I'm building a small array dynamically and then picking the maximum value. I have problems with each formulation. Which choice is less evil?

conditional_array.rb

# This option has the most clarity, but I generally hate local variables in a
# method, preferring to use tap.
def calculated_foo
  available_foos = []
  available_foos << bar
  available_foos << baz.qux if baz
  available_foos.max
end

# It's a little less clear, unless you know what tap is, and `end.max` is
# painful to look at.
def calculated_foo
  [].tap do |available_foos|
    available_foos << bar
    available_foos << baz.qux if baz
  end.max
end

# This avoids the painful `end.max` for a more-common `}.max`, but goes against
# the culture of using do/end for multi-line blocks.
def calculated_foo
  [].tap { |available_foos|
    available_foos << bar
    available_foos << baz.qux if baz
  }.max
end

[bar, baz ? baz.qux : 0].max

Because my code is for the right ppl only, knowhadimean?

Oh, and I fail in the presence of negative numbers because I'm too cool to care.

But if you're one of those guys who actually want their code to be readable, then may I suggest something completely different:

def good_method_name_that_exposes_intent
  return bar unless baz
  [bar, baz.qux].max
end

function calculatedFoo() {
  return Math.max.apply(Math, [bar, baz ? baz.qux : 0])
}

module NullObj
  def self.qux;0;end
end

[bar, (baz || NullObj).qux].max

i personally would just pick one of your options and go (problem the first one with the local), but i figured i'd add something to the discussion.

[bar, baz ? baz.qux : nil].compact.max