Re: Error recovery spec

James Clark scripsit:

> Before getting into the details of the algorithm, I think it's useful to
> start by considering what element relationships/properties it makes sense
> to use.

I'm not sure I agree with this philosophy, but I'm willing to try
running with it.

> My general feeling is the PossibleChild/NonPossibleChild (ExcludedChild?)

I like PermittedChild/ForbiddenChild as names.  I agree that it is
foundational; without even that, one can do little.

> Beyond this, things become less clear.  I can certainly see the use of
> PreferredParent in HTML, but how would I get this from a schema?  In some
> cases (where there is only one possible parent, eg html/head/title) I can
> infer it from the PossibleChild properties.

In TagSoup Schema Language, A is the PreferredParent of B iff B is defined
textually within the definition of A.  This also has the advantage that
it's impossible for there to be a loop in the PreferredParent graph,
which as I noted is essential to keep TagSoup from looping on bad input,
since it is impossible for two elements to be defined within each others'
definitions.  The PreferredParent of text has to be defined by other means.

-- 
John Cowan        http://ccil.org/~cowan   cowan@ccil.org
Lope de Vega: "It wonders me I can speak at all.  Some caitiff rogue
did rudely yerk me on the knob, wherefrom my wits yet wander."
An Englishman: "Ay, belike a filchman to the nab'll leave you
crank for a spell." --Harry Turtledove, Ruled Britannia

Received on Tuesday, 18 December 2012 10:21:27 UTC