activerecord-jdbc-adapter 0.9.1 → 0.9.2

data/History.txt

@@ -1,3 +1,36 @@
+== 0.9.2
+
+- The main, highly awaited fix for this release is a solution to the
+  rake db:create/db:drop issue. The main change is a new 'jdbc' rails
+  generator that should be run once to prepare a Rails application to
+  use JDBC. The upside of this generator is that you no longer will
+  need to alter database.yml for JDBC. See the README.txt for details.
+- Cleanup and reconnect if errors occur during begin/rollback
+  (Jean-Dominique Morani, Christian Seiler)
+- ACTIVERECORD_JDBC-1: Add #drop_database method for oracle (does the
+  same thing as recreate_database)
+- Sqlite3 and MSSQL fixes (Jean-Dominique Morani)
+- JRUBY-3512: Treat LONGVARCHAR as a CLOB for Mssql
+- JRUBY-3624: Upgrade Derby to 10.5.3.0 and add native limit/offset
+  support (Christopher Saunders)
+- JRUBY-3616: Fix postgres non-sequence primary keys (David Kellum)
+- JRUBY-3669: Fix Oracle case with unconfigured schema (Dan Powell)
+- Fixed quote_column_name of jdbc_oracle to accept numbers (Marcelo
+  Murad)
+- Fix for mysql tables with non standard primary keys such that the
+  schema dump is correct (Nick Zalabak)
+- MSSQL fixes from Mike Luu:
+  - add support for MSSQL uniqueidentifier datatype
+  - always quote strings using unicode identifier for MSSQL
+- Changes primary_key generation to use always instead of by default
+  for DB2 (Amos King)
+- Improves the SQLite adapter by fixing rename_column, change_column,
+  change_column_default, changing remove_column, and adding
+  remove_columns (Ryan Baumann)
+- More oracle love courtesy Ben Browning and Jens Himmelreich
+- JRUBY-3608: Add missing change_column_null method for postgres
+- JRUBY-3508: Fix quoting of integer and float columns
+
 == 0.9.1
 
 - We did a lot of internal cleanup this release in the hopes of

data/Manifest.txt

@@ -33,7 +33,9 @@ lib/jdbc_adapter/rake_tasks.rb
 lib/jdbc_adapter/tsql_helper.rb
 lib/jdbc_adapter/version.rb
 lib/jdbc_adapter.rb
+lib/pg.rb
 lib/jdbc_adapter/jdbc_adapter_internal.jar
+test/abstract_db_create.rb
 test/activerecord/connection_adapters/type_conversion_test.rb
 test/activerecord/connections/native_jdbc_mysql/connection.rb
 test/cachedb_simple_test.rb
@@ -58,6 +60,7 @@ test/derby_simple_test.rb
 test/generic_jdbc_connection_test.rb
 test/h2_simple_test.rb
 test/has_many_through.rb
+test/helper.rb
 test/hsqldb_simple_test.rb
 test/informix_simple_test.rb
 test/jdbc_adapter/jdbc_db2_test.rb
@@ -79,11 +82,16 @@ test/models/data_types.rb
 test/models/entry.rb
 test/models/mixed_case.rb
 test/models/reserved_word.rb
+test/models/validates_uniqueness_of_string.rb
 test/mssql_simple_test.rb
+test/mysql_db_create_test.rb
 test/mysql_multibyte_test.rb
+test/mysql_nonstandard_primary_key_test.rb
 test/mysql_simple_test.rb
 test/oracle_simple_test.rb
+test/postgres_db_create_test.rb
 test/postgres_mixed_case_test.rb
+test/postgres_nonseq_pkey_test.rb
 test/postgres_reserved_test.rb
 test/postgres_simple_test.rb
 test/simple.rb
@@ -94,7 +102,16 @@ src/java/jdbc_adapter/JdbcAdapterInternalService.java
 src/java/jdbc_adapter/JdbcConnectionFactory.java
 src/java/jdbc_adapter/JdbcDerbySpec.java
 src/java/jdbc_adapter/JdbcMySQLSpec.java
+src/java/jdbc_adapter/MssqlRubyJdbcConnection.java
 src/java/jdbc_adapter/PostgresRubyJdbcConnection.java
 src/java/jdbc_adapter/RubyJdbcConnection.java
 src/java/jdbc_adapter/SQLBlock.java
 src/java/jdbc_adapter/Sqlite3RubyJdbcConnection.java
+rakelib/compile.rake
+rakelib/package.rake
+rakelib/rails.rake
+rakelib/test.rake
+rails_generators/jdbc_generator.rb
+rails_generators/templates
+rails_generators/templates/jdbc.rake
+rails_generators/templates/jdbc.rb

data/README.txt

@@ -2,6 +2,13 @@ activerecord-jdbc-adapter is a database adapter for Rails' ActiveRecord
 component that can be used with JRuby[http://www.jruby.org/]. It allows use of
 virtually any JDBC-compliant database with your JRuby on Rails application.
 
+== Project Info
+
+* Mailing Lists: http://kenai.com/projects/activerecord-jdbc/lists
+* Issues: http://kenai.com/jira/browse/ACTIVERECORD_JDBC
+* Source: git://kenai.com/activerecord-jdbc~main 
+  	  git://github.com/nicksieger/activerecord-jdbc-adapter.git
+
 == Databases
 
 What's there, and what is not there:
@@ -49,13 +56,23 @@ To use activerecord-jdbc-adapter with JRuby on Rails:
   * derby (<tt>activerecord-jdbcderby-adapter</tt>)
   * hsqldb (<tt>activerecord-jdbchsqldb-adapter</tt>)
   * h2 (<tt>activerecord-jdbch2-adapter</tt>)
+  * sqlite3 (<tt>activerecord-jdbcsqlite3-adapter</tt>)
+
+2. Run the "jdbc" generator to prepare your Rails application for
+   JDBC.
 
-2. If you're using Rails 2.0, you may skip to the next step. For Rails prior to
-   version 2.0, you'll need to add one-time setup to your config/environment.rb
-   file in your Rails application. Add the following lines just before the
-   <code>Rails::Initializer</code>. (If you're using activerecord-jdbc-adapter
-   under the old gem name used in versions 0.5 and earlier (ActiveRecord-JDBC),
-   replace 'activerecord-jdbc-adapter' with 'ActiveRecord-JDBC' below.)
+    jruby script/generate jdbc
+    
+   The initializer and rake task files generated are guarded such that
+   they won't be loaded if you still run your application un C Ruby.
+
+   Legacy: If you're using Rails prior to version 2.0, you'll need to
+   add one-time setup to your config/environment.rb file in your Rails
+   application. Add the following lines just before the
+   <code>Rails::Initializer</code>. (If you're using
+   activerecord-jdbc-adapter under the old gem name used in versions
+   0.5 and earlier (ActiveRecord-JDBC), replace
+   'activerecord-jdbc-adapter' with 'ActiveRecord-JDBC' below.)
 
     if RUBY_PLATFORM =~ /java/
       require 'rubygems'
@@ -63,11 +80,11 @@ To use activerecord-jdbc-adapter with JRuby on Rails:
       require 'jdbc_adapter'
     end
 
-3. Configure your database.yml to use the <code>jdbc</code> adapter. For mysql,
-   postgres, derby, oracle, hsqldb, h2, and informix you can simply configure
-   the database in the normal Rails style. If you use one of the convenience
-   'activerecord-jdbcXXX-adapter' adapters, be sure and put a 'jdbc' prefix in
-   front of the databas adapter name as below.
+3. Configure your database.yml in the normal Rails style. 
+
+   Legacy configuration: If you use one of the convenience
+   'activerecord-jdbcXXX-adapter' adapters, you can still put a 'jdbc'
+   prefix in front of the databas adapter name as below.
 
     development:
       adapter: jdbcmysql
@@ -76,8 +93,8 @@ To use activerecord-jdbc-adapter with JRuby on Rails:
       hostname: localhost
       database: weblog_development
 
-For other databases, you'll need to know the database driver class and URL.
-Example:
+   For other databases, you'll need to know the database driver class
+   and URL. Example:
 
     development:
       adapter: jdbc
@@ -126,12 +143,23 @@ The source for activerecord-jdbc-adapter is available using git.
 
   git clone git://github.com/nicksieger/activerecord-jdbc-adapter.git
 
+== Feedback
+
+Please file bug reports at
+http://kenai.com/jira/browse/ACTIVERECORD_JDBC. If you're not sure if
+something's a bug, feel free to pre-report it on the mailing lists.
+
 == Running AR-JDBC's Tests
 
-Drivers for 4 open-source databases are included. Provided you have MySQL
+Drivers for 6 open-source databases are included. Provided you have MySQL
 installed, you can simply type <tt>jruby -S rake</tt> to run the tests. A
 database named <tt>weblog_development</tt> is needed beforehand with a
-connection user of "blog" and password empty.
+connection user of "blog" and an empty password.
+
+If you also have PostgreSQL available, those tests will be run if the
+`psql' executable can be found. Also ensure you have a database named
+<tt>weblog_development</tt> and a user named "blog" and an empty
+password.
 
 If you want rails logging enabled during these test runs you can edit 
 test/jdbc_common.rb and add the following line:
@@ -140,11 +168,11 @@ require 'db/logger'
 
 == Running AR Tests
 
-  # If you want to run MRI against Rails remember to re-export RUBYLIB to be
-  # empty (or whatever you normally have it set to when you are done).
-  export RUBYLIB=$(find $HOME/NetbeansProjects/activerecord-jdbc-adapter -name lib -type d | grep -v pkg | ruby -e 'puts $stdin.readlines.map{|l| l.chomp}.join(":")')
-  cd active_record_source_dir
-  jruby -S rake test_jdbcmysql (or specific adapter you want to test)
+To run the current AR-JDBC sources with ActiveRecord, just use the
+included "rails:test" task. Be sure to specify a driver and a path to
+the ActiveRecord sources.
+
+  jruby -S rake rails:test DRIVER=mysql RAILS=/path/activerecord_source_dir
 
 == Authors
 

data/Rakefile

@@ -1,177 +1,10 @@
-require 'rake'
 require 'rake/testtask'
+require 'rake/clean'
+CLEAN.include 'derby*', 'test.db.*','test/reports', 'test.sqlite3','lib/**/*.jar','manifest.mf', '*.log'
 
 task :default => [:java_compile, :test]
 
-def java_classpath_arg # myriad of ways to discover JRuby classpath
-  begin
-    cpath  = Java::java.lang.System.getProperty('java.class.path').split(File::PATH_SEPARATOR)
-    cpath += Java::java.lang.System.getProperty('sun.boot.class.path').split(File::PATH_SEPARATOR)
-    jruby_cpath = cpath.compact.join(File::PATH_SEPARATOR)
-  rescue => e
-  end
-  unless jruby_cpath
-    jruby_cpath = ENV['JRUBY_PARENT_CLASSPATH'] || ENV['JRUBY_HOME'] &&
-      FileList["#{ENV['JRUBY_HOME']}/lib/*.jar"].join(File::PATH_SEPARATOR)
-  end
-  jruby_cpath ? "-cp \"#{jruby_cpath}\"" : ""
-end
-
-desc "Compile the native Java code."
-task :java_compile do
-  pkg_classes = File.join(*%w(pkg classes))
-  jar_name = File.join(*%w(lib jdbc_adapter jdbc_adapter_internal.jar))
-  mkdir_p pkg_classes
-  sh "javac -target 1.5 -source 1.5 -d pkg/classes #{java_classpath_arg} #{FileList['src/java/**/*.java'].join(' ')}"
-  sh "jar cf #{jar_name} -C #{pkg_classes} ."
-end
-file "lib/jdbc_adapter/jdbc_adapter_internal.jar" => :java_compile
-
 task :filelist do
   puts FileList['pkg/**/*'].inspect
 end
 
-if defined?(JRUBY_VERSION)
-  # TODO: add more databases into the standard tests here.
-  task :test => [:test_mysql, :test_jdbc, :test_derby, :test_hsqldb, :test_h2, :test_sqlite3]
-else
-  task :test => [:test_mysql]
-end
-
-FileList['drivers/*'].each do |d|
-  next unless File.directory?(d)
-  driver = File.basename(d)
-  Rake::TestTask.new("test_#{driver}") do |t|
-    files = FileList["test/#{driver}*test.rb"]
-    if driver == "derby"
-      files << 'test/activerecord/connection_adapters/type_conversion_test.rb'
-    end
-    t.test_files = files
-    t.libs = []
-    if defined?(JRUBY_VERSION)
-      t.ruby_opts << "-rjdbc/#{driver}"
-      t.libs << "lib" << "#{d}/lib"
-      t.libs.push *FileList["adapters/#{driver}*/lib"]
-    end
-    t.libs << "test"
-    t.verbose = true
-  end
-end
-
-Rake::TestTask.new(:test_jdbc) do |t|
-  t.test_files = FileList['test/generic_jdbc_connection_test.rb', 'test/jndi_callbacks_test.rb']
-  t.libs << 'test' << 'drivers/mysql/lib'
-end
-
-Rake::TestTask.new(:test_jndi) do |t|
-  t.test_files = FileList['test/jndi_test.rb']
-  t.libs << 'test' << 'drivers/derby/lib'
-end
-
-task :test_postgresql => [:test_postgres]
-task :test_pgsql => [:test_postgres]
-
-# Ensure driver for these DBs is on your classpath
-%w(oracle db2 cachedb mssql informix).each do |d|
-  Rake::TestTask.new("test_#{d}") do |t|
-    t.test_files = FileList["test/#{d}_simple_test.rb"]
-    t.libs = []
-    t.libs << 'lib' if defined?(JRUBY_VERSION)
-    t.libs << 'test'
-  end
-end
-
-# Tests for JDBC adapters that don't require a database.
-Rake::TestTask.new(:test_jdbc_adapters) do | t |
-  t.test_files = FileList[ 'test/jdbc_adapter/jdbc_sybase_test.rb' ]
-  t.libs << 'test'
-end
-
-# Ensure that the jTDS driver is in your classpath before launching rake
-Rake::TestTask.new(:test_sybase_jtds) do |t|
-  t.test_files = FileList['test/sybase_jtds_simple_test.rb']
-  t.libs << 'test' 
-end
-
-# Ensure that the jConnect driver is in your classpath before launching rake
-Rake::TestTask.new(:test_sybase_jconnect) do |t|
-  t.test_files = FileList['test/sybase_jconnect_simple_test.rb']
-  t.libs << 'test' 
-end
-
-MANIFEST = FileList["History.txt", "Manifest.txt", "README.txt", 
-  "Rakefile", "LICENSE.txt", "lib/**/*.rb", "lib/jdbc_adapter/jdbc_adapter_internal.jar", "test/**/*.rb",
-   "lib/**/*.rake", "src/**/*.java"]
-
-file "Manifest.txt" => :manifest
-task :manifest do
-  File.open("Manifest.txt", "w") {|f| MANIFEST.each {|n| f << "#{n}\n"} }
-end
-Rake::Task['manifest'].invoke # Always regen manifest, so Hoe has up-to-date list of files
-
-require File.dirname(__FILE__) + "/lib/jdbc_adapter/version"
-begin
-  require 'hoe'
-  Hoe.new("activerecord-jdbc-adapter", JdbcAdapter::Version::VERSION) do |p|
-    p.rubyforge_name = "jruby-extras"
-    p.url = "http://jruby-extras.rubyforge.org/activerecord-jdbc-adapter"
-    p.author = "Nick Sieger, Ola Bini and JRuby contributors"
-    p.email = "nick@nicksieger.com, ola.bini@gmail.com"
-    p.summary = "JDBC adapter for ActiveRecord, for use within JRuby on Rails."
-    p.changes = p.paragraphs_of('History.txt', 0..1).join("\n\n")
-    p.description = p.paragraphs_of('README.txt', 0...1).join("\n\n")
-  end.spec.dependencies.delete_if { |dep| dep.name == "hoe" }
-rescue LoadError
-  puts "You really need Hoe installed to be able to package this gem"
-rescue => e
-  puts "ignoring error while loading hoe: #{e.to_s}"
-end
-
-def rake(*args)
-  ruby "-S", "rake", *args
-end
-
-%w(test package install_gem release clean).each do |task|
-  desc "Run rake #{task} on all available adapters and drivers"
-  task "all:#{task}" => task
-end
-
-(Dir["drivers/*/Rakefile"] + Dir["adapters/*/Rakefile"]).each do |rakefile|
-  dir = File.dirname(rakefile)
-  prefix = dir.sub(%r{/}, ':')
-  tasks = %w(package install_gem debug_gem clean)
-  tasks << "test" if File.directory?(File.join(dir, "test"))
-  tasks.each do |task|
-    desc "Run rake #{task} on #{dir}"
-    task "#{prefix}:#{task}" do
-      Dir.chdir(dir) do
-        rake task
-      end
-    end
-    task "#{File.dirname(dir)}:#{task}" => "#{prefix}:#{task}"
-    task "all:#{task}" => "#{prefix}:#{task}"
-  end
-  desc "Run rake release on #{dir}"
-  task "#{prefix}:release" do
-    Dir.chdir(dir) do
-      version = nil
-      if dir =~ /adapters/
-        version = ENV['VERSION']
-      else
-        Dir["lib/**/*.rb"].each do |file|
-          version ||= File.open(file) {|f| f.read =~ /VERSION = "([^"]+)"/ && $1}
-        end
-      end
-      rake "release", "VERSION=#{version}"
-    end
-  end
-  # Only release adapters synchronously with main release. Drivers are versioned
-  # according to their JDBC driver versions.
-  if dir =~ /adapters/
-    task "adapters:release" => "#{prefix}:release"
-    task "all:release" => "#{prefix}:release"
-  end
-end
-
-require 'rake/clean'
-CLEAN.include 'derby*', 'test.db.*','test/reports', 'test.sqlite3','lib/**/*.jar','manifest.mf', '*.log'

data/lib/active_record/connection_adapters/jdbc_adapter.rb

@@ -1,14 +1,10 @@
+require 'active_record/version'
 require 'active_record/connection_adapters/abstract_adapter'
 require 'java'
 require 'active_record/connection_adapters/jdbc_adapter_spec'
 require 'jdbc_adapter/jdbc_adapter_internal'
 require 'bigdecimal'
 
-begin
-  require 'jdbc_adapter/rake_tasks'
-rescue LoadError
-end if defined?(RAILS_ROOT)
-
 # AR's 2.2 version of this method is sufficient, but we need it for
 # older versions
 if ActiveRecord::VERSION::MAJOR <= 2 && ActiveRecord::VERSION::MINOR < 2
@@ -618,6 +614,16 @@ module ActiveRecord
         @connection.write_large_object(*args)
       end
 
+      def pk_and_sequence_for(table)
+        result_set = @connection.connection.get_meta_data.get_primary_keys(nil, nil, table)
+        if result_set.next
+          keys = [result_set.getString("COLUMN_NAME"), nil]
+        end
+        keys.blank? ? nil : keys
+      ensure
+        result_set.close
+      end
+
       private
       def select(sql, name=nil)
         execute(sql,name)

data/lib/active_record/connection_adapters/jdbc_adapter_spec.rb

@@ -1,3 +1,15 @@
+module JdbcSpec
+  module ActiveRecordExtensions
+    def self.add_method_to_remove_from_ar_base(meth)
+      @methods ||= []
+      @methods << meth
+    end
+
+    def self.extended(klass)
+      (@methods || []).each {|m| (class << klass; self; end).instance_eval { remove_method(m) rescue nil } }
+    end
+  end
+end
 
 require 'jdbc_adapter/jdbc_mimer'
 require 'jdbc_adapter/jdbc_hsqldb'

data/lib/jdbc_adapter.rb

@@ -1,4 +1,4 @@
-if RUBY_PLATFORM =~ /java/
+if defined?(JRUBY_VERSION)
   begin
     tried_gem ||= false
     require 'active_record/version'

data/lib/jdbc_adapter/jdbc.rake

@@ -6,7 +6,7 @@ def redefine_task(*args, &block)
       public :instance_variable_set
       attr_reader :actions
     end
-    existing_task.instance_variable_set "@prerequisites", FileList[]    
+    existing_task.instance_variable_set "@prerequisites", FileList[]
     existing_task.actions.shift
     enhancements = existing_task.actions
     existing_task.instance_variable_set "@actions", []
@@ -16,46 +16,59 @@ def redefine_task(*args, &block)
 end
 
 namespace :db do
-  if Rake::Task["db:create"]
-    redefine_task :create => :environment do
-      create_database(ActiveRecord::Base.configurations[RAILS_ENV])
+  redefine_task :create => :environment do
+    create_database(ActiveRecord::Base.configurations[RAILS_ENV])
+  end
+  task :create => :load_config if Rake.application.lookup(:load_config)
+
+  redefine_task :drop => :environment do
+    config = ActiveRecord::Base.configurations[RAILS_ENV]
+    begin
+      ActiveRecord::Base.establish_connection(config)
+      db = ActiveRecord::Base.connection.database_name
+      ActiveRecord::Base.connection.drop_database(db)
+    rescue
+      drop_database(config.merge('adapter' => config['adapter'].sub(/^jdbc/, '')))
     end
+  end
+  task :drop => :load_config if Rake.application.lookup(:load_config)
 
-    class << self; alias_method :previous_create_database, :create_database; end
-    def create_database(config)
-      begin
-        ActiveRecord::Base.establish_connection(config)
-        ActiveRecord::Base.connection
-      rescue
-        begin
-          url = config['url']
-          if url
-            if url =~ /^(.*\/)/
-              url = $1
-            end
-          end
+  namespace :create do
+    task :all => :environment
+  end
 
-          ActiveRecord::Base.establish_connection(config.merge({'database' => nil, 'url' => url}))
-          ActiveRecord::Base.connection.create_database(config['database'])
-          ActiveRecord::Base.establish_connection(config)
-        rescue
-          previous_create_database(config)
-        end
-      end
-    end
+  namespace :drop do
+    task :all => :environment
+  end
 
-    redefine_task :drop => :environment do
-      config = ActiveRecord::Base.configurations[RAILS_ENV]
+  class << self
+    alias_method :previous_create_database, :create_database
+    alias_method :previous_drop_database, :drop_database
+  end
+
+  def create_database(config)
+    begin
+      ActiveRecord::Base.establish_connection(config)
+      ActiveRecord::Base.connection
+    rescue
       begin
+        if url = config['url'] && url =~ /^(.*(?<!\/)\/)(?=\w)/
+          url = $1
+        end
+
+        ActiveRecord::Base.establish_connection(config.merge({'database' => nil, 'url' => url}))
+        ActiveRecord::Base.connection.create_database(config['database'])
         ActiveRecord::Base.establish_connection(config)
-        db = ActiveRecord::Base.connection.database_name
-        ActiveRecord::Base.connection.drop_database(db)
       rescue
-        drop_database(config)
+        previous_create_database(config.merge('adapter' => config['adapter'].sub(/^jdbc/, '')))
       end
     end
   end
 
+  def drop_database(config)
+    previous_drop_database(config.merge('adapter' => config['adapter'].sub(/^jdbc/, '')))
+  end
+
   namespace :structure do
     redefine_task :dump => :environment do
       abcs = ActiveRecord::Base.configurations