On Fri, 26 May 2000, Kai Gro�johann <[EMAIL PROTECTED]>
wrote:
> "Daniel Pittman" <[EMAIL PROTECTED]> writes:
>
>> I think that what we really need is to get the coding system used for
>> filenames. I don't really know the LANG stuff thought. 'ls' should
>> produce, like, a list in the local coding system for filenames, yes?
>
> Omigod. Hadn't thought about this at all. I don't see how to find
> out about the coding system used for file names, since / is the only
> directory which we know exists on all machines, and files in that
> directory are most likely to only have ascii names.
>
> I see that this requires some serious thinking. I'll ponder it for a
> few weeks, but I'm making no promises as to what will be the result.
Yup. I thought 'bout it briefly, then paled a little and shuddered.
I am almost tempted to suggest some sort of per-machine configuration
for RCP, detailing the type of connection, coding systems, etc. I think
this has been requested before as well.
That kind of changes the way the RCP works though, and it's not really
very Emacs-ish, I think.
The other option, I think, is to add a 'coding system' member to the
connection type object. That could have the right line ending, coding
system for input and output, etc, defined.
They would, I guess, default to either the local coding system
preferences or to 'undefined-*' with the right EOL conversion.
Thoughts on that?
Daniel
--
The real object of education is to have a man in
the condition of continually asking questions.
-- Mandell Creighton
