UserAgent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.108 Safari/537.36

Example URL:

Steps to reproduce the problem:
1. Download a file 
2. The transfer get somehow stopped by the remote server
3. The downloaded file is only partially written to disk as a .crdownload file
4. There is no extended file attribute (xattr) written to the .crdownload file, and there is no way to know which was the original URI for that file (after several month and the download history has been cleared).

What is the expected behavior?
The extended file attribute should be written to the ".crdownload" file like any completely downloaded and written file in order to help track original URI for the broken file.

What went wrong?
The extended file attribute is not written to the .crdownload file, which makes it nearly impossible to recover the file form its original URI without having it written somewhere prior.

Did this work before? No 

Chrome version: 63.0.3239.108  Channel: stable
OS Version: 
Flash Version: 

This is responsible for a lot of data loss on my side, this shouldn't be too hard to implement since the extended file attributes are already written to downloaded files which have completed successfully.