Re: mobileOK spec unclear on nested objects parsing

OK I see, sounds like a good proposal.

There is still quite the open question of what we do with these
changes -- I assume that anything except editorial and clarifications
has to be saved for the next revision.

On Jan 24, 2008 11:34 AM, Jo Rabin <jrabin@mtld.mobi> wrote:
>
>
> I wrote it the way it was - though it could be clearer, I agree, to take
> account of, say:
>
> <object>
>         <div>
>                 <p>Here is an alternative
>                         <object>
>                                 <div>
>                                         <p>but that one did not work
> either
>                                                 <object> ...
>                                                 </object>
>                                         </p>
>                                 </div>
>                         </object>
>                         <object>
>                                 Etc.
>                         </object>
>                 </p>
>         </div>
> </object>
>
>
>
> > -----Original Message-----
> > From: Sean Owen [mailto:srowen@google.com]
> > Sent: 24 January 2008 16:27
> > To: Jo Rabin
> > Cc: Dominique Hazael-Massieux
> > Subject: Re: mobileOK spec unclear on nested objects parsing
> >
>
> > ISSUE-230: we need to come up with a resolution on this to propose.
> >
> > On Jan 18, 2008 10:19 AM, Jo Rabin <jrabin@mtld.mobi> wrote:
> > > For each object element that has no object element ancestor
> > >         Request the object (ingoring the type attribute)
> > >         Apply the Object Processing Rule
> > >
> > > Object Processing Rule
> > >
> > > If the content type of the retrieved object is not image/jpeg or
> > image/gif
> > >         If it is empty, warn
> > >         If it consists only of white space, FAIL
> > >         For each of its descendant object elements that is not a
> > descendant of another descendant object element, apply the object
> > processing rule
> >
> > OK by me though I might state the last line as:
> >
> > For each direct child object element, apply the object processing
> rule.
> >
> > How about this then as the proposed change?
>
>