[ISSUE-67] definition of regular expression for allowed characters from Yves Savourel on 2013-01-09 (public-multilingualweb-lt-comments@w3.org from January 2013)

From: Yves Savourel <ysavourel@enlaso.com>
Date: Wed, 9 Jan 2013 13:16:38 -0700
To: <public-multilingualweb-lt-comments@w3.org>
Message-ID: <assp.0721813d83.assp.072171dca2.001501cdeea6$3afd19e0$b0f74da0$@com>

Hi all,

To answer one question Christian had about whether or not we could be sure the proposed common sub-set would cover all possible use cases:

I thought about it more, and yes, it can be demonstrated: The sub-set allows the enumeration of all characters as well as ranges of characters. This means any character can be defined as allowed (or, because you can also negate the set, defined as not-allowed).

What we would not be able to do anymore would be to use shortcuts and groups to simplify the expression. That is certainly a drawback, but IMO it is less of a problem than using a syntax that is not interoperable between RE engines.

cheers,
-yves

Received on Wednesday, 9 January 2013 20:29:39 UTC