From: "Mariusz Pękala" <skoot@...>
Date: 2007-06-21T18:54:09+09:00
Subject: Re: if == any item in an array

--gKMricLos+KVdGMg
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On 2007-06-21 09:17:24 +0900 (Thu, Jun), 12 34 wrote:
> > Give this code a shot. You almost had it -- the syntax you were
> > looking for was the leading splat "*" in front of the PhotoEndings and
> > MovieEndings in the branches of the case statement.
> >=20
> Thank you and all the others that answered. The array.include was what I=
=20
> was thinking too. But removing a couple of commas and adding a couple of=
=20
> astericks (splats) is nice. Ruby is great.
>=20
> I don't understand the meaning of the *. I imagine it comes from the=20
> wildcard usage. What do I look up in my books to understand this.

When the splat-solution appeared here an alarm bell rung in my head:
If the array is small, like in this problem, everything is OK. But what
if the array is huge?

I was wondering whether stack overflows could occur. I havent found
them, but the benchmark shows that .include?() is faster.


require 'benchmark'
LIMIT =3D 10_000_000
STEP =3D LIMIT / 10 + 1
BIG_ARRAY =3D (1..LIMIT).to_a; nil

def look_by_splash(item)
  case item
  when *BIG_ARRAY
    'found by splash'
  else
    'not found by splash'
  end
end

def look_by_include(item)
  case
  when BIG_ARRAY.include?(item)
    'found by include'
  else
    'not found by include'
  end
end

Benchmark.bmbm do |x|
  x.report('by_splash')  do
    1.step(LIMIT + STEP, STEP) { |i| look_by_splash(i) }
  end
  x.report('by_include') do
    1.step(LIMIT + STEP, STEP) { |i| look_by_include(i) }
  end
end

Rehearsal ---------------------------------------------
by_splash  13.080000   0.000000  13.080000 ( 13.212631)
by_include  6.750000   0.000000   6.750000 (  6.817161)
----------------------------------- total: 19.830000sec

                user     system      total        real
by_splash  12.930000   0.000000  12.930000 ( 13.060550)
by_include  6.750000   0.000000   6.750000 (  6.809274)



I tested with LIMIT =3D 100_000_000, and the proportions are the same.

--=20
Ceterum censeo Internet Explorer esse delendam.

--gKMricLos+KVdGMg
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7-ecc0.1.6 (GNU/Linux)

iD8DBQFGekrAsnU0scoWZKARAhOnAKDApJD6zJDnIXcDjrQu6eYhr+wNvwCcDHvY
sdEQycx8P27MYOOsQn9R54M=
=qmqg
-----END PGP SIGNATURE-----

--gKMricLos+KVdGMg--