transliteration in Python

[Martin v. Loewis]

"Frederick H. Bartlett" <fbartlet at optonline.net> writes:

> But transliteration isn't what encodings do. 

Atleast not the way that they are normally understood. In Python 2.2,
"base64" and "rot13" are accessible as encodings, as well, so one
could argue that a transliteration is also some kind of encoding.

But you are probably right that Giorgi's problem most likely is a
misunderstanding of the term "encoding".

> My experience with encoding issues comes from TeX, where I used to
> design my own encodings in .tfm files so that I could type Classical
> Greek, Russian, and Georgian in American ascii with reasonable clarity.
> (Clear to me, anyway.) But in TeX one can use ligatures to accomplish
> neat encoding effects; no other encoding works that way.

Actually, Unicode has a number of pre-combined ligatures (e.g. U+0132,
U+04A4, U+04D4, U+FCB7, etc); in addition, you have the joiner and
non-joiner characters.

Regards,
Martin