How do I convert a Block to a Proc in a Ruby 1.9 C extension?

Question

I'm writing a Ruby 1.9 C extension and I want to do the following in ruby:

notifier = Notifier.new
notifier.on 'click' do
  puts "clicked!"
end

Now the problem with this is that on the C method, I only "receive" a block, and, as far as I know, it's not even a parameter: I just can call with with rb_yield.

So my question is: is there a way on a Ruby 1.9 C extension, to transform a block into a proc or something, so I can store it inside my module, and call it later whenever I want/need them? Like an async callback!

I already implemented this with Procs/lambdas, but it's just ugly not to use the block syntax directly.

Eslinger

Have you seen this article (especially "explicit blocks" paragraph)? It might be outdated but otherwise it looks like what you need. – Culbertson

Sorry, I can't answer your question, because I don't know C nor the YARV C API, but as a clarification to other readers, your question basically is: "how do I do def on(&blk) end from C", right? – Jukebox

Answer 1

In the Ruby C source you'll see this in proc.c:

/*
 * call-seq:
 *   proc   { |...| block }  -> a_proc
 *
 * Equivalent to <code>Proc.new</code>.
 */

VALUE
rb_block_proc(void)
{
    return proc_new(rb_cProc, FALSE);
}

and Proc.new does this:

Creates a new Proc object, bound to the current context. Proc::new may be called without a block only within a method with an attached block, in which case that block is converted to the Proc object.

So you'd do something like this:

VALUE p = rb_block_proc();
/* and then store `p` somewhere convenient */

and then later on, to call the block/Proc:

rb_funcall(p, rb_intern("call"), 0);

That rb_funcall is pretty much the C version of p.send(:call).

Boutis

@rubenfonseca: Simple once you know what to do :) You need to have a grepping-familiarity of the C source to write C extensions. – Boutis