To come back to the initial point, I found out what in dragevent.js, onmousedown was 
making my form elements acting strangely in IE when I was clicking in the layer they 
are in, layer which is dragable. It's the onselectstart. I moved it to the onmousemove 
handler, just
before we fire the drag start event. I think it makes more sens, but It fixed my 
issues anyway !

Benoit Marchant

Joachim Lundgren wrote:

> At 2001-07-12 00:13, you wrote:
> >Well it looks like your right, but there are a few issues, firstly, this is
> >only for NS4, the other browser code seems to be missing all together.
> >I think the whole thing, crossbrowser should look like this:
> >
> >This replaces the line nr 147 in mouse.js:
> >
> >
> >
> >  // Click on links and form elements
> >        if (e && e.target.handleEvent && e.target!=this)evt.browserReturn =
> >e.target.handleEvent(e);
> >    }else{
> >     if(e&&e.srcElement!=this)evt.browserReturn =e.returnValue
> >
> >
> >
> >The closing bracket is already in the code.
> >I tested this, and it solves the problem in IE5.5, Mozilla, and NS4.
> >
> >It does however cause NS4 to fire two onclick events over a regular
> >dynlayer, I think this might be something Jordi could solve easily (?)
> >
> >Cheers,
> >Richard Bennett
>
> For some reason I thought the problem only appeared in NS4...
>
> Now that this is being considered for inclusion in the codebase I suggest that the 
>end of that function to be (indented so it looks better in email):
>
>         // Click on links and form elements
>         if(e && e.target.handleEvent && e.target!=this) {
>             evt.browserReturn = (e.target.handleEvent(e) == false)
>                                 ? false
>                                 : evt.browserReturn;
>         }
>     }
>     else {
>         if(e && e.srcElement!=this)
>             evt.browserReturn = e.returnValue;
>     }
>
>     return evt.browserReturn;
> }
>
> The reason would be to allow the (in this case) onclick-handler to cancel the click 
>with a "return false". If it instead returned "true" or nothing at all it should 
>respect the evt.browserReturn value which is true by default unless changed by 
>evt.cancelBrowserEvent().
>
> This is, of course, untested. Haven't even checked what should be done in IE/Mozilla.
>
> /Lunna
>
> _______________________________________________
> Dynapi-Help mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/lists/listinfo/dynapi-help


_______________________________________________
Dynapi-Help mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/dynapi-help

Reply via email to