Re: [STUMP] UTF8 related bug

[Stefan Reichör]

Vladimir Sedach <address@hidden> writes:

> You can check which encoding system Emacs uses for a particular buffer
> by doing C-h C Ret
> (http://www.delorie.com/gnu/docs/emacs/emacs_220.html). It looks like
> the shell is doing the "right" thing by displaying the '?'

(My) Emacs uses iso-latin-1-unix as coding system for file names:
,----
| default-file-name-coding-system is a variable defined in `C source code'.
| Its value is iso-latin-1-unix
| 
| Documentation:
| Default coding system for encoding file names.
| This variable is used only when `file-name-coding-system' is nil.
| 
| This variable is set/changed by the command `set-language-environment'.
| User should not set this variable manually,
| instead use `file-name-coding-system' to get a constant encoding
| of file names regardless of the current language environment.
`----

I will experiment setting file-name-coding-system to 'utf-8.

However, I found a way to reproduce the stumpwm crash by just opening a
pdf file, using evince.

The attached tarball contains the same file, using two different names.
When using the one that contains the iso-latin-1-unix coded ö, stumpwm
crashes.

Stefan.


> Looking quickly at the backtrace, it looks like the place to fix the
> crash in stumpwm is utf8-to-string in wrapper.lisp:212
>
> The utf-8 encoding/decoding functions right now are pretty ugly and
> have a bunch of implementation-dependent stuff in them. Replacing
> those with either flexi-streams or babel is going to cut out a lot of
> the code, provide more portability, and most important give a uniform
> error interface. Now it's just a matter of doing it.
>
> Vladimir

stump-test.tar.gz
Description: Binary data