Revolution debugger issues

[J. Landman Gay]

Update on the debugger thing.

I walked through my wonky stack with someone else on the RR team and 
they were able to reproduce some of the problems. The problems, in my 
case, boil down to this:

Breakpoints do not always break
Handlers/functions called from other handlers do not always run
Variables can contain inaccurate values in certain specific cases

He was able to see the same problems, but oddly enough, they occured at 
different places in the script even though he and I were running through 
the same debugging procedures in tandem.

The good news is that they are on it, and hopefully can solve it. I have 
learned that a sequence of two errors in a script do different things in 
MC and Rev. In Rev, the second one causes the debugger to abort; in MC 
it does not. That is why in MC I get a recursive error dialog that 
requires a force-quit. Rev is better at solving this and does exit 
before that happens. However, the result of the exit is that sometimes 
handlers with errors will continue to run to the end using incorrect 
data; sometimes they just skip over things, or indulge in other odd 
behaviors.

Anyway, they're looking at it. Fingers crossed.


-- 
Jacqueline Landman Gay         |     jacque at hyperactivesw.com
HyperActive Software           |     http://www.hyperactivesw.com