Closed GoogleCodeExporter closed 9 years ago
Well, conceptually that makes no sense at all.
Ids are used for unique HTML elements, while with class you are able to target
multiple elements. And SWFObject only deals with unique instances: 1 piece of
alternative content (HTML element) maps to 1 SWF (HTML object element).
Original comment by bobbyvandersluis
on 4 Oct 2008 at 9:52
I think I wasn't clear enough about how this is supposed to work: The class name
would only be used by the main function too find all elements which have this
class
name. These objects would then be added to regAttrObj, like the the objeccts
that
have been added with register_object(). Basically this is like doing
register_object() for all objects that have a ceraatin class name but without
the
need to know the ids of these objects (they don't even necessarily need to have
one).
Original comment by Spotligh...@gmail.com
on 4 Oct 2008 at 10:03
Ok, static publishing only, I get it :-) I will add it to the enhancement list.
Original comment by bobbyvandersluis
on 4 Oct 2008 at 10:31
Original comment by bobbyvandersluis
on 4 Oct 2008 at 10:31
Is there any progress on this?
I need this for a WordPress plugin, since there's no easy access to the content
of the posts on a given
WordPress page at the time of creation of the page header.
That is, it would be much easier if SWFObject could just find all flash objects
with a given classname on its
own instead of me looping through all WordPress posts, extracting ids,
registering them with SWFObject when
the header of the page is created only to loop through all posts again later
on.
The easy solution is to give each static flash object an ID, loop through every
object, check the classname, get
the ID and register it with SWFObject, but ideally SWFObject shouldn't need an
ID at all, just a DOM-node.
Is there any potential danger of just sticking DOM-nodes into regObjArr instead
of ids?
Original comment by b...@benderydt.com
on 8 Dec 2008 at 4:22
Although I sincerely like the idea of this method, it didn't make the selection
of
SWFObject 2.2 alpha, and to be honest, I don't think we will include this in the
future either. The main reason for this decision is that we have agreed to keep
the
minimized version of SWFObject under 10Kb and that this feature would cost us at
least an additional Kb, while it is a feature that is only requested by a few
and
priority-wise at the bottom of the requested features list.
Original comment by bobbyvandersluis
on 5 Jan 2009 at 4:47
Would you be interested in an experimental patch which tries to get rid of ID's
in swfObject as much as possible?
I.e. refactor the internal code so it works on DOM nodes instead? I could try
to knock something together next
week, but only if has a slight chance at success. So please stop me before I
jump into the abyss...
Original comment by b...@benderydt.com
on 13 Jan 2009 at 11:32
We are always interested in new ideas, especially if they are good, and if we
can do
things in a smarter way, we will. However I can tell you beforehand that we are
unlikely to change the current SWFObject API. We will also unlikely add
features that
require a lot of additional code and only will be used by a few. Also, please
keep in
mind that in IE the HTML object element (and its nested elements) is only
partially
implemented in the DOM, so you have to keep in mind additional built-in
behavior and
accessibility issues.
Original comment by bobbyvandersluis
on 14 Jan 2009 at 10:08
Basically I'd like to replace createSWF and removeSWF by createSWFNode and
removeSWFNode and let the old
createSWF and removeSWF functions call the new one after their initial
getElementById. Since both functions
start with getElementById I see no immediate trouble (famous last words), aside
from the fact that createSWF
stores ID's in objIdArray for cleanup. I will have to change this to an array
of references to nodes. Do you see
any trouble in that?
I'll also have to change some calls of setVisibility to setVisibilityNode.
(Same system, aside from the dynamic
CSS trick).
I know that I'm assuming that any calls to createSWFNode, removeSWFNode and
setVisibilityNode happen after
the dom is loaded, but a) embedSWF, the major defined API, takes care of that
and b) the documentation for
createSWF uses it in an onDomLoad event handler and c) I can't see the point of
removeSWF before the DOM
has loaded...
Original comment by b...@benderydt.com
on 14 Jan 2009 at 8:19
Original issue reported on code.google.com by
Spotligh...@gmail.com
on 30 Sep 2008 at 7:35