backfiller 0.0.1

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 3c3e9ee72c70a3b89caba10d6073a87058290680
+  data.tar.gz: 22071f36c634b6353787cac65a8c2984e8eb1ade
+SHA512:
+  metadata.gz: 3931af99bdc3771c1636c24e4ce1374ca0af0739090c6f56defd530f4fa26f6cd3ac497feb9cd2824fa91a7f8c6a2c4e3cef0b64fb51b4d76cd4f13d90c033c1
+  data.tar.gz: c6bd78c26fd968f0df688bb89b52d22dc7681e9f781825300c4d5183310db95aaab56a6e1efc3b0ed8f0419ec20dd271e2b46d4114f2fbba89aed424e266debb

data/.gitignore

@@ -0,0 +1,12 @@
+/.bundle/
+/.yardoc
+/Gemfile.lock
+/_yardoc/
+/coverage/
+/doc/
+/pkg/
+/spec/reports/
+/tmp/
+
+# rspec failure tracking
+.rspec_status

data/.rspec

@@ -0,0 +1,2 @@
+--format documentation
+--color

data/Gemfile

@@ -0,0 +1,6 @@
+source "https://rubygems.org"
+
+git_source(:github) {|repo_name| "https://github.com/#{repo_name}" }
+
+# Specify your gem's dependencies in backfiller.gemspec
+gemspec

data/README.md

@@ -0,0 +1,117 @@
+# Backfiller
+
+The backfill machine for null database columns.
+This gem maybe handly for `no-downtime` deployment especially when you need to fill columns for table with huge amount for records without locking the table.
+
+## Typical no-downtime and non-locking cycle
+
+* add migaration that adds new column (null: true)
+* deploy and run migration task
+* deploy code that starts filling new column in corresponding flows
+* add backfill task
+* deploy and run backflill task
+* [optional] add migration that invokes backfill task asn so keep all environments consistent (except production environment because we already backfilled data)
+* add migration that disallow null values (null: false)
+* deploy code that starts using new column
+
+
+## Concept
+
+Idea is to prepare all data in selection method on database server and fetch all data using CURSOR and then build simple UPDATE queries.
+With this way we minimize db server resources usage and we lock only one record (atomic update).
+We use two connections to database:
+* master - to creates cursor in transaction and fetch data in batches.
+* worker - to execute small atomic update queries (no wrapper transaction)
+
+Even if backfill process crashes you may resolve issue and run it again to process remaining amount of data.
+
+## Connection adapters
+
+Curently it supports only PostgreSQL ActiveRecord adapter.
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+```ruby
+gem 'backfiller'
+```
+
+And then execute:
+
+    $ bundle
+
+Or install it yourself as:
+
+    $ gem install backfiller
+
+## Usage
+
+Assume we we want to backfill `profiles.name` column from `users.first_name`, `users.last_name` columns.
+
+Create backfill task into `db/backfill/profile_name.rb` and defined required methods:
+
+```ruby
+class Backfill::ProfileName
+  def select_sql
+    <<-SQL.strip_heredoc
+      SELECT
+        profile.id AS profile_id,
+        CONCAT(users.first_name, ' ', users.last_name) AS profile_name
+      FROM profiles
+      INNER JOIN users ON
+        users.id = profiles.user_id
+      WHERE
+        profiles.name IS NULL
+    SQL
+  end
+
+  def update_sql(connection, row)
+    <<-SQL.strip_heredoc
+      UPDATE profiles SET
+        name = #{connection.quote(row['profile_name'])}
+     WHERE
+       id = #{connection.quote(row[:profile_id])}
+    SQL
+  end
+end
+```
+
+And then just run rake task:
+
+```bash
+$ rails db:backfill[profile_name]
+```
+
+
+## Configuration
+
+For Rails application backfiller is initialized with next options
+
+* task_directory: `RAILS_ROOT/db/backfill`
+* task_namespace: `Backfill`
+* batch_size - `1_000`
+* connection_pool: `ApplicationRecord.connection_pool`
+* logger: `ApplicationRecord.logger`
+
+You may change it globally via `config/initializers/backfiller.rb`:
+
+```ruby
+Backfiller.configure do |config|
+  config.foo = bar
+end
+```
+
+Or specify some options in certain backfill task
+
+```ruby
+class Backfill::Foo
+  def batch_size
+    100
+  end
+end
+```
+
+## Authors
+
+* [Andriy Yanko](http://ayanko.github.io)

data/Rakefile

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

data/backfiller.gemspec

@@ -0,0 +1,27 @@
+# coding: utf-8
+lib = File.expand_path("../lib", __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+
+Gem::Specification.new do |spec|
+  spec.name          = "backfiller"
+  spec.version       = "0.0.1"
+  spec.authors       = ["Andriy Yanko"]
+  spec.email         = ["andriy.yanko@railsware.com"]
+
+  spec.summary       = %q{Backfiller for null database columns}
+  spec.homepage      = "https://github.com/railsware/backfiller"
+  spec.license       = "MIT"
+
+  spec.files         = `git ls-files -z`.split("\x0").reject do |f|
+    f.match(%r{^(test|spec|features)/})
+  end
+  spec.bindir        = "exe"
+  spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
+  spec.require_paths = ["lib"]
+
+  spec.add_dependency "activerecord", ">= 5.0.0"
+
+  spec.add_development_dependency "bundler", "~> 1.15"
+  spec.add_development_dependency "rake", "~> 10.0"
+  spec.add_development_dependency "rspec", "~> 3.0"
+end

data/bin/console

@@ -0,0 +1,14 @@
+#!/usr/bin/env ruby
+
+require "bundler/setup"
+require "backfiller"
+
+# 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"
+# Pry.start
+
+require "irb"
+IRB.start(__FILE__)

data/bin/setup

@@ -0,0 +1,8 @@
+#!/usr/bin/env bash
+set -euo pipefail
+IFS=$'\n\t'
+set -vx
+
+bundle install
+
+# Do any other automated setup that you need to do here

data/lib/backfiller.rb

@@ -0,0 +1,5 @@
+require 'backfiller/configuration'
+require 'backfiller/cursor'
+require 'backfiller/runner'
+
+require 'backfiller/railtie' if defined?(Rails)

data/lib/backfiller/configuration.rb

@@ -0,0 +1,27 @@
+module Backfiller
+
+  class << self
+    def configure
+      yield self
+    end
+
+    attr_accessor :task_directory
+
+    attr_accessor :task_namespace
+
+    attr_accessor :connection_pool
+
+    attr_accessor :batch_size
+
+    attr_accessor :logger
+
+    def run(task_name)
+      Backfiller::Runner.new(task_name).run
+    end
+
+    def log(message)
+      logger.info "[Backfiller] #{message}" if logger
+    end
+  end
+
+end

data/lib/backfiller/cursor.rb

@@ -0,0 +1,16 @@
+require 'backfiller/cursor/postgresql'
+
+module Backfiller
+  module Cursor
+
+    def self.new(connection, *args)
+      case connection
+      when ActiveRecord::ConnectionAdapters::PostgreSQLAdapter
+        Backfiller::Cursor::Postgresql.new(connection, *args)
+      else
+        raise "Unsupported connection #{connection.inspect}"
+      end
+    end
+
+  end
+end

data/lib/backfiller/cursor/postgresql.rb

@@ -0,0 +1,27 @@
+module Backfiller
+  module Cursor
+    class Postgresql
+
+      attr_reader :connection
+
+      def initialize(connection, name, query)
+        @connection = connection
+        @name = name
+        @query = query
+      end
+
+      def open
+        @connection.execute "DECLARE #{@name} NO SCROLL CURSOR WITHOUT HOLD FOR #{@query}"
+      end
+
+      def fetch(count)
+        @connection.select_all "FETCH #{count} FROM #{@name}"
+      end
+
+      def close
+        @connection.execute "CLOSE #{@name}"
+      end
+
+    end
+  end
+end

data/lib/backfiller/railtie.rb

@@ -0,0 +1,28 @@
+module Backfiller
+  class Railtie < Rails::Railtie
+
+    rake_tasks do
+      load 'backfiller/tasks/db.rake'
+    end
+
+    initializer 'backfiller.configure' do
+      Backfiller.configure do |config|
+        config.task_directory = Rails.root.join('db', 'backfill')
+
+        config.task_namespace = 'backfill'
+
+        config.batch_size = 1_000
+
+        config.connection_pool = defined?(ApplicationRecord) ? ApplicationRecord.connection_pool : ActiveRecord::Base.connection_pool
+
+        config.logger = defined?(ApplicationRecord) ? ApplicationRecord.logger : ActiveRecord::Base.logger
+      end
+    end
+
+    config.after_initialize do
+      task_module = Backfiller.task_namespace.classify
+      Object.const_set(task_module, Module.new) unless Object.const_defined?(task_module)
+    end
+
+  end
+end

data/lib/backfiller/runner.rb

@@ -0,0 +1,88 @@
+module Backfiller
+  class Runner
+
+    attr_reader \
+      :task,
+      :connection_pool,
+      :batch_size
+
+    def initialize(task_name)
+      @task = build_task(task_name)
+      @connection_pool = @task.respond_to?(:connection_pool) ? @task.connection_pool : Backfiller.connection_pool
+      @batch_size = @task.respond_to?(:batch_size) ? @task.batch_size : Backfiller.batch_size
+    end
+
+    def run
+      master_connection = acquire_connection
+      worker_connection = acquire_connection
+
+      fetch_each(master_connection) do |row|
+        update_row(worker_connection, row)
+      end
+
+      release_connection(master_connection)
+      release_connection(worker_connection)
+    end
+
+    private
+
+    def build_task(task_name)
+      Backfiller.log "Build #{task_name} task"
+      require File.join(Backfiller.task_directory, task_name)
+      "#{Backfiller.task_namespace}/#{task_name}".classify.constantize.new
+    end
+
+    ###########################################################################
+
+    def acquire_connection
+      connection_pool.checkout
+    end
+
+    def release_connection(connection)
+      connection_pool.checkin(connection)
+    end
+
+    ###########################################################################
+
+    def build_cursor(connection)
+      Backfiller::Cursor.new(connection, 'backfill_cursor', task.select_sql)
+    end
+
+    def fetch_each(master_connection, &block)
+      cursor = build_cursor(master_connection)
+
+      cursor.connection.transaction do
+        Backfiller.log "Open cursor"
+        cursor.open
+
+        Backfiller.log "Start fetch loop"
+        fetch_loop(cursor, &block)
+
+        Backfiller.log "Close cursor"
+        cursor.close
+      end
+    end
+
+    def fetch_loop(cursor, &block)
+      count = 0
+
+      loop do
+        result = cursor.fetch(batch_size)
+
+        break if result.empty?
+
+        result.each do |row|
+          block.call(row)
+          count += 1
+        end
+
+        Backfiller.log "Processed #{count}"
+      end
+    end
+
+    def update_row(connection, row)
+      connection.execute task.update_sql(connection, row)
+    end
+
+  end
+end

data/lib/backfiller/tasks/db.rake

@@ -0,0 +1,10 @@
+namespace :db do
+
+  desc 'Run database backfill task'
+  task :backfill, [:name] => :environment do |t, args|
+    raise 'Please specify backfill task name' unless args[:name]
+    Backfiller.logger.level = :info if Backfiller.logger
+    Backfiller.run(args[:name])
+  end
+
+end

metadata

@@ -0,0 +1,116 @@
+--- !ruby/object:Gem::Specification
+name: backfiller
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- Andriy Yanko
+autorequire: 
+bindir: exe
+cert_chain: []
+date: 2017-09-30 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: activerecord
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 5.0.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 5.0.0
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.15'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.15'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '10.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '10.0'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+description: 
+email:
+- andriy.yanko@railsware.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".gitignore"
+- ".rspec"
+- Gemfile
+- README.md
+- Rakefile
+- backfiller.gemspec
+- bin/console
+- bin/setup
+- lib/backfiller.rb
+- lib/backfiller/configuration.rb
+- lib/backfiller/cursor.rb
+- lib/backfiller/cursor/postgresql.rb
+- lib/backfiller/railtie.rb
+- lib/backfiller/runner.rb
+- lib/backfiller/tasks/db.rake
+homepage: https://github.com/railsware/backfiller
+licenses:
+- MIT
+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: 
+rubygems_version: 2.6.11
+signing_key: 
+specification_version: 4
+summary: Backfiller for null database columns
+test_files: []
+has_rdoc: