sequel-jdbc-hive2-adapter 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 1fa83dab077f1190d708a412b83b2f6e08cec8d6
+  data.tar.gz: 8cf8c718583586cc214f4d31e72f57a3c207798b
+SHA512:
+  metadata.gz: 1738dc25532f26927556aa78fe357a0d182ec621241ae2dd41830ba24ddfcedae8269d77ddc1faebc88b8912f0ffa3efc4de366dfbe46344b1e886a0839dfdb6
+  data.tar.gz: 8c0afdd6fb59a037f6b66d441a7adbdd78fbe43dbafa7c5334d81333eccbc339ec58afc5721173e032dafdb91fd2eb76e6405c0ac2f4556374ef9c33d77e7664

data/.gitignore

@@ -0,0 +1,2 @@
+Gemfile.lock
+spec/database.yml

data/.ruby-version

@@ -0,0 +1 @@
+jruby-1.7.9

data/Gemfile

@@ -0,0 +1,3 @@
+source "http://rubygems.org"
+
+gemspec

data/README.md

@@ -0,0 +1,4 @@
+* sequel-jdbc-hive2-adapter
+
+A sequel adapter for querying Hive/Hive2 and Impala clusters.
+JRuby is required when using this adapter. It will not work in any other Ruby interpreter.

data/Rakefile

@@ -0,0 +1,6 @@
+require "bundler/gem_tasks"
+require 'rspec/core/rake_task'
+
+RSpec::Core::RakeTask.new(:spec)
+
+task :default => :spec

data/lib/sequel-jdbc-hive2-adapter.rb

@@ -0,0 +1,12 @@
+# encoding: utf-8
+require "sequel-jdbc-hive2-adapter/version"
+
+# module Sequel
+#   module Jdbc
+#     module Hive2
+#       module Adapter
+#         # Your code goes here...
+#       end
+#     end
+#   end
+# end

data/lib/sequel-jdbc-hive2-adapter/version.rb

@@ -0,0 +1,9 @@
+module Sequel
+  module Jdbc
+    module Hive2
+      module Adapter
+        VERSION = "0.0.1"
+      end
+    end
+  end
+end

data/lib/sequel/adapters/jdbc/hive2.rb

@@ -0,0 +1,11 @@
+# encoding: utf-8
+
+module Sequel
+  module JDBC
+    module Hive2
+      module DatabaseMethods
+        # Add in hive specific database methods
+      end
+    end
+  end
+end

data/lib/sequel/jdbc_hive2_adapter.rb

@@ -0,0 +1,29 @@
+# encoding: utf-8
+require 'sequel'
+require 'sequel/adapters/jdbc'
+
+Sequel::JDBC::DATABASE_SETUP[:hive] = proc do |db|
+  require 'jdbc/hive2'
+  Jdbc::Hive2.load_driver
+
+  org.apache.hadoop.hive.jdbc.HiveDriver
+
+  require 'sequel/adapters/jdbc/hive2'
+  # db.extend(Sequel::JDBC::Postgres::DatabaseMethods)
+  # db.dataset_class = Sequel::JDBC::Postgres::Dataset
+
+  org.apache.hadoop.hive.jdbc.HiveDriver
+end
+
+Sequel::JDBC::DATABASE_SETUP[:hive2] = proc do |db|
+  require 'jdbc/hive2'
+  Jdbc::Hive2.load_driver
+
+  org.apache.hive.jdbc.HiveDriver
+
+  require 'sequel/adapters/jdbc/hive2'
+  # db.extend(Sequel::JDBC::Postgres::DatabaseMethods)
+  # db.dataset_class = Sequel::JDBC::Postgres::Dataset
+
+  org.apache.hive.jdbc.HiveDriver
+end

data/sequel-jdbc-hive2-adapter.gemspec

@@ -0,0 +1,22 @@
+Gem::Specification.new do |s|
+  s.name        = 'sequel-jdbc-hive2-adapter'
+  s.version     = '0.0.1'
+  s.authors     = ['Josh Ferguson']
+  s.email       = %w(josh@modeanalytics.com)
+  s.homepage    = 'https://github.com/mode/sequel-hive2-adapter'
+  s.summary     = %q{Sequel adapter for Hive/Impala}
+  s.description = %q{Use this adapter to use sequel to query Hive and Impala clusters}
+
+  s.rubyforge_project = 'sequel-jdbc-hive2-adapter'
+
+  s.files         = `git ls-files`.split("\n")
+  s.test_files    = `git ls-files -- {test,spec,features}/*`.split("\n")
+  s.executables   = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }
+  s.require_paths = %w(lib)
+
+  s.add_runtime_dependency 'jdbc-hive2'
+
+  s.add_development_dependency 'rake'
+  s.add_development_dependency 'rspec'
+  s.add_development_dependency 'sequel'
+end

data/spec/cb_clean.csv

@@ -0,0 +1,10 @@
+start_time,type
+12/31/10 23:49,Casual
+12/31/10 23:37,Registered
+12/31/10 23:27,Registered
+12/31/10 23:21,Registered
+12/31/10 23:20,Casual
+12/31/10 23:18,Registered
+12/31/10 23:18,Registered
+12/31/10 23:17,Casual
+12/31/10 23:16,Casual

data/spec/database.template.yml

@@ -0,0 +1,7 @@
+adapter: jdbc:hive2
+host: ec2-ip.us-west-2.compute.amazonaws.com
+port: 10000
+ssl: false
+database: default
+username: test
+password: test

data/spec/jdbc_hive2_adapter_spec.rb

@@ -0,0 +1,21 @@
+require 'spec_helper'
+
+describe 'Sequel::JDBC::Hive2' do
+  let(:conn) { Sequel.connect(HiveConfig.connection_url) }
+
+  it 'should show tables' do
+    tables = []
+    conn.dataset.fetch_rows("SHOW TABLES") do |row|
+      tables << row[:tab_name]
+    end
+    tables.length.should be > 0
+  end
+
+  it 'should select rows from a table' do
+    rows = []
+    conn.dataset.fetch_rows("SELECT * FROM cb_clean LIMIT 100") do |row|
+      rows << row
+    end
+    rows.length.should be > 0
+  end
+end

data/spec/schema.sql

@@ -0,0 +1,2 @@
+CREATE TABLE cb_clean(start_time STRING, type STRING) ROW FORMAT DELIMITED FIELDS TERMINATED BY ',';
+LOAD DATA LOCAL INPATH 'cb_clean.csv' INTO TABLE cb_clean;

data/spec/spec_helper.rb

@@ -0,0 +1,45 @@
+require 'yaml'
+require 'sequel'
+require 'sequel/jdbc_hive2_adapter'
+
+class HiveConfig
+  class << self
+    def adapter
+      config['adapter']
+    end
+
+    def host
+      config['host']
+    end
+
+    def port
+      config['port']
+    end
+
+    def database
+      config['database']
+    end
+
+    def username
+      config['username']
+    end
+
+    def password
+      config['password']
+    end
+
+    def ssl
+      config['ssl'] == true ? true : false
+    end
+
+    def connection_url
+      "#{adapter}://#{host}:#{port}/#{database};user=#{username};password=#{password};ssl=#{ssl}"
+    end
+
+    private
+
+    def config
+      @config ||= YAML.load_file(File.join(File.dirname(__FILE__), 'database.yml'))
+    end
+  end
+end

metadata

@@ -0,0 +1,119 @@
+--- !ruby/object:Gem::Specification
+name: sequel-jdbc-hive2-adapter
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- Josh Ferguson
+autorequire:
+bindir: bin
+cert_chain: []
+date: 2014-04-02 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: jdbc-hive2
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  prerelease: false
+  type: :runtime
+- !ruby/object:Gem::Dependency
+  name: rake
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  prerelease: false
+  type: :development
+- !ruby/object:Gem::Dependency
+  name: rspec
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  prerelease: false
+  type: :development
+- !ruby/object:Gem::Dependency
+  name: sequel
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  prerelease: false
+  type: :development
+description: Use this adapter to use sequel to query Hive and Impala clusters
+email:
+- josh@modeanalytics.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- .gitignore
+- .ruby-version
+- Gemfile
+- README.md
+- Rakefile
+- lib/sequel-jdbc-hive2-adapter.rb
+- lib/sequel-jdbc-hive2-adapter/version.rb
+- lib/sequel/adapters/jdbc/hive2.rb
+- lib/sequel/jdbc_hive2_adapter.rb
+- sequel-jdbc-hive2-adapter.gemspec
+- spec/cb_clean.csv
+- spec/database.template.yml
+- spec/jdbc_hive2_adapter_spec.rb
+- spec/schema.sql
+- spec/spec_helper.rb
+homepage: https://github.com/mode/sequel-hive2-adapter
+licenses: []
+metadata: {}
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: sequel-jdbc-hive2-adapter
+rubygems_version: 2.1.9
+signing_key:
+specification_version: 4
+summary: Sequel adapter for Hive/Impala
+test_files:
+- spec/cb_clean.csv
+- spec/database.template.yml
+- spec/jdbc_hive2_adapter_spec.rb
+- spec/schema.sql
+- spec/spec_helper.rb