Drag and Drop

[Richard Gaskin]

Thierry wrote:

> Le 8 févr. 2011 à 15:56, Richard Gaskin a écrit :
>
>> The only problem I can find with this approach is that it "snaps back" the dragImage to the source, even though the allowDrop has been set to true.
>>
>>  if anyone here knows a reliable way to prevent that I'm be most grateful.
>
> Hi Richard,
>
> I add one more  line in your mousedown  script.
> See below
>
> Please, let me know if it works.
>
> Regards,
> Thierry
>
>
> on mouseDown
>   put the long id of me into tObj
>   export snapshot from tObj to tVar as PNG
>   set the width of img 1 to the width of tObj
>   set the height of img 1 to the height of tOBj
>   set the visible of img 1 to false
>   put tVar into img 1
> end mouseDown


Unfortunately setting the image visibility to false has no effect. I had 
it visible only because it's a demo and it's fun to watch the image 
change; I have production code with such images false and still get the 
snap-back.

RunRev's Ben Beaumont offered this tip on the dev list which works a 
treat, also noting that this bug will be fixed in v4.6:

---------------------------

    Firstly, we've tracked down the problem and fixed it for
    the next release. In the meantime and for those using
    older version of LC here is a workaround:

    on dragEnd
       remove script of btn "ObjectDragFrontScript" from front
       put gObjectDragInfoA["stack"] into tDestStack
       send "doDragEnd tDestStack" to me in 0 millisecs
    end dragEnd

    on doDragEnd tDestStack
       if tDestStack <> empty then
         put gObjectDragInfoA["loc"] into tLoc
         set the defaultStack to tDestStack
         set the properties of the templateButton to \
            the properties of me
         lock screen
         create btn
         set the loc of last control to tLoc
         select last control
       end if
     end doDragEnd

    The problem occurs because creating a button and setting the
    loc causes (at some point) the dragAction global property to
    be reset, which is then what the engine is returning to the
    OS - hence the 'drag didn't succeed' snap-back.

    The engine should really return the value of 'dragAction' at the
    point the drop occured (before any messages are sent), rather
    than after the drop had been processed. This is the underlying bug.

----------------------------------

Thanks, Ben!

--
  Richard Gaskin
  Fourth World
  LiveCode training and consulting: http://www.fourthworld.com
  Webzine for LiveCode developers: http://www.LiveCodeJournal.com
  LiveCode Journal blog: http://LiveCodejournal.com/blog.irv