Re: File name completion with non-ascii filenames

[Kenichi Handa]

Eli Zaretskii <address@hidden> writes:
> On Tue, 25 Dec 2001, Kenichi Handa wrote:

>>  Ah, yes, that is a good idea.  In C code, we can use
>>  Ffind_coding_systems_region to get a list of coding systems
>>  that can encode FILE.
>>  
>>  If `base coding system' of file-name-coding-system is not
>>  included in the returned list, we can return nil.
>>  
>>  Otherwise, we can use the current method.

> This still leaves the issue of downcasing the encoded file name.  The 
> situation that bothers me is this: suppose the user's language 
> environment is Latin-1, but file-name-coding-system is Latin-2 or one of 
> the cpNNN.  Will the DOWNCASE macro DTRT with the encoded file name in 
> this case?

Ah! I see your point.  You are right.  In such a situation,
DOWNCASE macro on encoded file names doesn't work well.  In
addition, I've just noticed that DOWNCASE anyway doesn't
work well for such a language environment that is based on
2-byte charsets (e.g. Japanese lang. env.).

So, it seems that we must use a slow method at least in the
case that completion-ignore-case is non-nil (it is nil by
default).

---
Ken'ichi HANDA
address@hidden