|
Maybe John Cowan and/or *you*, dear reader, can help me understand
the following quote from http://www.w3.org/TR/xml-c14n#Example-Chars
: ---- Note: The last element, normId, is well-formed
but
violates a validity constraint for attributes of type ID. For
testing
canonical XML implementations based on validating processors, remove
the
line containing this element from the input and canonical form. In
general,
XML consumers should be discouraged from using this feature of XML.---- Specifically, I'm wondering what is "this feature" that XML consumers should be discouraged from using, and why? Some possibilities have occurred to me: (0) Validating processors? (1) The validity constraint for values of attributes of type ID? (2) The ID attribute type? (3) The fact that whitespaces and/or apostrophes can be made to appear in attribute values? (4) Or is it just that, in general, people should be discouraged from doing philosophically aberrant things in XML-land, even if XML permits them? (But if that's the intended import, why say so in this particular Note?) ----------------------------------------------------------------------- FYI: The reason I got into the XML canonicalization document is that I was surprised to discover that Firefox dislikes carriage-returns and linefeeds between attribute specs in start tags in HTML, at least in that it can't trivially pre-discover certain info in certain <meta> elements so represented. I was relieved to see that, anyway, Firefox accepts any whitespace before the TAGC (tag-close, '>') character at the end of a start tag (or end tag, for that matter). I haven't made a systematic study of browsers on these points, thinking I'd start, anyway, with W3C's take on how things should be. And so I ran into this bit of ambiguity. |