workaround for exec err dialog?

[Richard Gaskin]

Wouter wrote:

>     workaround for exec err dialog?
>     Richard Gaskin ambassador at fourthworld.com
>     Mon May 3 11:33:38 EDT 2004
> 
>     I may have found a workaround for the engine issue in which the error
>     data param sent with errorDialog contains erroneous information:
> 
>     I've found that for each erroneous entry the line and column items
>     (#s 2
>     and 3 respectively) contain 0 (zero).
> 
>     Can you think of any valid circumstances in which that would be the
>     case?
> 
> 
> Yes by test.
> 
> I don't think it's a good idea to take out all lines where item 2 and 3 
> are zero.
> A faulty testhandler placed in a button and on stacklevel can show this:
> 
> on mouseup ### opencard
> repeat with i = 1 to 100 step 2
> put i into foo[i]
> end repeat
> set the innards of me to foo
> end mouseup ### opencard
> 
> The errordialog handler got 1 extra line of code to put its parameter.
> 
> the which parameter contains (when the reported error is correct):
> openstack called from a button:
> 620,0,0
> 90,5,19
> 449,5,3
> 535,5,1
> 241,5,1,openstack
> 353,0,0,stack "/Volumes/MacOS9/Dev 
> Folder/Teststacks/metacard_error_tester.mc"
> 573,2,1,openstack
> 241,2,1,mouseUp
> 353,0,0,button id 1007 of card id 1002 of stack "/Volumes/MacOS9/Dev 
> Folder/Teststacks/metacard_error_tester.mc"
> 
> mouseup in a button:
> 620,0,0
> 90,5,19
> 449,5,3
> 535,5,1
> 241,5,1,mouseUp
> 353,0,0,button id 1009 of card id 1002 of stack "/Volumes/MacOS9/Dev 
> Folder/Teststacks/metacard_error_tester.mc"
> 
> Line 620 of field "Messages" of stack "Execution Error" indicates a 
> "Array: bad index expression"
> Line 353 of field "Messages" provides "Object Name:"
> 
> If these lines are taken out by:
> 
>     ###!
>     if item 2 of line i of ee = 0 AND item 3 of line i of ee = 0 then
>     next repeat
>     ###!
> 
> 
> then the error dialog displayed won't be complete

I've gone further in my revision to only excluding those lines where ite 
ms 2 and 3 are zero AND where the error message corrresponding to item 1 
does not start with "Object Name:".

It's interesting that lines of code causing errors with arrays are not 
noted in the error data. That should be easily accounted for as well.

Any others?

Unfortunately I don't have code from the Exec Err dialog to post for 
this, as I gave up on it and wrote my own (which I needed to do for 
client deliverables anyway; the current one is too ugly to include in 
standalones <g>).

-- 
  Richard Gaskin
  Fourth World Media Corporation
  ___________________________________________________________
  Ambassador at FourthWorld.com       http://www.FourthWorld.com