Zugg Software :: View topic - Should Aliases be allowed to change the scope/context

Posted: Tue Nov 20, 2007 12:04 am

OK, I hope this is another easy question that we can get consensus on. Consider the following script:

Posted: Tue Nov 20, 2007 1:39 am

It's always been the case in the past that you need to be careful to use #class 0 after you've used #class to change the current class (unless you deliberately intend not to). I don't see why this should be an exception.

Posted: Tue Nov 20, 2007 1:49 am

Posted: Tue Nov 20, 2007 1:54 am

My gut reaction was that the scope should have remain unchanged, but after thinking about it more, scripts should do an automatic '#CLASS 0'. It's more in line with other programming languages and I wouldn't want other errant scripts changing my scope access. In general, I don't assume any context unless I explicitly set it within my scripts.

I did want to confirm that if I had this code

SubAdmin Joined: 18 Nov 2001 Posts: 5182

I think Tech has hit a good point. Rather then being an automatic #CLASS 0, it really needs to be something that is put into the call stack. That way at the start of a script it pushes the current module and class path, then at the end of a script it pops them. This will likely give the most consistent behavior.

Doing just #CLASS 0 leads to a possible situation where the a #CLASS command by the end user is getting lost or ignored.

Wizard Joined: 11 Jul 2002 Posts: 1265 Location: USA

OK, if I get this right.

I start out my alias script with a class scope change.

#class A
Run a few commands that I want to execute within this scope.
Call another alias that will run in its own scope and do its own thing.
Run more commands that I want executed within this scope.
#class 0

If an automatic #class 0 is placed at the end of an alias then I'm going to lose my scope for the line.

"Run more commands that I want executed within this scope."

I think that Vijilante's Idea of having a record of where the scope was when it wasn't closed is a better idea. But we need some sort of end game that draws the line on when the scope is set to 0.

Perhaps with the last } of the uppermost alias is called.

Wizard Joined: 11 Jul 2002 Posts: 1265 Location: USA

Maybe it would be cleaner to just state that if you call an alias/function within a #class block that you need to start a new #class because you have lost your scope. Confused

Posted: Tue Nov 20, 2007 5:39 pm

Sorry I got people sidetracked with the #CLASS 0 comment. What I really meant was what Vijilante already mentioned...that an alias call would push the current context and then pop it when it was done. Sorry I wasn't more specific.

Anyway, it sounds like I'm on the right track here and that nobody wants the context changed by an alias to effect the calling script.

GURU Joined: 10 Oct 2000 Posts: 873 Location: USA

Posted: Tue Nov 20, 2007 7:46 pm

I could see an alias moving an item between classes with a new command.

#MOVEITEM {ID} {NewClassName}