Re: BibTeX-mode: Key generation when latin-1 characters appear in author field

[Roland Winkler]

On Sat Apr 16 2005 Richard Stallman wrote:
> I think there is something in Emacs already to do conversions
> like this, but I don't recall what.  It might be in enriched mode,
> or it might be in international/iso-*.el.

There is international/iso-cvt.el which translates (La)TeX and HTML
files to ISO 8859-1 and back. However, I think this package does not
work beyond ISO 8859-1. At least, it would be necessary to provide
the appropriate conversion tables. Furthermore, it seems to me that
this package is not actively maintained. I just discovered that in
CVS emacs the docstring of the autoloaded command iso-iso2duden is
definitely wrong, as it was already in GNU emacs 20.6.1. Is anybody
using this package?

In the context of BibTeX keys, the problem is a bit more subtle
because ideally, one would like to translate (La)TeX (or unicode or
ISO 8859) umlaute to (sequences of) plain ascii characters, for
example, `ß' -> `ss' and `ä' -> `ae' (but depending on the context
some people might prefer `ä' -> `a'). I do not know whether such
rules are established for other languages besides German. Here,
international/iso-cvt.el contains only the conversion table
iso-iso2duden-trans-tab for German umlaute, but not for any other
languages -- unless I overlooked them.

Roland