First implementation (Version)

Author: magec

.gitignore

@@ -11,6 +11,7 @@ capybara-*.html
 *.orig
 rerun.txt
 pickle-email-*.html
+*log
 
 # TODO Comment out this rule if you are OK with secrets being uploaded to the repo
 config/initializers/secret_token.rb
@@ -42,3 +43,5 @@ bower.json
 
 # Ignore Byebug command history file.
 .byebug_history
+
+*~

Gemfile.lock

@@ -0,0 +1,131 @@
+PATH
+  remote: .
+  specs:
+    utf8mb4rails (0.1.0)
+      departure (~> 4.0, >= 4.0.1)
+      mysql2 (~> 0.4.0)
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    actionmailer (4.2.9)
+      actionpack (= 4.2.9)
+      actionview (= 4.2.9)
+      activejob (= 4.2.9)
+      mail (~> 2.5, >= 2.5.4)
+      rails-dom-testing (~> 1.0, >= 1.0.5)
+    actionpack (4.2.9)
+      actionview (= 4.2.9)
+      activesupport (= 4.2.9)
+      rack (~> 1.6)
+      rack-test (~> 0.6.2)
+      rails-dom-testing (~> 1.0, >= 1.0.5)
+      rails-html-sanitizer (~> 1.0, >= 1.0.2)
+    actionview (4.2.9)
+      activesupport (= 4.2.9)
+      builder (~> 3.1)
+      erubis (~> 2.7.0)
+      rails-dom-testing (~> 1.0, >= 1.0.5)
+      rails-html-sanitizer (~> 1.0, >= 1.0.3)
+    activejob (4.2.9)
+      activesupport (= 4.2.9)
+      globalid (>= 0.3.0)
+    activemodel (4.2.9)
+      activesupport (= 4.2.9)
+      builder (~> 3.1)
+    activerecord (4.2.9)
+      activemodel (= 4.2.9)
+      activesupport (= 4.2.9)
+      arel (~> 6.0)
+    activesupport (4.2.9)
+      i18n (~> 0.7)
+      minitest (~> 5.1)
+      thread_safe (~> 0.3, >= 0.3.4)
+      tzinfo (~> 1.1)
+    arel (6.0.4)
+    builder (3.2.3)
+    concurrent-ruby (1.0.5)
+    departure (4.0.1)
+      mysql2 (~> 0.4.0)
+      rails (~> 4.2.0)
+    diff-lcs (1.3)
+    erubis (2.7.0)
+    globalid (0.4.0)
+      activesupport (>= 4.2.0)
+    i18n (0.8.6)
+    loofah (2.0.3)
+      nokogiri (>= 1.5.9)
+    mail (2.6.6)
+      mime-types (>= 1.16, < 4)
+    mime-types (3.1)
+      mime-types-data (~> 3.2015)
+    mime-types-data (3.2016.0521)
+    mini_portile2 (2.2.0)
+    minitest (5.10.3)
+    mysql2 (0.4.9)
+    nokogiri (1.8.0)
+      mini_portile2 (~> 2.2.0)
+    rack (1.6.8)
+    rack-test (0.6.3)
+      rack (>= 1.0)
+    rails (4.2.9)
+      actionmailer (= 4.2.9)
+      actionpack (= 4.2.9)
+      actionview (= 4.2.9)
+      activejob (= 4.2.9)
+      activemodel (= 4.2.9)
+      activerecord (= 4.2.9)
+      activesupport (= 4.2.9)
+      bundler (>= 1.3.0, < 2.0)
+      railties (= 4.2.9)
+      sprockets-rails
+    rails-deprecated_sanitizer (1.0.3)
+      activesupport (>= 4.2.0.alpha)
+    rails-dom-testing (1.0.8)
+      activesupport (>= 4.2.0.beta, < 5.0)
+      nokogiri (~> 1.6)
+      rails-deprecated_sanitizer (>= 1.0.1)
+    rails-html-sanitizer (1.0.3)
+      loofah (~> 2.0)
+    railties (4.2.9)
+      actionpack (= 4.2.9)
+      activesupport (= 4.2.9)
+      rake (>= 0.8.7)
+      thor (>= 0.18.1, < 2.0)
+    rake (10.5.0)
+    rspec (3.6.0)
+      rspec-core (~> 3.6.0)
+      rspec-expectations (~> 3.6.0)
+      rspec-mocks (~> 3.6.0)
+    rspec-core (3.6.0)
+      rspec-support (~> 3.6.0)
+    rspec-expectations (3.6.0)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.6.0)
+    rspec-mocks (3.6.0)
+      diff-lcs (>= 1.2.0, < 2.0)
+      rspec-support (~> 3.6.0)
+    rspec-support (3.6.0)
+    sprockets (3.7.1)
+      concurrent-ruby (~> 1.0)
+      rack (> 1, < 3)
+    sprockets-rails (3.2.1)
+      actionpack (>= 4.0)
+      activesupport (>= 4.0)
+      sprockets (>= 3.0.0)
+    thor (0.20.0)
+    thread_safe (0.3.6)
+    tzinfo (1.2.3)
+      thread_safe (~> 0.1)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  bundler (~> 1.12)
+  rake (~> 10.0)
+  rspec (~> 3.0)
+  utf8mb4rails!
+
+BUNDLED WITH
+   1.12.5

README.md

@@ -1,2 +1,53 @@
-# utf8mb4rails
-Adds a couple of rake tasks to deal with utf8 to utf8mb4 migrations for mysql.
+# Utf8mb4rails
+
+A simple gem that adds a rake task to deal with utf8 to utf8mb4 migrations for mysql.
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+```ruby
+gem 'utf8mb4rails'
+```
+
+And then execute:
+
+    $ bundle
+
+Or install it yourself as:
+
+    $ gem install utf8mb4rails
+
+## Usage
+
+This gem will add in your rails project a new task called utf8mb4. This tast is aimed to easily convert tables in utf8 into
+utf8mb4 without truncating column contents and maintaining its original schema (without changing max sizes).
+
+**USE AT YOUR OWN RISK!!!**
+
+    $ rake -T
+
+```
+...
+rake db:utf8mb4                          # migrates a table (ENV["TABLE"]) to utf8mb encoding
+...
+
+```
+
+You can modify its behavior using env vars.
+
+- **TABLE**: The table you want to migrate, if no column is specified every single column will be migrated, also the default charset
+of the table is altered.
+- **COLUMN**: The column to be migrated (needs TABLE defined). It will only migrate that column definition.
+- **COLLATION**: The collation (utf8mb4_unicode_520_ci by default)
+
+
+## Contributing
+
+Bug reports and pull requests are welcome on GitHub at https://github.com/magec/utf8mb4rails.
+
+
+## License
+
+The gem is available as open source under the terms of the [MIT License](http://opensource.org/licenses/MIT).
+

lib/utf8mb4rails.rb

@@ -1,5 +1,43 @@
-require "utf8mb4rails/version"
+require 'utf8mb4rails/version'
+require 'utf8mb4rails/migrator'
+require 'departure'
+
+Departure.configure do |_config|; end
+def usage
+  puts
+  puts 'Usage: '
+  puts
+  puts 'For migrating an entire table use:'
+  puts '$ TABLE=table_name rake db:utf8mb4'
+  puts
+  puts 'For migrating an specified column table:'
+  puts '$ TABLE=table_name COLUMN=column_name rake db:utf8mb4'
+  puts
+  puts 'You can also specify the COLLATION (utf8mb4_unicode_520_ci)'
+end
 
 module Utf8mb4rails
-  # Your code goes here...
+  extend Rake::DSL
+
+  namespace :db do
+    desc 'migrates a table (ENV["TABLE"]) to utf8mb encoding'
+    task utf8mb4: :environment do
+      table = ENV['TABLE']
+      unless table
+        puts 'Please specify a table with TABLE='
+        usage
+        exit 1
+      end
+
+      column = ENV['COLUMN']
+      runner = Migrator::Runner.new
+      if column
+        puts 'Migrating #{table}.#{column}'
+        runner.migrate_column!(table, column)
+      else
+        puts 'No column specified, will migrate the entire table'
+        runner.migrate_table!(table)
+      end
+    end
+  end
 end

lib/utf8mb4rails/migrator.rb

@@ -0,0 +1,4 @@
+# Use percona
+require_relative 'migrator/db_inspector'
+require_relative 'migrator/runner'
+require_relative 'migrator/column_info'

lib/utf8mb4rails/migrator/column_info.rb

@@ -0,0 +1,38 @@
+module Utf8mb4rails
+  module Migrator
+
+    # Abstraction for dealing with column information
+    # received from the database
+    class ColumnInfo
+      TEXT_TYPES = %w(CHAR VARCHAR TINYTEXT TEXT MEDIUMTEXT LONGTEXT).freeze
+      attr_accessor :info
+
+      # Receives a hash with (:type, :default, :max_length, :charset)
+      def initialize(info)
+        @info = info
+      end
+
+      # @return Bool : True if the column is in utf8mb4
+      def utf8mb4?
+        info[:charset] =~ /utf8mb4/
+      end
+
+      # @return String : the sql part of the new type of the column definition
+      def new_type_for_sql
+        return "#{info[:type]}(#{info[:max_length]})" if info[:type] =~ /CHAR/
+        info[:type]
+      end
+
+      # @return String : the sql part of the default value of the column definition
+      def default_value_for_sql
+        return nil unless info[:default]
+        "default '#{info[:default]}'"
+      end
+
+      # @return Bool : True if the column is of a text type
+      def text_column?
+        TEXT_TYPES.include?(info[:type])
+      end
+    end
+  end
+end

lib/utf8mb4rails/migrator/db_inspector.rb

@@ -0,0 +1,44 @@
+module Utf8mb4rails
+  module Migrator
+    class DBInspector
+      # Initializes AR
+      def initialize
+        ActiveRecord::Base.establish_connection(
+          ActiveRecord::Base.connection_config.merge(adapter: 'percona')
+        )
+      end
+
+      ##
+      # @returns [Array<String>] An array with column names
+      def columns(table)
+        ActiveRecord::Base.connection.columns(table).map(&:name)
+      end
+
+      # Returns a hash with information about the column
+      #
+      # @param [String] table
+      # @param [String] column
+      #
+      # @returns [Hash] { type: String, default: String or nil, charset: String or nil }
+      def column_info(table, column)
+        sql = "SELECT DATA_TYPE, COLUMN_DEFAULT, CHARACTER_SET_NAME, CHARACTER_MAXIMUM_LENGTH
+        FROM INFORMATION_SCHEMA.COLUMNS WHERE table_name = '#{table}'
+        AND COLUMN_NAME = '#{column}'
+        AND TABLE_SCHEMA='#{database_name}'"
+        result = ActiveRecord::Base.connection.execute(sql).first
+        ColumnInfo.new(
+          type: result[0].upcase,
+          default: result[1],
+          charset: result[2],
+          max_length: result[3]
+        )
+      end
+
+      private
+
+      def database_name
+        ActiveRecord::Base.connection.current_database
+      end
+    end
+  end
+end

lib/utf8mb4rails/migrator/runner.rb

@@ -0,0 +1,43 @@
+module Utf8mb4rails
+  module Migrator
+    class Runner
+      NEW_COLLATION = ENV.fetch('COLLATION', 'utf8mb4_unicode_520_ci').freeze
+
+      attr_accessor :inspector
+
+      def initialize
+        @inspector = DBInspector.new
+      end
+
+      def migrate_column!(table, column)
+        column_info = inspector.column_info(table, column)
+        if column_info.utf8mb4?
+          puts "          Skipping column #{column} (already in utf8mb4)'"
+          return
+        end
+        return unless column_info.text_column?
+        sql = "ALTER TABLE `#{table}` DROP COLUMN \`#{column}`\ ,
+       ADD COLUMN \`#{column}`\ #{column_info.new_type_for_sql} CHARACTER SET utf8mb4
+       COLLATE #{NEW_COLLATION} #{column_info.default_value_for_sql}"
+        ActiveRecord::Base.connection.execute(sql)
+      end
+
+      def migrate_table_schema!(table)
+        sql = "ALTER TABLE `#{table}` CONVERT TO CHARACTER SET utf8mb4 COLLATE #{NEW_COLLATION}"
+        ActiveRecord::Base.connection.execute(sql)
+      end
+
+      def migrate_table!(table)
+        inspector.columns(table).each do |column|
+          puts "      migrating #{table}:#{column}"
+          migrate_column!(table, column)
+        end
+        migrate_table_schema!(table)
+      rescue
+        puts "Problems accesing the table #{table}"
+        puts $!
+        exit 1
+      end
+    end
+  end
+end

spec/fixtures/migrate/01_create_activities.rb

@@ -0,0 +1,9 @@
+class CreateActivities < ActiveRecord::Migration
+  def change
+    create_table :activities do |t|
+      t.string :name
+      t.text :description
+      t.timestamps null: false
+    end
+  end
+end