[Laszlo-dev] setters and LzDataElement

[Henry Minsky]

I'll  look at the devildog branch data classes you've worked on for
swf9, and see if I can make something in trunk
that is as close as possible. After that gets merged back up to
devildog, then we can try to merge the swf9 and swf8 versions.




On Mon, Mar 10, 2008 at 7:40 PM, Philip Romanik
<promanik at laszlosystems.com> wrote:
>
>  Hi Henry,
>
>  In the swf9 branch I kept the methods defined in the mixin classes, but
> moved where the setters are added to the real objects (LzDataElement,
> LzDataset, LzDataText). Are you planning on doing the same thing? This
> alleviates the mixin's from having to deal with the static setters list.
>
>  I don't think the extensible mechanism is needed to the data objects
> either. However, I'm still debugging to get the data code to work.
>
>  Thanks!
>
>  Phil
>
>
>
>
> As an optimization, I think we should not try to give LzDataElement the
> extensible setter mechanism that LzNode has.
>
>  There are a couple of setters defined in LzDataElement, which would look
> like
>
>  attributes = "setAttrs";
>  childNodes = "setChildNodes";
>  nodeName = "setNodeName";
>  and maybe
>  ownerDocument = "setOwnerDocument";
>
>  and LzDataText has a setter for
>  data = "setData";
>
>  I can't think of any case where the user would want to add setters to these
> classes, so I am thinking of just hardcoding these into a static class var
> setters list, or possibly even hardcoding them into a custom setAttribute
> method on LzDataElement.
>
>  All the other crud (earlySetters, DelayedSetters, getters, options,
> defaultattrs), are not needed for LzDataElement I think.
>
>  Classes like LzDatapointer which derive from LzNode will still get all that
> stuff. I'm just talking about LzDataElement and other LzMiniNode subclasses.
>
>  Thoughts?
>
>
>
>
>
>
>
>  --
>  Henry Minsky
>  Software Architect
>  hminsky at laszlosystems.com



-- 
Henry Minsky
Software Architect
hminsky at laszlosystems.com