Andrew E. Mileski (
Tue, 18 Nov 1997 09:09:43 -0500 (EST)

> > what are the alternatives? [We've come full circle.]
> This highly depends on your file system. If UDF works similar to
> 9660, I see two options:
> a) use the block number of the first block of the file
> b) use a hash value of the file name

UDF is used on much bigger disks (like DVD), it also can span media
(up to 64k volumes). It really neads a 48-bit inode number.

I think I found a passable solution though. Use the lower half of the
64-bit unique ID of the _first_ file entry for a file.

On write-once media, a file entry is revised by creating a new entry
which must have a different unique ID [unless the entry is identical,
which is useless]. On re-writable or overwritable media, there only
needs to be one file entry for a file.

The UDF spec recommends the unique ID be less than 2^32 for Macintosh
compatibility, but it doesn't insist on it. [This limits the number of
files on all media, and the number of file revisions on write-once media.]
So we'll have to scream if the upper half of the ID is ever > 0.

It is possible to access a file by this inode number, but it would be
slow to open as the file system must be searched.

The perfect solution though would be for Linux to move to 64-bit
inode numbers.

Andrew E. Mileski