Trying to explain the difference in how blocks "bind" to methods in ruby

gistfile1.rb

def foo *args
  p args
  puts "foo: " + block_given?.to_s
  return "foo"
end

def bar
  puts "bar: " + block_given?.to_s
  return "bar"
end

foo do
end
# no args
# block given

foo {}
# no args
# foo got block

foo bar {}
# args: "bar"
# foo didn't get block
# bar got block

foo bar do
end
# args: "bar"
# foo got block
# bar didn't get block

# Basically {} binds to the immediate method before it,
# do;end takes the easy option out and assumes the preceeding
# method is an argument.

# So if we rewrite the last two methods with brackets as
# ruby interprets it:

# Bar gets the block and 
# foo gets all of that as an argument.
foo( bar {})

# Foo gets both the block and bar as an argument.
foo( bar ) do
end