- From: Joe Kesselman <keshlam@alum.mit.edu>
- Date: Tue, 12 Jun 2001 14:21:22 -0400
- To: www-dom@w3.org
The DOM doesn't actually restrict the contents of most of its strings, though individual implementations may do so. However, the XML grammar _does_ restrict what characters may appear in an XML document. Numeric character references can't violate those rules. CDATASections don't change what's legal; they just provide an alternative to the numeric character referencs. If you need to represent characters which XML wouldn't permit, you have to implement an encoding scheme in your applications. The usual approach is to use a base-64 encoder/decoder, which is a fairly traditional solution to expressing arbitary binary data as character streams. Also remember that in some languages (such as C), the zero character is taken as end-of-string; you may find that you need to use a different string library to process the contents of this file.
Received on Tuesday, 12 June 2001 14:05:19 UTC