[Laszlo-dev] For Review: Change 20090508-bargull-QAO Summary: animator bug due to implicit coercion

[André Bargull]

I've created LPP-8178.


On 5/8/2009 9:22 PM, P T Withington wrote:
> Approved!
> 
> But maybe we also ought to file a feature request that if you don't give 
> an animator an attribute to animate, it shouldn't try to animate the 
> attribute named 'null'?
> 
> On 2009-05-08, at 14:24EDT, André Bargull wrote:
> 
>> Change 20090508-bargull-QAO by bargull at dell--p4--2-53 on 2009-05-08 
>> 20:09:33
>> in /home/Admin/src/svn/openlaszlo/trunk
>> for http://svn.openlaszlo.org/openlaszlo/trunk
>>
>> Summary: animator bug due to implicit coercion
>>
>> New Features:
>>
>> Bugs Fixed: LPP-8172 ('empty' animators do not run in swf9)
>>
>> Technical Reviewer: ptw
>> QA Reviewer: (pending)
>> Doc Reviewer: (pending)
>>
>> Documentation:
>>
>> Release Notes:
>>
>> Details:
>> Empty animators caused a subtle bug in swf9 because of implicit 
>> coercion in swf9:
>> In swf9, 'undefined' can only be used for the any-type ('*'), for any 
>> other type 'undefined' is coerced to 'null'. This is a problem if you 
>> intend to use strict typing, but then forget some casts. In this case, 
>> the "attribute"-property was casted to String in some methods, but in 
>> several methods it wasn't. And if you later use string concatenation 
>> (here: "e_" + attribute), you may get "e_null" (coerced) or 
>> "e_undefined" (not coerced).
>>
>>
>>
>> Tests:
>> testcase from bugreport
>>
>> Files:
>> M WEB-INF/lps/lfc/controllers/LaszloAnimation.lzs
>>
>> Changeset: 
>> http://svn.openlaszlo.org/openlaszlo/patches/20090508-bargull-QAO.tar
>>
> 
>