Revolution debugger issues

[Robert Brenstein]

>I can't answer your question directly, however, one of the things you wrote
>struck a chord:
>
>>  Because I can't debug it, I can't figure out where the problem is.
>
>You may recall, I wrote to Rev Support a few weeks ago vaguely describing a
>problem where garbage characters were intermittently winding up in global
>variables (that would otherwise evaluate correctly in the message box and
>simple scripts).  This did not involve the debugger specifically, but as in
>your case, it was something that took me days if not weeks to track down.
>And I still have no explanation for why it happens.
>

This won't help solve Jacque's problem but I have a year-old bug 
entry in bugzilla for a bug which involves bizzaire misbehavior with 
variables. I wonder whether there is a relation. My error started 
occuring only when the script reached certain level of complexity (as 
I kept adding additional functionality) and exhibits itself by 
finding non-existing word in a variable. I debugged this by writing 
values to text file (not using debugger) getting something like

   found word 17 of 15

where 17 is returned by wordOffset() and 15 is the number of words. I 
have reproduced this behavior in a few different versions of the 
engine on two platforms, so it is quite persistent. Changing the 
script, though, shifts the error: it is reported for different 
database records (the records processed were always the same) which 
implies things being messed up in memory somehow.

Robert