[Webkit-unassigned] [Bug 100626] Support semantics annotation-xml elements

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=100626





--- Comment #12 from David Carlisle <davidc at nag.co.uk>  2012-10-29 16:09:53 PST ---
(In reply to comment #11)
> (In reply to comment #9)
> > Not quite.
> > 
> > If the first child is a Content-MML element, it DOESN'T HAVE a default rendering. Instead, the UA is supposed come up with a rendering, based on the subsequent <annotation-xml> or <annotation> child(ren).
> 
> It seems to me that if a browser is able to render content MathML, then it is allowed to display this first child or choose among the annotation.


Yes certainly.

The usage with no base expression and just annotations is really an abuse of notation that was valid MathML2 because of loose schema enforcement and works in gecko because the spec, when intending to say the default behaviour is to render the "first child" wasn't really expecting a case where there is no base expression and only annotations. Since Gecko simply renders the first child it of course picks up the first annotation in this case.

The MathML3 schema (which is generally rather tighter than the mathml2 schema which was very loose, allowing many things such as a fraction with more than 2 children) makes this invalid.

The MathML3 Schema is:

semantics = element semantics {semantics.attributes,
                               MathExpression, 
                              (annotation|annotation-xml)*}

I became aware of this usage with only annotations somewhat later when I think some of Jacques' examples were given as examples of integration for html5.

I suppose if there is really a lot of usage of this construct a case could be made for a bug report changing the schema to   MathExpression? so making it optional but really I'm not sure it is good usage to be annotating nothing. It has the benefit of working in firefox but structurally, annotating nothing isn't really what is wanted here.

As far as I can see, none of the semantics elements in

http://www.w3.org/Math/Documents/Notes/graphics.xml#svg-in-mathml-guidelines

use a semantics element with only annotations, in fact it explitly says the forst (non annotation) child is mandatatory:

The first child of the semantics element is the MathML
    object whith[sic] which the additional information is associated, it can be
    a presentation- or content MathML representation. This child is
    mandatory in a semantics element, it can be seen as a fallback
    representation in MathML, if the MathML application processing the
    semantics element cannot process any of the annotation and
    annotation-xml children. In our case, we have chosen to represent the
    dice as a csymbol element, another choice would have been to give
    an alternate text represented in a mtext element.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.