Re: Emacs 22 branch created.

[Stefan Monnier <monnier@iro.umontreal.ca>]

>     The non-decoded part of the buffer should be in unibyte mode.
> Why does it have to be in unibyte mode?

I didn't say it has to, just that it should: for efficiency, for clarity,
etc...

> Decoding can be done in a multibyte buffer.

Yes, binary data stored in a multibyte buffer works, but has to be handled
with more care.  Experience shows it's more likely to lead to bugs which
ultimately result in things like \NNN escapes shown to the user instead of
accented chars.

>     And of course very inefficient when you'll constantly be editing a very
>     large rmail buffer.
> Not really, because the gap makes such operations efficient.

It just reduces the inefficiency.

> The other approach also needs peculiar changes in lower-level features
> to work right.  Various operations on the message buffer would have to
> operate on the file buffer as well.  These include set-buffer-file-name,
> rename-buffer, as well as saving.

Those other features can all be done at the UI-level, where they belong, not
at a low level.  At the very least, it's all done in elisp, without any need
to fiddle with C code.


        Stefan