Re: File name completion with non-ascii filenames

[Eli Zaretskii]

On Mon, 14 Jan 2002, Richard Stallman wrote:

>       - The completion functions could have been called by a Lisp program,
>       such as somethink ffap-like, that gleans the partial file names
>       from the buffer text.  If the buffer text has mule-unicode-*
>       characters, but file-name-coding-system is latin-2, we have the
>       same situation again.
> 
> Is it really a serious problem to fail in that case?

The completion could fail to display some of the files.  It's not a 
disaster, but is quite annoying, I think.  It's also possible (I didn't 
try) that you'll get "New file" if you type a full name of an existing 
file.

> If we wanted to handle that case, how could we do it?

If we had a way to downcase encoded text, we could solve this with a 
suitable modification to the DOWNCASE macro.  The problem with the 
current handling of unibyte characters is that case-conversion tables 
have place for only one set of unibyte characters, so we set them up for 
the preferred coding system defined by the current language environment.  
What is needed here is many case-conversion tables for unibyte 
characters, one each for every encoding we support.