Unicode program representation

[François Pinard]

fredrik at pythonware.com writes:

> what scares me is if we end up with multiple encodings in the
> same source file.

Oh, I agree with you that is not going to be usual, at least at first.

I foresee it might happen quite naturally in Python scripts generating
multilingual announcements, say.  Or more generally, in Python scripts
holding translation correspondences between languages, or anything in which
we have substantial amount of editing to do in one language followed by a
substantial amount in another language.  Emacs Mule/MULE is likely able to do
this, without resorting to UTF-8.  Unicode on-the-fly combining is not well
supported at the font rendering level (despite it is increasingly needed),
and Unicode raises other disputes, like for the Han unification it imposed.

As for the "simpler" cases like of a Vietnamese sentence quoting a German
proverb, it is likely in the long run to see a common encoding like UTF-8.

> (and no, it's not because we're developing a python development environment
> -- supporting this in our python editor would be fairly trivial ;-)

It would be?  You should address the matter then.  It is a worth goal.

-- 
François Pinard   http://www.iro.umontreal.ca/~pinard