Rearrange things a bit

.rubocop.yml

@@ -0,0 +1,20 @@
+AllCops:
+  TargetRubyVersion: 2.4
+
+# Configuration parameters: CountComments, ExcludedMethods.
+Metrics/BlockLength:
+  Max: 26
+
+# Configuration parameters: CountComments.
+Metrics/MethodLength:
+  Max: 18
+
+# Offense count: 2
+Style/Documentation:
+  Exclude:
+    - '**/*'
+
+# Configuration parameters: AllowHeredoc, AllowURI, URISchemes, IgnoreCopDirectives, IgnoredPatterns.
+# URISchemes: http, https
+Metrics/LineLength:
+  Max: 102

Gemfile

@@ -1,4 +1,4 @@
-source "https://rubygems.org"
+source 'https://rubygems.org'
 
 # Specify your gem's dependencies in es_index.gemspec
 gemspec

Rakefile

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

bin/console

@@ -1,11 +1,11 @@
 #!/usr/bin/env ruby
 
-require "bundler/setup"
-require "es_index"
+require 'bundler/setup'
+require 'es_index'
 
 # You can add fixtures and/or initialization code here to make experimenting
 # with your gem easier. You can also use a different console, if you like.
 
 # (If you use this, don't forget to add pry to your Gemfile!)
-require "pry"
+require 'pry'
 Pry.start

es_index.gemspec

@@ -1,40 +1,40 @@
 
-lib = File.expand_path("../lib", __FILE__)
+lib = File.expand_path('../lib', __FILE__)
 $LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
-require "es_index/version"
+require 'es_index/version'
 
 Gem::Specification.new do |spec|
-  spec.name          = "es_index"
+  spec.name          = 'es_index'
   spec.version       = EsIndex::VERSION
-  spec.authors       = ["Isaac Seymour"]
-  spec.email         = ["i.seymour@oxon.org"]
+  spec.authors       = ['Isaac Seymour']
+  spec.email         = ['i.seymour@oxon.org']
 
-  spec.summary       = %q{Zero-downtime (re-)indexing of ActiveRecord models into Elasticsearch.}
-  spec.description   = %q{An index manager for Elasticsearch and ActiveRecord}
-  spec.homepage      = "https://github.com/carwow/es_index"
-  spec.license       = "MIT"
+  spec.summary       = 'Zero-downtime (re-)indexing of ActiveRecord models into Elasticsearch.'
+  spec.description   = 'An index manager for Elasticsearch and ActiveRecord'
+  spec.homepage      = 'https://github.com/carwow/es_index'
+  spec.license       = 'MIT'
 
   # Prevent pushing this gem to RubyGems.org. To allow pushes either set the 'allowed_push_host'
   # to allow pushing to a single host or delete this section to allow pushing to any host.
   if spec.respond_to?(:metadata)
-    spec.metadata["allowed_push_host"] = "TODO: Set to 'http://mygemserver.com'"
+    spec.metadata['allowed_push_host'] = "TODO: Set to 'http://mygemserver.com'"
   else
-    raise "RubyGems 2.0 or newer is required to protect against " \
-      "public gem pushes."
+    raise 'RubyGems 2.0 or newer is required to protect against ' \
+      'public gem pushes.'
   end
 
-  spec.files         = `git ls-files -z`.split("\x0").reject do |f|
+  spec.files = `git ls-files -z`.split("\x0").reject do |f|
     f.match(%r{^(test|spec|features)/})
   end
-  spec.bindir        = "exe"
+  spec.bindir        = 'exe'
   spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
-  spec.require_paths = ["lib"]
+  spec.require_paths = ['lib']
 
-  spec.add_dependency "activesupport"
+  spec.add_dependency 'activesupport'
 
-  spec.add_development_dependency "bundler"
-  spec.add_development_dependency "pry"
-  spec.add_development_dependency "rubocop"
-  spec.add_development_dependency "rspec"
-  spec.add_development_dependency "rake"
+  spec.add_development_dependency 'bundler'
+  spec.add_development_dependency 'pry'
+  spec.add_development_dependency 'rake'
+  spec.add_development_dependency 'rspec'
+  spec.add_development_dependency 'rubocop'
 end

lib/es_index/config.rb

@@ -0,0 +1,51 @@
+module EsIndex
+  class Config
+    attr_reader :client, :data_source
+
+    def initialize(
+      client:,
+      data_source:,
+      mapping:,
+      **overrides,
+      &index_data
+    )
+      @client = client
+      @data_source = data_source
+      @mapping = mapping
+      @index_data = index_data
+      @overrides = overrides
+    end
+
+    def index_data(model)
+      @index_data.call(model)
+    end
+
+    def read_alias
+      @read_alias ||= overrides.fetch(:read_alias) { data_source.table_name }
+    end
+
+    def write_alias
+      @write_alias ||= overrides.fetch(:write_alias) { [read_alias, 'write'].join('_') }
+    end
+
+    def type
+      @type ||= overrides.fetch(:type, read_alias.singularize)
+    end
+
+    def logger
+      return Rails.logger if defined?(Rails)
+      @logger ||= Logger.new(STDOUT)
+    end
+
+    def index_definition
+      {
+        settings: overrides.fetch(:index_settings, {}),
+        mappings: { type => mapping }
+      }
+    end
+
+    private
+
+    attr_reader :overrides, :mapping
+  end
+end

lib/es_index/index_manager.rb

@@ -1,22 +1,32 @@
 module EsIndex
   class IndexManager
-    attr_reader :config
-
     def initialize(config)
       @config = config
     end
 
     def create_index(unique_name)
       full_name = [config.read_alias, unique_name].join('_')
 
-      client.indices.create(
+      config.client.indices.create(
         index: full_name,
         body: config.index_definition
       )
 
       config.client.indices.put_alias(index: full_name, name: config.write_alias)
     end
 
+    def populate_index(unique_name = nil, batch_size: 3000)
+      index_name = if unique_name
+                     [config.read_alias, unique_name].join('_')
+                   else
+                     config.write_alias
+                   end
+
+      config.data_source.find_in_batches(batch_size: batch_size) do |batch|
+        indexer.index_batch(batch, index_name: index_name)
+      end
+    end
+
     def switch_read_index(new_name)
       new_index = [config.read_alias, new_name].join('_')
 
@@ -29,27 +39,27 @@ def switch_read_index(new_name)
         ({ remove: { index: old_index, alias: config.read_alias } } if old_index)
 
       config.client.indices.update_aliases(body: {
-        actions: [
-          remove_action,
-          { add: { index: new_index, alias: config.read_alias } }
-        ].compact
-      })
+                                             actions: [
+                                               remove_action,
+                                               { add: { index: new_index, alias: config.read_alias } }
+                                             ].compact
+                                           })
     end
 
     def stop_dual_writes
-      Rails.logger.info('Stopping dual writes - making index read and write aliases the same')
+      logger.info('Stopping dual writes - making index read and write aliases the same')
       current_index = config.client.indices.get_alias(name: config.read_alias).keys.first
 
-      Rails.logger.info("Currently used index is #{current_index}")
+      logger.info("Currently used index is #{current_index}")
 
       other_write_indices = config.client.indices.get_alias(name: config.write_alias).keys
-        .reject { |name| name == current_index }
+                                  .reject { |name| name == current_index }
 
       if other_write_indices.none?
-        Rails.logger.info("No write indexes that aren't the read index. Nothing to do!")
+        logger.info("No write indexes that aren't the read index. Nothing to do!")
         return
       end
-      Rails.logger.info("Stopping writes to #{other_write_indices.count} old ES indices: " \
+      logger.info("Stopping writes to #{other_write_indices.count} old ES indices: " \
                         "#{other_write_indices.join(', ')}")
 
       actions = other_write_indices.map do |index|
@@ -59,36 +69,33 @@ def stop_dual_writes
     end
 
     def cleanup_old_indices
-      Rails.logger.info('Cleaning up old indices in Elasticsearch')
+      logger.info('Cleaning up old indices in Elasticsearch')
       current_index = config.client.indices.get_alias(name: config.read_alias).keys.first
 
-      Rails.logger.info("Currently used index is #{current_index}")
+      logger.info("Currently used index is #{current_index}")
 
       indices_to_delete = config.client
-        .cat
-        .indices(format: :json)
-        .map { |index| index['index'] }
-        .select { |name| name.start_with?(config.read_alias) }
-        .reject { |name| name == current_index }
+                                .cat
+                                .indices(format: :json)
+                                .map { |index| index['index'] }
+                                .select { |name| name.start_with?(config.read_alias) }
+                                .reject { |name| name == current_index }
 
       if indices_to_delete.none?
-        Rails.logger.info('Nothing to do: no old indices')
+        logger.info('Nothing to do: no old indices')
         return
       end
-      Rails.logger.info("Deleting #{indices_to_delete.count} old indices: #{indices_to_delete.join(', ')}")
+      logger.info("Deleting #{indices_to_delete.count} old indices: #{indices_to_delete.join(', ')}")
       config.client.indices.delete(index: indices_to_delete)
     end
 
-    def populate_index(unique_name = nil, batch_size: 3000)
-      index_name = if unique_name
-                     [config.read_alias, unique_name].join('_')
-                   else
-                     config.write_alias
-                   end
+    private
 
-      config.data_source.find_in_batches(batch_size: batch_size) do |batch|
-        index_batch(batch, index_name: index_name)
-      end
+    attr_reader :config
+    delegate :logger, to: :config
+
+    def indexer
+      @indexer ||= Indexer.new(config)
     end
   end
 end

lib/es_index/indexer.rb

@@ -62,7 +62,7 @@ def delete_by_id(id)
     end
 
     def delete_by_ids(ids)
-      Rails.logger.info('ES: Deleting batch records')
+      logger.info('ES: Deleting batch records')
 
       indices = config.client.indices.get_alias(name: config.write_alias).keys
       commands = []
@@ -86,14 +86,15 @@ def delete_by_ids(ids)
     end
 
     def delete_by_query(query)
-      Rails.logger.info('ES: Deleting batch records')
+      logger.info('ES: Deleting batch records')
 
       config.client.delete_by_query(index: config.write_alias, body: { query: query })
     end
 
     private
 
     attr_reader :config
+    delegate :logger, to: :config
 
     def current_version
       config.data_source.connection.select_one('SELECT txid_current()').fetch('txid_current')

lib/es_index/version.rb

@@ -1,3 +1,3 @@
 module EsIndex
-  VERSION = "0.1.0"
+  VERSION = '0.1.0'.freeze
 end

spec/es_index_spec.rb

@@ -1,9 +1,9 @@
 RSpec.describe EsIndex do
-  it "has a version number" do
+  it 'has a version number' do
     expect(EsIndex::VERSION).not_to be nil
   end
 
-  it "does something useful" do
+  it 'does something useful' do
     expect(false).to eq(true)
   end
 end

spec/spec_helper.rb

@@ -1,8 +1,8 @@
-require "bundler/setup"
-require "es_index"
+require 'bundler/setup'
+require 'es_index'
 
 RSpec.configure do |config|
-  config.example_status_persistence_file_path = ".rspec_status"
+  config.example_status_persistence_file_path = '.rspec_status'
 
   config.disable_monkey_patching!