df: Is printing the raw string guarenteed to have the same cell width as using the lossy string?

[ChrisDenton]

See this code:

coreutils/src/uu/df/src/table.rs

Lines 265 to 271 in 30fd234

	/// Create a cell from an `OsString`
	fn from_os_string(os: &OsString) -> Self {
	Self {
	bytes: uucore::os_str_as_bytes(os).unwrap().to_vec(),
	width: UnicodeWidthStr::width(os.to_string_lossy().as_ref()),
	}
	}

It gets the width of the lossy string but stores the raw bytes. I think this at least deserves a comment explaining this is intentional. Because it's not entirely clear to me if doing a lossy conversion may use more or less cells.