ietf-nntp Simplification of wildmat

[Clive D.W. Feather]

Maurizio Codogno said:
> I prefer to avoid the sets, ant therefore I vote for version 5b.
> The reason is that I prefer a simpler grammar: I shudder when I see
>    [^a-dx-p]  all one character strings except "a", "b", "c", and "d"

That's an implication of the "z-a means nothing" rule.

> ISO 10646/UTF8 collating is a mess, IMO, and I don't want to try it
> on a wildmat.

We collate in strict numerical order. UTF8 collates simply and obviously
when you do that (that is, strcmp() will produce the correct result).

> Besides, I don't see examples where sets are actually needed: 
> in practice, it should always be possible to unfold the set and 
> obtain a not-too-long wildmat.

How do you unfold "alt.[^x]*" into something not-too-long ?

-- 
Clive D.W. Feather  | Work:  <clive at demon.net>   | Tel:  +44 20 8371 1138
Internet Expert     | Home:  <clive at davros.org>  | Fax:  +44 20 8371 1037
Demon Internet      | WWW: http://www.davros.org | DFax: +44 20 8371 4037
Thus plc            |                            | Mobile: +44 7973 377646