Re: File name completion with non-ascii filenames

[Eli Zaretskii]

On 17 Jan 2002, Kim F. Storm wrote:

> Eli Zaretskii <address@hidden> writes:
> 
> > On Wed, 16 Jan 2002, Richard Stallman wrote:
> > > 
> > > It is possible to downcase encoded text by decoding it into multibyte,
> > > downcasing it, and re-encoding it.  Will this do the job?
> > 
> > Yes, I think so.  It is slower, of course.
> 
> But you would only need to do that when the normal unibyte case conversion
> table cannot be used - so the slowdown would be an exception I think.

I thought about this optimization, and asked Handa-san whether it was 
possible.  Unfortunately, the answer (AFAIU)was NO.

The problem, as I understand it, is that there's no way of telling which 
file-name-coding-system corresponds to the unibyte case-conversion table.