rbhive 0.2.2 → 0.2.3

data/lib/rbhive/connection.rb

@@ -37,6 +37,7 @@ module RBHive
       @client = ThriftHive::Client.new(@protocol)
       @logger = logger
       @logger.info("Connecting to #{server} on port #{port}")
+      @mutex = Mutex.new
     end
     
     def open
@@ -52,8 +53,7 @@ module RBHive
     end
     
     def execute(query)
-      @logger.info("Executing Hive Query: #{query}")
-      client.execute(query)
+      execute_safe(query)
     end
     
     def priority=(priority)
@@ -70,21 +70,31 @@ module RBHive
     end
     
     def fetch(query)
-      execute(query)
-      ResultSet.new(client.fetchAll, client.getSchema)
+      safe do
+        execute_unsafe(query)
+        rows = client.fetchAll
+        schema = SchemaDefinition.new(client.getSchema, rows.first)
+        ResultSet.new(rows, schema)
+      end
     end
     
-    def fetch_in_batch(query, batch_size=100)
-      execute(query)
-      schema = client.getSchema
-      until (next_batch = client.fetchN(batch_size)).empty?
-        yield ResultSet.new(next_batch, schema)
+    def fetch_in_batch(query, batch_size=1_000)
+      safe do
+        execute_unsafe(query)
+        until (next_batch = client.fetchN(batch_size)).empty?
+          schema ||= SchemaDefinition.new(client.getSchema, next_batch.first)
+          yield ResultSet.new(next_batch, schema)
+        end
       end
     end
     
     def first(query)
-      execute(query)
-      ResultSet.new([client.fetchOne], client.getSchema)
+      safe do
+        execute_unsafe(query)
+        row = client.fetchOne
+        schema = SchemaDefinition.new(client.getSchema, row)
+        ResultSet.new([row], schema).first
+      end
     end
     
     def create_table(schema)
@@ -107,5 +117,22 @@ module RBHive
     def method_missing(meth, *args)
       client.send(meth, *args)
     end
+    
+    private
+    
+    def execute_safe(query)
+      safe { execute_unsafe(query) }
+    end
+    
+    def execute_unsafe(query)
+      @logger.info("Executing Hive Query: #{query}")
+      client.execute(query)
+    end
+    
+    def safe
+      ret = nil
+      @mutex.synchronize { ret = yield }
+      ret
+    end
   end
 end

data/lib/rbhive/result_set.rb

@@ -1,7 +1,7 @@
 module RBHive
   class ResultSet < Array
-    def initialize(rows, schema=[])
-      @schema = SchemaDefinition.new(schema, rows.first)
+    def initialize(rows, schema)
+      @schema = schema
       super(rows.map {|r| @schema.coerce_row(r) })
     end
     
@@ -14,16 +14,16 @@ module RBHive
     end
     
     def to_csv(out_file=nil)
-      output(",", out_file)
+      to_seperated_output(",", out_file)
     end
     
     def to_tsv(out_file=nil)
-      output("\t", out_file)
+      to_seperated_output("\t", out_file)
     end
     
     private
     
-    def output(sep, out_file)
+    def to_seperated_output(sep, out_file)
       rows = self.map { |r| @schema.coerce_row_to_array(r).join(sep) }
       sv = rows.join("\n")
       return sv if out_file.nil?

data/lib/rbhive/schema_definition.rb

@@ -3,19 +3,19 @@ module RBHive
     attr_reader :schema
   
     TYPES = { 
-      :boolean => :to_s, 
-      :string => :to_s, 
-      :bigint => :to_i, 
-      :float => :to_f, 
-      :double => :to_f, 
-      :int => :to_i, 
-      :smallint => :to_i, 
-      :tinyint => :to_i
+      :boolean  => :to_s,
+      :string   => :to_s,
+      :bigint   => :to_i,
+      :float    => :to_f,
+      :double   => :to_f,
+      :int      => :to_i,
+      :smallint => :to_i,
+      :tinyint  => :to_i,
     }
   
     def initialize(schema, example_row)
       @schema = schema
-      @example_row = example_row.split("\t")
+      @example_row = example_row ? example_row.split("\t") : []
     end
   
     def column_names

metadata

@@ -1,34 +1,48 @@
---- !ruby/object:Gem::Specification
+--- !ruby/object:Gem::Specification 
 name: rbhive
-version: !ruby/object:Gem::Version
-  version: 0.2.2
+version: !ruby/object:Gem::Version 
+  hash: 17
   prerelease: 
+  segments: 
+  - 0
+  - 2
+  - 3
+  version: 0.2.3
 platform: ruby
-authors:
+authors: 
 - Forward Internet Group
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2010-12-07 00:00:00.000000000 +00:00
+
+date: 2010-12-07 00:00:00 +00:00
 default_executable: 
-dependencies:
-- !ruby/object:Gem::Dependency
+dependencies: 
+- !ruby/object:Gem::Dependency 
   name: thrift
-  requirement: &2160558040 !ruby/object:Gem::Requirement
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
-    - - ! '>='
-      - !ruby/object:Gem::Version
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        hash: 15
+        segments: 
+        - 0
+        - 4
+        - 0
         version: 0.4.0
   type: :runtime
-  prerelease: false
-  version_requirements: *2160558040
+  version_requirements: *id001
 description: Simple lib for executing Hive queries
 email: andy@forward.co.uk
 executables: []
+
 extensions: []
+
 extra_rdoc_files: []
-files:
+
+files: 
 - lib/rbhive.rb
 - lib/rbhive/connection.rb
 - lib/rbhive/table_schema.rb
@@ -52,26 +66,36 @@ files:
 has_rdoc: true
 homepage: http://github.com/trfficbroker/rbhive
 licenses: []
+
 post_install_message: 
 rdoc_options: []
-require_paths:
+
+require_paths: 
 - lib
-required_ruby_version: !ruby/object:Gem::Requirement
+required_ruby_version: !ruby/object:Gem::Requirement 
   none: false
-  requirements:
-  - - ! '>='
-    - !ruby/object:Gem::Version
-      version: '0'
-required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
+required_rubygems_version: !ruby/object:Gem::Requirement 
   none: false
-  requirements:
-  - - ! '>='
-    - !ruby/object:Gem::Version
-      version: '0'
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
 requirements: []
+
 rubyforge_project: 
 rubygems_version: 1.6.2
 signing_key: 
 specification_version: 3
 summary: Simple lib for executing Hive queries
 test_files: []
+