From: Brian Candler <b.candler@...>
Date: 2009-08-20T22:49:23+09:00
Subject: Re: #has_arguments?

Robert Klemme wrote:
> I do not see the connection of this to what I posted.

What I mean is, when you have an argument with a default value, you may 
want to use the default value even when nil is explicitly passed in, 
e.g. m(nil)

This is usually because the function is being called from a higher-level 
function, which passes through a value from a higher level again, and 
doesn't want to be concerned with knowing what the default value is for 
the low-level function. e.g.

# Option 1 - not very DRY

def z(a = 123)
  puts "Handling #{a}"
end

def y(count = 3, a = 123)
  count.times { z(a) }
end

def x(count = 3, a = 123)
  y(count, a)
end

x()

# Option 2 - not always possible

def z(a = 123)
  puts "Handling #{a}"
end

def y(count = 3, *args)
  count.times { z(*args) }
end

def x(*args)
  y(*args)
end

x()

# Option 3 - not pretty

def z(a = 123)
  puts "Handling #{a}"
end

def y(count = 3, a = nil)
  count.times { a ? z(a) : z }
end

def x(count = nil, a = nil)
  count ? (a ? y(count,a) : y(count)) : y
end

x()

# Option 4

def z(a = nil)
  a ||= 123
  puts "Handling #{a}"
end

def y(count = nil, a = nil)
  count ||= 3
  count.times { z(a) }
end

def x(count = nil, a = nil)
  y(count, a)
end

x()

This last option is the one which was called "silly", but I think in 
this case it's actually quite a reasonable way to solve the problem. The 
defaults are specified in exactly one place, and passing nil means 'use 
the default'

Regards,

Brian.

-- 
Posted via http://www.ruby-forum.com/.