Variable typing

[Alex Tweedly]

Robert Sneidar wrote:

> The problem is, Revolution script is soooo friendly with variable  
> typing that just passing an undefined variable name as an arguement  
> to a function will simply pass the variable name as a literal to the  
> parameter of the called function, so within the function, no matter  
> what, the value of the parameter will be something. It would be nice  
> if Revolution did not do that, but I understand they need to be  
> compatible with prior xCard solutions.
>
If you set "explicitVariables" in the IDE, you will get errors for 
undefined variables (and if you declare the variable as a local within 
the handler, then it will be initialized to empty) - so avoiding some of 
the difficulties you have to deal with below. Of course, you still can't 
distinguish between a variable which is undefined and one which has been 
deliberately set to empty :-(

> Once inside the function, variableNames and globalNames knows nothing  
> of a local variable in the calling handler. And there is no way to  
> distinguish if the parameter passed is the value of an existing  
> variable, or the literal name of the variable I thought I was passing  
> it. Revolution is simply being too friendly. Passing the variable as  
> a reference doesn't help either. The parameter is still resolved to  
> the value of the passed variable regardless, (which begs the question  
> "what's the difference?") In short, there is no way for a called  
> function to tell if a variable belonging to the scope of the calling  
> script exists or not, and what it is. Until now.
>
Hmmmm - how about the case of :

on myHandler
  local myVar
  put "myVar" into myVar
  put "myVar" into varName
  put theType(myVar, varName) into myType
  -- function theType() returns a type of "U" which is wrong
end myHandler


-- 
Alex Tweedly       http://www.tweedly.net



-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.385 / Virus Database: 268.3.5/300 - Release Date: 03/04/2006