code blocks and C-like lambda expressions

[confuseme]

A while back, I started using Ruby for some personal software projects. I was immediately impressed with its "code block" syntax. It's a fairly natural Algol-like syntax for lambda expressions, and is used to great effect in the language.

Comments

[ch]

you might wish to check out Forsythe (ref: CMU CS Tech Report cmu-cs-88-159). dylan too. there has been lots of work on have higher order functions in algol-derived languages. the smalltalk approach is not the only way. (i personally prefer full closures over blocks.)

[confuseme]

Thanks for pointing out Forsythe, I hadn't paid any attention to it before!

I think Ruby's blocks are full closures, although I'm not super clear on the meaning of that phrase. But, for example:

irb(main):001:0> i = 1
=> 1
irb(main):002:0> p = Proc.new { i += 1 }
=> #

irb(main):003:0> p.call()
=> 2
irb(main):004:0> i
=> 2

[ch]

I'm not a Ruby expert, but I understood the Proc object as the closure and a block is just a block.

It seems roughly equivalent to the difference between lamba and progn.

[snej]

But you can convert a block to a Proc object with a single method call, so there isn't really much distinction between them. It's primarily syntactic sugar plus some optimization. (Smalltalk-80 implementations also tried to be lazy about converting block parameters into full-fledged objects, generally doing so only if the block got assigned to a non-local variable.)

[nescafe]

You might also want to check out http://www.iolanguage.com/about/