Nicolas Williams wrote:
> On Wed, Mar 04, 2009 at 09:23:49PM -0800, Joyce McIntosh wrote:
>
>> Such a file really shouldn't exist on Solaris.
>> The translation of '/' is listed in the table because the table is pretty
>> much a defacto standard. A filename with a '/' in it just shouldn't occur in
>> our environment, so the translation shouldn't be necessary.
>>
>
> If a filesystem has been used by other operating systems it might.
> Think of FAT, or NTFS (since it looks like the project to make NTFS
> usable through FUSE seems to be making progress).
>
> Does Windows normally allow the use of '/' in file names?
>
No. That's why it gets translated *IF* it occurs.
> In any case, I don't think this is going to ever really be an issue, and
> I suspect that the conversion for '/' was included in the spec for
> completeness.
>
Yes, the translation is included in the spec for completeness. It should
never occur.
Theoretically this particular translation ('/' <-> 0x00f8) should not be
required.
The impact of not supporting it would be that if SOMEHOW a file exists with
a '/' in it the CATIA windows client would not be able to access that
file. This
is extremely unlikely.
Joyce.
