Whoops.

Looks like a bug in Debug.objectOwnProperties.  Somehow it appears to be saying 
that there is a property named 'undefined' in the object named LzError, which 
is causing the inspect loop to blow up.

Not sure how to construct a test case.  But if you want to try to do that, I 
can look at the Debug code and see if I can figure out where things are going 
wrong...

On 2010-11-30, at 12:17, Henry Minsky wrote:

> Does this ring a bell?
> 
> [SWF]
> /trunk2/ibm/OLBacktrace/src/lzx/commerce/shell/restricted/ManagementCenter.lzx
> - 6,710 bytes after decompression
> [Fault] exception, information=ReferenceError: Error #1069: Property
> undefined not found on LzError and there is no default value.
> Fault, inspectInternal() at LzDebugService.as:1633
> 1633  val_$g = obj_$0[key_$l];
> 
> 
> 
> On Tue, Nov 30, 2010 at 12:10 PM, P T Withington 
> <[email protected]>wrote:
> 
>> You should be able to inspect the recursive error (the `#...`), which
>> hopefully points to where the internal error is.
>> 
>> Maybe I can move the try/catch "in" to have it not lose _everything_ when
>> it trips over an error.  I'll look at that.
>> 
>> IM me if you want.
>> 
>> On 2010-11-30, at 11:58, Henry Minsky wrote:
>> 
>>> So, where would I start to look for what's causing this error in printing
>>> out a backtrace
>>> 
>>> 
>>> lzx> Debug.inspect(«LzError(128)| @../../foundation/restricted/D…»)
>>> «LzError(128)#3| @../../foundation/restricted/DefinitionLoader.lzx#137:
>>> TypeError: Error #1010: A term is undefined and has no …» {
>>> Error: ReferenceError: Error #… computing inspectInternal}
>>> 
>>> 
>>> 
>>> 
>>> 
>>> --
>>> Henry Minsky
>>> Software Architect
>>> [email protected]
>> 
>> 
> 
> 
> -- 
> Henry Minsky
> Software Architect
> [email protected]


Reply via email to