[Laszlo-dev] [Laszlo-user] core/LzNode errors breaks debug=false version of swf9 app
Henry Minsky
henry.minsky at gmail.com
Tue Apr 6 07:25:10 PDT 2010
Hmm, I wonder if there is anything better that the compiler or runtime can
do to warn about
constraints that cannot work.
The runtime error you were seeing was a clue, maybe we can make a beefed up
check in the
constraint execution code in debug mode to try and help flag cases where
the constraint
is ill-formed.
On Tue, Apr 6, 2010 at 10:10 AM, Roger Yarrow <ryarrow at gmail.com> wrote:
> Thanks everyone for the prompt replies. With lots of commenting out
> sections and trial/error, I found the problematic constraint. I had a
> constraint inside a class that was based upon an array. The array is defined
> in the class such as this:
>
> <attribute name="range_array" type="expression" value="$once{new
> Array()}"/>
>
> The constraint was like this:
>
> <view visible="${classroot.range_array.length>1}">
>
> This was the point of failure. To resolve, I made an attribute that held
> the current value of the array's length. I then modified the constraint to
> use this attribute value instead, and it works! Problem solved!
>
>
>
> On Tue, Apr 6, 2010 at 1:33 AM, Max Carlson <max at openlaszlo.org> wrote:
>
>> Wow - congratulations on the port!
>>
>> I'd suggest running in debug mode with backtrace enabled. Then, you can
>> click on the errors and see exactly what's causing them. If it's something
>> in our code you can click the bug report button and use that to submit a
>> report to http://jira.openlaszlo.org/ so we can get it fixed for you.
>>
>> Let me know what you find!
>>
>> Regards,
>> Max Carlson
>> OpenLaszlo.org
>>
>>
>> On 4/5/10 8:33 PM, Roger Yarrow wrote:
>>
>>> Hi,
>>>
>>> I have a large app (18000+ lines) that I migrated from 4.1.1 swf8 to
>>> 4.7.1 swf9. I have it all working (finally!) with debug=true. When
>>> debug=true, the application will compile and execute but I get about a
>>> handful of errors like this:
>>>
>>> ERROR @core/LzNode.lzs˜1518: TypeError: Error #1034: Type Coercion
>>> failed: cannot convert []@2c2b5821 to LzEventable.
>>> ERROR @core/LzNode.lzs˜1518: TypeError: Error #1034: Type Coercion
>>> failed: cannot convert []@b297f59 to LzEventable.
>>> ERROR @core/LzNode.lzs˜1518: TypeError: Error #1034: Type Coercion
>>> failed: cannot convert []@c91ec11 to LzEventable.
>>> ERROR @core/LzNode.lzs˜1518: TypeError: Error #1034: Type Coercion
>>> failed: cannot convert []@a6610b1 to LzEventable.
>>>
>>> The errors are during initialization as I have lots of views/classes
>>> being instantiated in a variety of ways.
>>>
>>> When I attempt to deploy the app using debug=false, it will not fully
>>> execute. Adobe Flash (debugger version) pops up with errors such as:
>>>
>>> TypeError: Error #1034: Type Coercion failed: cannot convert []@1d6a2661
>>> to .LzEventable.
>>> at LzNode/applyConstraintMethod()
>>> at LzNode/applyConstraintExpr()
>>> at LzNode/__LZresolveReferences()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZclearPreventInit()
>>> at LzState/apply()
>>> at LzState/$lzc$set_applied()
>>> at LzEventable/setAttribute()
>>> at MethodInfo-2268()
>>> at LzNode/applyConstraintMethod()
>>> at LzNode/applyConstraintExpr()
>>> at LzNode/__LZresolveReferences()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZinstantiationDone()
>>> at LzView/__LZinstantiationDone()
>>> at LzInstantiatorService/makeSomeViews()
>>> at LzInstantiatorService/checkQ()
>>> at Function/http://adobe.com/AS3/2006/builtin::call()<http://adobe.com/AS3/2006/builtin::call%28%29>
>>> at LzEvent/sendEvent()
>>> at LzIdleKernel$/__update()
>>>
>>> And:
>>>
>>> TypeError: Error #1034: Type Coercion failed: cannot convert []@1fc27079
>>> to .LzEventable.
>>> at LzNode/applyConstraintMethod()
>>> at LzNode/applyConstraintExpr()
>>> at LzNode/__LZresolveReferences()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZclearPreventInit()
>>> at LzState/apply()
>>> at LzState/$lzc$set_applied()
>>> at LzEventable/setAttribute()
>>> at MethodInfo-2286()
>>> at LzNode/applyConstraintMethod()
>>> at LzNode/applyConstraintExpr()
>>> at LzNode/__LZresolveReferences()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZcallInit()
>>> at LzNode/__LZinstantiationDone()
>>> at LzView/__LZinstantiationDone()
>>> at LzInstantiatorService/makeSomeViews()
>>> at LzInstantiatorService/createImmediate()
>>> at LzNode/createChildren()
>>> at LzNode()
>>> at LzView()
>>> at $lzc$class_generic_widget()
>>> at $lzc$class_countdown_timer()
>>> at $lzc$class_m3107/reload()
>>> at $lzc$class_m3100/show_desktop()
>>> at $lzc$class_m1974/clearMessage()
>>> at Function/http://adobe.com/AS3/2006/builtin::call()<http://adobe.com/AS3/2006/builtin::call%28%29>
>>> at LzDelegate/execute()
>>> at MethodInfo-437()
>>> at Function/http://adobe.com/AS3/2006/builtin::apply()<http://adobe.com/AS3/2006/builtin::apply%28%29>
>>> at LzTimeKernelClass/timerHandler()
>>> at flash.utils::Timer/_timerDispatch()
>>> at flash.utils::Timer/tick()
>>>
>>> The nature of the errors (Error #1034) are the same so I'm assuming they
>>> are one in the same. When debug=true, Flash can "handle" the error and
>>> allows my app to execute normally. When debug=false, it cannot handle
>>> the error and the app dies during instantiation and executes but only
>>> "halfway" alive.
>>>
>>> How do I figure out where the problem is? I looked in core/LzNode and
>>> the problem area is related to applyConstraintMethod, so I'm guessing I
>>> have a constraint going haywire somewhere. Perhaps a constraint is
>>> based upon a view not yet instantiated? How do I "dig deeper" to find
>>> out where the issue is?
>>>
>>> I am desperate here... if I have to, I can just deploy using a
>>> debug=true and hide the debug window. Obviously, I'd rather not do that.
>>>
>>> Thanks!!
>>>
>>> - Roger
>>>
>>
>
--
Henry Minsky
Software Architect
hminsky at laszlosystems.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.openlaszlo.org/pipermail/laszlo-dev/attachments/20100406/e5244c42/attachment.html
More information about the Laszlo-dev
mailing list