- From: <jpederse@wiley.com>
- Date: Wed, 6 Aug 2003 11:03:07 -0400
- To: David Carlisle <davidc@nag.co.uk>
- Cc: www-math@w3.org
David, > If you start an entity > definition with a combining character than a text base unicode > normaliser will attempt to combine the character with the markup rather > than with the intended character. (and will actually succeed in the case > of a combining / and a >, which would make the file non-well formed. > So for MathML we've tried as far as possible to avoid combining > characters, > see: > http://www.w3.org/2003/entities/iso9573-2003doc/9573sec6.html#sec6.2 Thanks for the instruction on this: I hadn't previously realized about the parsing ambiguity problem with the combining diacrticial marks. A bit of a shame, really, considering all of the nice symbols they have in those blocks... At the end of the document you point to, it lists two combining characters only (tdot and DotDot). Isn't DownBreve (̑) also in this category (no alternative, and needs a space before it)? Thanks again. John. ----------------------------- John Pedersen Content Systems, John Wiley & Sons, Inc.
Received on Wednesday, 6 August 2003 11:02:49 UTC