[ruby-talk:5488] RE: Some newbye question

Davide wrote such intriguing questions to be a real newbie :)

> Which is the rationale of the fact that block variables are 
> non-binding if some variable with the same name already 
> exists ? Isn't it confusing?

I think I'm reading your question wrongly, but it seems you have a mistake
here. I guess a block variable binds if there exists a variable of same name
in outer scope. 

And it *is* confusing sometimes. For me the only bothersome case was when
there wasn't a variable, and I tried to access a variable local to a block.
Like:

  def foo
    loop {
      a = "initialize"
      break
    }
    puts a
  end
  foo

But even that was easy afterall. I'm sure I'll hit really hard runtime
weirdness in the future, but so far so good.

> What is supposed to do the following fragment of code?
> 
> a = 1 ; x = 1 ; f = proc {|a| a += 1}
> f.call(x)

I have to say your following example makes me dizzy, but for this I can give
my explanation, even though it might be completely wrong.

When we call the method call of the proc object referenced by f, we pass the
object referenced by x as an argument. That object is number 1. In the block
the first thing is to assign arguments to the parameter list, so 'a' starts
to refer to object 1. Then we send a message '+' to object passed object 1
with and argument 1, and get a reference to a object 2, which is assigned to
variable a again. Then we fade out from a proc, and return from the call,
continuing from the next line after your example.

In my understanding your example shouldn't touch the "top-level" variable
'a' at all, as the proc has a variable 'a' already in scope before "a+=1" -
namely the 'a' in the proc's parameter list.

So I'm really puzzled why this example of yours actually changes the value
of the top-level variable 'a' based on if it's passed to the block or not.
Maybe the clever guys will enlight us.

> a = 1 ; x = 1 ; f = proc {|a| a += 1}
> f.call(x)
> f.call(a)
> a # results in 2
> 
> a = 1 ; x = 1 ; f = proc {|a| a += 1}
> f.call(a)
> f.call(x)
> a # results in 3

And what it comes to your last question

> Are the following statement equivalent?
> 
> p() if (b1 and b2)
> p() if (b1 && b2)
> 
> If so, why are operators `and', `or' needed? 

I have to say there are differences, and I'd like to have both, while maybe
both are not actually needed.

Their precedence is different, and 'and' is lower, thus:

  ruby -v -e "def foo; true; end; p foo and false; p foo && false;"
  ruby 1.6.2 (2000-10-11) [i386-cygwin]
  true
  false

The first p is actually 

   p(foo) and false;  # evaluating to (nil and false) which is nil

The second

   p( foo && false ); # evaluating to nil

Usually you don't have to think about the difference so much. It's quite
much the same as you think when to put extra parentheses to calculations.
Nevertheless, use extra parentheses when in doubt.

My personal opinion is that 'and' is much more nicer looking, and should be
used most of the time.

	- Aleksi