Who cares if the name mangling scheme is different? As long as an
alias maps to long name with some level of probability, then who cares
if the 10th time you have to mangle a filename matching "biteme123*"
you get something different on NT, Linux, and Win95? Before switching
to a new collision avoidance scheme, I think NT goes up to 'biteme~6',
linux goes up to 'biteme~9', and I have no idea what Win98 does up to.
To me, capitalization is a far more annoying difference between how
Win95 and NT implement vfat. On NT, there are two bits for 8.3 names
that determine capitalization. One specifies the case the 8 part of
the name, and one specifies the case of the 3 part of the name. Win95
ignores these bits, so try writing filenames that fit in the 8.3 space.
Their capitalization will be inverted on NT. (And don't look at it in
Explorer which does its own filename case handling that has nothing to
do with what is on the disk.)
> To be 100%
> compliant with all uses of a FAT partition, it's probably better not to use
> these names.
What a bunch of crap! OS/2 defines a bunch of reserved names, so does
NT, so does Win95, and so do any other versions of DOS. This has nothing
to do with compliance. Worse yet, you can define new devices that get
reserved names. So the only 'compliant' way would be to allow no names
to be written to the disk. I should probably throw out the reserved
name checking entirely because it varies so much between systems that
have FAT filesystems.
- Gordon
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/