W3C home > Mailing lists > Public > www-math@w3.org > October 2006

Re: MathML-in-HTML5

From: David Carlisle <davidc@nag.co.uk>
Date: Tue, 3 Oct 2006 23:33:08 +0100
Message-Id: <200610032233.k93MX8H1007412@edinburgh.nag.co.uk>
To: ian@hixie.ch
Cc: www-math@w3.org, dev-tech-mathml@lists.mozilla.org


Ian,
> What are the rules for handling non-well-formed content?
not sure what the "rules" are (as in whether they are published
anywhere), perhaps someone from DS (or Microsoft for that matter) could
give more information, but empirically what I think happens is that
if you register the namespce on m: with a component then any top level
element (<m:math>..</m:math> in our case) gets handed over to the
component, well formed or not, and then it's up to the component what it
does with it.  Mathplayer for example usually tries to make something
out of incorrect (including non well formed)  content, but always
renders it in a red error box.

>  (Could you show me an example of this?
yes.

Taking the example whose markup is shown and described here

http://www.dessci.com/en/products/mathplayer/author/creatingpages.htm#AnatomyMathPlayerWebPage

I've cut out the example and made it into a file served here:
http://www.dcarlisle.demon.co.uk/mml1.html

Here's the same file made gratuitously non-well formed (all end tags
made into start tags)

http://www.dcarlisle.demon.co.uk/mml2.html

It renders in red, and if you use the right menu to "copy mathml"
and paste the copied markup into a text editor you will see

<math>
  <!-- Error encountered and repaired: 
    Too few children in <msup> node
   -->
  <msup>
    <!-- Error encountered and repaired: 
      Too many children in <mi> node
     -->
    <mi>x 2 + 9 x + 9 = 0</mi>
    <mrow>
    </mrow>
  </msup>
</math>


So basically in the case of mathplayer any non well formed text is
displayed as an error, although I suspect that isn't enforced by the API
exposed by IE for XML fragments.

I should stress I'm just a user here I have no inside knowledge about
any of the components being discussed.

David
Received on Tuesday, 3 October 2006 22:33:29 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:12:59 GMT