install cascalog, hadoop, etc. all in one go. a few minutes and we have a testable, queryable, hadoop instance in place.

simple_steps_to_cascalog.txt

simple steps to clojure cascalog

is ~/bin in your $PATH

at term type: echo $PATH

if not there add this to .bash_profile:

export PATH=$HOME/bin:$PATH

then:

cd ~/bin

then:

wget https://raw.github.com/technomancy/leiningen/stable/bin/lein

then:

chmod +x lein

then:

lein


will autoinstall clojure and all deps asssuming you have Java jdk in path


then:

cd ~/

then:

lein new [project_name]


then:

cd project_name

then:

vi project.clj


then:

add this to project.clj


:repositories {"conjars" "http://conjars.org/repo"}
:dependencies [[cascalog/cascalog-core "2.0.0"][org.clojure/clojure
"1.5.1"]]
:profiles { :dev {:dependencies [[org.apache.hadoop/hadoop-core "1.1.2"]]}})


you will overwrite only original deps definition

then:

lein repl

and it should install cascalog for you and start the repl!

at the repl you get a user prompt

try this:

user=> (use 'cascalog.playground) (bootstrap)

then try your first query:

user=> (?<- (stdout) [?person] (age ?person 25))

the data will be in playground.clj

next we will add testing!