sorenmacbeth

package forma.tap;

import backtype.hadoop.pail.PailStructure;
import java.util.Collections;
import java.util.List;
import org.apache.thrift.TBase;
import org.apache.thrift.TDeserializer;
import org.apache.thrift.TException;
import org.apache.thrift.TSerializer;

sorenmacbeth

(defn hfs-report
  [path]
  "Loads the log data from an HDFS path into Hbase."
  (?<- (hbase-tap "urls" "?url-hash" "urls" "?url" "?crawl-date"
                  "?crawl-time" "?response-code" "?status" "?host")
       [?url-hash ?url ?crawl-date ?crawl-time ?response-code ?status ?host]
       ((hfs-textline path) ?text)
       (fetch-value-hash ?text :url  :> ?url-hash)
       (fetch-value ?text :url  :> ?url)
       (fetch-value ?text :crawl-date  :> ?crawl-date)

sorenmacbeth

# mac port installs bash_completion in /opt/local
if [ -f /opt/local/etc/bash_completion ]; then
  . /opt/local/etc/bash_completion
  # *
  export GIT_PS1_SHOWDIRTYSTATE=1 
  # $
  export GIT_PS1_SHOWSTASHSTATE=1 
  # %
  # export GIT_PS1_SHOWUNTRACKEDFILES=1 
  export PS1='\[\e[32m\]λ \w\[\e[36m\]$(__git_ps1 " (%s)") [$(~/.rvm/bin/rvm-prompt i v)]\[\e[0m\]\n\[\e[32m\]→\[\e[0m\] '

sorenmacbeth

(ns gist.globhfs
  (:import [cascading.tap GlobHfs]))

;; ### Bucket to Cluster
;;
;;; To get tuples back out of our directory structure on S3, we employ
;; Cascading's [GlobHFS] (http://goo.gl/1Vwdo) tap, along with an
;; interface tailored for datasets stored in the MODIS sinusoidal
;; projection. For details on the globbing syntax, see
;; [here](http://goo.gl/uIEzu).

sorenmacbeth

diff -Naur whirr-0.3.0-incubating/contrib/python/src/py/hadoop/cloud/cli.py whirr-0.3.0-incubating-backtype/contrib/python/src/py/hadoop/cloud/cli.py
--- whirr-0.3.0-incubating/contrib/python/src/py/hadoop/cloud/cli.py	2011-01-15 23:03:44.000000000 -0800
+++ whirr-0.3.0-incubating-backtype/contrib/python/src/py/hadoop/cloud/cli.py	2011-02-15 11:51:49.000000000 -0800
@@ -296,7 +296,7 @@
                          opt.get('user_data_file'),
                          opt.get('availability_zone'), opt.get('user_packages'),
                          opt.get('auto_shutdown'), opt.get('env'),
-                         opt.get('security_group'))
+                         opt.get('security_group'), opt.get('spot_price'))
     service.launch_master(template, config_dir, opt.get('client_cidr'))

sorenmacbeth

class TwitterUser
  def calculate_tunkrank(p=0.05)
    self.followers.inject(0.0) do |sum, follower|
      sum + ((1.0 + (p * follower.tunkrank_score)) / (1.0 + follower.num_friends))
    end
  end
end

sorenmacbeth

SELECT 1.0 + SUM((1.0 + #{p} * tunkrank_score) / (1.0 + num_friends)) AS tunkrank_score 
FROM twitter_users 
  INNER JOIN twitter_id_follows ON (twitter_users.twitter_id = twitter_id_follows.follower_twitter_id)
WHERE twitter_id_follows.user_twitter_id = #{twitter_id};

sorenmacbeth

from redis import Redis
import simplejson

class Resque(object):
    """Dirt simple Resque client in Python. Can be used to create jobs."""
    redis_server = 'localhost:6379'

    def __init__(self):
        host, port = self.redis_server.split(':')
        self.redis = Redis(host=host, port=int(port))