Conversation
| return cls( | ||
| serial=fs.ntfs.serial, | ||
| volume_uuid=get_volume_identifier(fs), | ||
| volume_uuid=fs.identifier, |
There was a problem hiding this comment.
I think the behavior silently changed of this. In get_volume_identifier, we preferred the volume GUID over the serial. However, in the new fs.identifier we prefer the serial over the volume GUID.
I'm not entirely sure which is nicer. Up for discussion.
There was a problem hiding this comment.
im fine with either.
honestly i dont mind keeping the old function as deprecated function, it just seems kinda silly as AFAIK its barely used anywhere.
we could also easily change the order of fs.identifier, if you prefer.
There was a problem hiding this comment.
On second thought, the serial is already available in the line above this. Could be make this to be exclusively the volume UUID?
There was a problem hiding this comment.
But wouldn't that put us back at square one? The serial wouldn't get yielded, so when an NTFS volume is using the old MBR approach, we wouldn't be able to differentiate between filesystems. Also, since the serial isn't used anywhere else, maybe we could just delete it altogether and rename this field?
| return cls( | ||
| serial=fs.ntfs.serial, | ||
| volume_uuid=get_volume_identifier(fs), | ||
| volume_uuid=fs.identifier, |
There was a problem hiding this comment.
On second thought, the serial is already available in the line above this. Could be make this to be exclusively the volume UUID?
3 participants
fixed depreciated function, since we have a better one implemented.