Re: Error bunzip2ing files with non-ASCI-names

[Heinz Rommerskirchen <Heinrich.Rommerskirchen@siemens.com>]

Kevin Rodgers <ihs_4664@yahoo.com> writes:

> Rommerskirchen, Heinrich wrote:
snip
>> On my GNU/Linux (SuSE 9.3) workstation emacs is unable to bunzip2
>> files whose names are not ASCII, maybe because file names are utf-8
>> encoded on
>> this system.
>> Steps to reproduce this bug:
>> In the shell:
>>   > echo xx > bä.txt     # the second letter of the filename is a-Umlaut
>>   > bzip2 bä.txt   > emacs -q --no-site-file
>> In emacs
>> - type "M-x auto-compression-mode"
>> - open a dired buffer which contains the file (the filename is displayed
>>   correctly with the Umlaut), - move point to the file and press enter
>> The error message "File exists, but cannot be read" is displayed and an
>> empty
>> buffer is created. If I rename the file with dired-do-rename to
snip
>
> What happens if you visit bä.txt with `C-x C-f' instead of via Dired?
>
> -- 
> Kevin Rodgers

I assumed you mean open bä.txt.bz2 with `C-x C-f'; this gives the same
behaviour (error message and empty buffer). bä.txt can be opened in any
way I can think of without problems

-- 
Regards

Heinz