat call to get the inode number and then find (with the -inum argument) to find the corresponding inode in the lower filesystem. I think Dustin is working on a simpler version which does something along these lines.