Lists Home |
Date Index |
- From: Kohsuke KAWAGUCHI <email@example.com>
- To: Eddie Robertsson <firstname.lastname@example.org>
- Date: Thu, 17 May 2001 18:04:15 -0700
> On a side issue: Why does this constaint on restriction apply to <xsd:all> and
> <xsd:choice> groups? It's logical for <xsd:sequence> since the sequence is
> important but in <xsd:choice> and <xsd:all> the order is insignificant.
I guess there is a computational difficulty issue involved. Otherwise
it's just an error somehow survived the reviews.
It becomes very hard to check that a given <all> is a proper subset of
another <all> if you allow arbitrary order of items. If <all> has x items,
then you have to check x^2 times in the worst case.