apache-spark 0.9.1: New formula

Apache Spark is a fast and general engine for large-scale data
processing.

Closes Homebrew/legacy-homebrew#29534.

Signed-off-by: Adam Vandenberg <flangy@gmail.com>

Formula/apache-spark.rb

@@ -0,0 +1,19 @@
+require "formula"
+
+class ApacheSpark < Formula
+  homepage "http://spark.apache.org/"
+  head "https://github.com/apache/spark.git"
+  url "http://d3kbcqa49mib13.cloudfront.net/spark-0.9.1-bin-hadoop2.tgz"
+  version "0.9.1"
+  sha1 "953cd042a8cc44d297af46cc6714e30d20554c4b"
+
+  def install
+    rm_f Dir["bin/*.cmd"]
+    libexec.install Dir["*"]
+    bin.write_exec_script Dir["#{libexec}/bin/*"]
+  end
+
+  test do
+    system "#{bin}/spark-shell <<<'sc.parallelize(1 to 1000).count()'"
+  end
+end