Re[2]: [cgmo-webcgm] onload solution

[Lofton Henderson <lofton@rockynet.com>]

At 09:48 AM 12/20/2005 -0500, Benoit Bezaire wrote:
>I'm okay with Dieter's wording.

Same here, with one qualification (below)...


>Tuesday, December 20, 2005, 4:28:33 AM, you wrote:
>
> > Working through my pile of emails...
>
> > see below
>
> >> -----Original Message-----
> >> From: Lofton Henderson [mailto:lofton@rockynet.com]
> >> Sent: Thursday, December 15, 2005 9:06 PM
> >> To: Benoit Bezaire; cgmo-webcgm@lists.oasis-open.org
> >> Subject: Re: [cgmo-webcgm] onload solution
> >>
> >> Hi Benoit,
> >>
> >> Looks good.  One question/suggestion...
> >>
> >>
> >> At 04:29 PM 12/14/2005 -0500, Benoit Bezaire wrote:
> >> >Hi all,
> >> >
> >> >   As agreed from today's telecon, here is the proposal for the
> >> >   'onload' issue. The solution only addresses the loading of WebCGM
> >> >   files via the 'object' tag. If other group members want to have
> >> >   'onload' event handling not using the 'object' tag, they will have
> >> >   to make their own proposal.
> >> >
> >> >   The spec wording of section 3.4 should change from:
> >> >   "The user supplies the name of an event handler that the viewer
> >> >   shall invoke upon the 'onload' event. This param element
> >> and handler
> >> >   allows for script writers to manipulate the WebCGM DOM at
> >> the point
> >> >   at which the user agent has fully parsed the object tag and its
> >> >   descendants and is ready to render the object to the screen."
> >> >
> >> >   to:
> >> >   "The user supplies the name of an event handler that the viewer
> >> >   shall invoke upon the @@'onload' event. Note: subsequent
> >> >   modification of the user agent's source file using the DOM will
> >> >   cause the 'onload' event to be dispatched."
> >>
> >> Do you think it is clear to all readers, what is meant by
> >> "subsequent modification of the user agent's source file
> >> using the DOM"?  E.g., some readers who haven't participated
> >> in the TC might take this to mean manipulation of the DOM tree.
> >>
> >> How about being very specific:  "Note:  Using DOM 'src'
> >> attribute (on the WebCGMMetafile interface) to define or
> >> redefine the user agent's WebCGM source file will cause the
> >> 'onload' event to be dispatched."
> > This is not sufficient in my opinion. Suggestion:
>
> > "Note: The 'load' event will be dispatched every time a CGM file will be
> > loaded by the user agent. This may be the result of a DOM operation or
> > the traversal of a link inside the CGM file."

And of course it may also be a result of the initial execution of the 
<object> with a 'data' attribute, which is where this all started (and 
where this modified wording will be placed).  So it if "may be the result 
of" is supposed to be inclusive, then perhaps we should append, ", in 
addition to initial specification of the metafile by the 'data' attribute 
on the 'object' element."

-Lofton.


> > Rationale:
> > A linkURI stored inside the CGM using "_replace" might also cause a
> > different
> > CGM file to be loaded.
>
> > Regards,
> > Dieter
>
> >>
> >>
> >> >   and the 'onload' event definition (5.7.10) should be changed from:
> >> >   "The load event occurs when the WebCGM DOM implementation finishes
> >> >   loading all content within a WebCGM metafile."
> >>
> >> I just noticed.  In 5.7.10, the event is named 'load', not
> >> 'onload' (and its companion is 'unload', not 'onunload').  It
> >> has always been so.  I guess it's a little late to change
> >> it's name, right?
> >>
> >>
> >> >   to:
> >> >   "The load event occurs when the user agent has fully loaded all
> >> >   content within a WebCGM metafile and is ready to render the
> >> >   WebCGM metafile to the screen."
> >>
> >> Regards,
> >> -Lofton.