(Oops... this was meant for the list, and I only sent it to Cameron Kaiser...)

OT WARNING....

Question for the list.

Given that the MacOS on the client side only knows about 31 character
filenames, (not 32) how would you approach this problem?

There is no "console" on the user/client side to tell you that there are
files on the server, but if I showed them to you I'd be lying.

If the three files below were displayed in a window, they'd all have
the same name.  I know of no mechanism in MacOS to deal with this
in any reasonable way.

What **should** Netatalk do.  Even logging - what should it log, and in
which cases?  Would you want it to log the name of every file that
has been in some way referenced, but has a name > 31 chars?
If a file request comes in to netatalk for a 31 char file, does it
log any potential ambiguity?  Does it refuse service if there is
ambiguity?  What error?

I'm not so certain that silently failing is such a horrible thing,
given that the idea of netatalk is that the files are created and
consumed on the Macintosh.  Hence, these >31 names can't be
created in the first place.

-dgl-

>> The fact is that Netatalk could be doing a more right thing than it
>> is, by truncating the names or at least by logging a message on the
>> server when a client tries to list a directory which contains file
>> names they won't be able to get.
>
>I would strongly prefer it didn't willy-nilly truncate. What if you had
>filenames like
>
>Document Discussing Some Inane Topic Of More than 32 Characters Part 1 of 114
>Document Discussing Some Inane Topic Of More than 32 Characters Part 2 of 114
>Document Discussing Some Inane Topic Of More than 32 Characters Part 3 of 114
>:
>:
>etc.
>
>You couldn't come up with a general rule for truncation in that regard
>unless you resort to filth like MICROS~1's long file name kludge.
>
>But I like the logging idea and I agree that ...
>
>> Failing silently is clearly wrong.
>
>Yes :-)
>
>-- 
>----------------------------- personal page: http://www.armory.com/~spectre/ --
> Cameron Kaiser, Point Loma Nazarene University * ckaiser@stockholm.ptloma.edu
>-- The world will end at 3 p.m. today, to be followed by a brief symposium. ---