- From: Philip Wadler <wadler@research.bell-labs.com>
- Date: Sat, 09 Dec 2000 20:09:48 -0500
- To: www-xml-schema-comments@w3.org
Volker Rennenberg, working under my guidance, completed an
implementation of the April 2000 version of Schema, written in Xduce.
He had the following observations about which parts of the
implementation were particularly difficult. I believe these all apply
to the CR version. I can solicit further details from Volker if there
are questions. -- P
- definition of the time datatype considering truncation is ambigous; part 2
- definition of restriction is ad hoc and complex; part I, 5.10 {particle valid(restriction)}
- assembling the content type of a type from its base type is complex; part I, 4.3.3 {content type}
- interaction between equivalence classes and import is complex.
- identity constraints were not implemented because of the work
- implementation of "all" groups is difficult because of free ordering and defaults
-----------------------------------------------------------------------
Philip Wadler wadler@avaya.com
Avaya Labs http://www.cs.bell-labs.com/~wadler
600 Mountain Ave, room 2T-402 office: +1 908 582 4004
Murray Hill, NJ 07974-0636 fax: +1 908 582 5857
USA home: +1 908 626 9252
-----------------------------------------------------------------------
Received on Saturday, 9 December 2000 20:10:04 UTC