Fw: [Dynapi-Dev] DynLayer and setClass()

[Kevin <ke...@ke...>]

Sorry didn't send reply to the list.

> 
> "Raymond Irving" <xw...@ya...> wrote:
> 
> > Hi,
> > 
> > 
> > When using setClass('whatever',true) you should always
> > create dynlayer with width and height or 0. This is to
> > prevent dynlayer from attempting to autosize the
> > layer. 

I disagree the w & h should be defaulted to null in the
constructor and not be touched until _assignElement.

> > It was suggested once that we get the width and the
> > height from the style of the layer, but that might not
> > be a good thing to do as style sheets can use
> > different values for width and height:
> > 
> > width:100px; height:100px;
> > width:100%; height:100%;
> > width:100cm; height:100cm;
 
A user/developer knowing that pixels are supported
would use px in style sheets as in the same way
lyr=new DynLayer('z',10,10,'100%',100%')
wouldn't be expected to work in any case.
 
> > If a layer's style width inside the style sheet was
> > set to 100% then the dynlayer would report the wrong
> > size, correct?
> > 
> > Any thoughts on this? Should we document that users
> > set the width and height to 0,0 when using style
> > sheets? Or should we enable DynLayer.assignElement()
> > to first check for a valid css width/height before
> > calling getContentWidth/Height? 
 
Defaults as explained above. Then in assignElement:
1) call new methods getElmWidth() & getElmHeight()
to satisfy the style sheet requirement
then if the user/developer hasn't specified a style object
or class
2) call getContentWidth() & getContentHeight() to satisfy
the auto size requirement (single character DynLayers etc)
 
-
Kevin.
 
> > 
> > --
> > Raymond Irving
> > 
> > __________________________________________________
> > Do you Yahoo!?
> > The New Yahoo! Search - Faster. Easier. Bingo
> > http://search.yahoo.com
> > 
> > 
> > -------------------------------------------------------
> > This sf.net email is sponsored by:ThinkGeek
> > Welcome to geek heaven.
> > http://thinkgeek.com/sf
> > _______________________________________________
> > Dynapi-Dev mailing list
> > Dyn...@li...
> > http://www.mail-archive.com/dyn...@li.../
>