MerouaneBen

# $Id: Nielsen2012Python_case.py,v 1.2 2012/09/02 16:55:25 fn Exp $

# Define a url as a Python string (note we are only getting 100 documents)
url = "http://wikilit.referata.com/" + \
    "wiki/Special:Ask/" + \
    "-5B-5BCategory:Publications-5D-5D/" + \
    "-3FHas-20author%3DAuthor(s)/-3FYear/" + \
    "-3FPublished-20in/-3FAbstract/-3FHas-20topic%3DTopic(s)/" + \
    "-3FHas-20domain%3DDomain(s)/" +  \
    "format%3D-20csv/limit%3D-20100/offset%3D0"

tamizhgeek

# setup for the hadoop + spark env for airflow

remote_file "/home/airflow/spark.tgz" do
  source "/remote/spark/download/url"
  owner "airflow"
  group "airflow"
  mode  '0755'
  not_if { File.exists?("/home/airflow/spark.tgz") }
end