I just migrated my Documents folder contents to a new installation on a desktop computer from a laptop. Disc drive on the laptop went out so I had to send it in for recycling.

In the process of migrating the contents, I created (out of ignorance, by the way) a gzip file that became corrupt in the process of getting it to my desktop hard drive.

Read around and tried a gzrecover utility, which did not work. I have tried many other things and have spent hours on it, getting to what seems to be nowhere.

Does anybody out there have a surefire way to uncorrupt the file or get out the remaining contents of the file that are still intact?

Much thanks in advance for your help.

Did you try the cpio option mentioned in the gzrecover site?
http://www.urbanophile.com/arenn/hacking/gzrt/gzrt.html

The tables used for compression are built using data previously encountered. Some of the files after the corruption won't be recovered correctly as a result. I think a sliding window is used so after so many bytes of corrupted files, you may start getting correct output.

If the file was truncated, you will be out of luck, as there is nothing to recover.

The cpio method is used because tar can't handle the region of file corruption from gzrecover.

Success in recovery depends on just how the corruption occurred. If the problem is due to an FTP ASCII-mode transfer from a Linux machine to Windows machine, then the corruption was non-destructive (inserting a CR character ahead of each LF character), and there is a program called fixgz that has a good chance of repairing the file. If the transfer was in the reverse direction (Windows to Linux), then the corruption was destructive (removal of every CR immediately followed by LF) and recovery may be impractical (not impossible, but probably beyond what is reasonable for data that wasn't important enough to be backed up).

gzrecover threw an error when I tried to get an output. It said

# gzrecover: Command not found.

I don't think I have it installed properly.

What is the proper way to install it?

Thanks!

I have done some reading around and have decided to give up on the effort of trying to extract my good data from the archive. Thanks anyway for your help, but for now, it is not needed.