Re: [I18ndir] I-D on filesystem I18N

[John Levine <[e-mail]>]

In article <20200706225139.GJ3100@localhost> you write:
>The I-D covers the background of the issue, documents the existence of
>the VFS, explains why I18N behaviors belong at the filesystem layer
>instead of the filesystem protocol server layer, the need to expose I18N
>behavior metadata to clients, and provides advice to protocol designers
>and filesystem implementors.  It only deals with path components, not
>any other I18N issues.

It's interesting, but if I were someone maintaining filesystem code,
my response would be "who are you and why should I pay any attention?"

You are right that NFS clients need some way to find out whether two
different names refer to the same file, but the idea of a registry
seems like a non-starter, both because nobody has any incentive to
register and because I expect there are a lot of strange corner cases
like unpaired UTF-16 surrogates that would be very hard to describe.

What clients need is a way to ask the server to map a filename to a
handle, so the client can see if it's the same handle as one it
already has. I'm pretty sure NFS already does that. Even if you had
the case folding maps, clients would still need to do this because
Unix/Linux files can have several unrelated names.

R's,
John