good_job 2.8.1 → 2.9.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: bca3de7432f33744351456a803dcbb74f73c0228f824dad035b51565c711e529
-  data.tar.gz: 0e0b3d2959f2db50c00d61a49ccad52be73711ac8e07cfd7cfffb186923fb5b2
+  metadata.gz: 815933ed8a435e9826bbb655e92c298d10d5bd74eb4f0474cb25d49a194d5cae
+  data.tar.gz: 29f7753f2401599feb16fc4b25e9c7073855f1fa135d88e640051533c2c24593
 SHA512:
-  metadata.gz: ad760ae64fe74a08943013b9359e675bf46ad366cb9a688937a182e2fc8b6abd336fa8b8408cc61e22cf7c3ab713d0fa9e0a6ae973196d653e5c763818ef1672
-  data.tar.gz: 630f2765587b5b675eaf1e33b6c5b39a83ada8980a0638011e2cc71f4ce963d6f6287db5d50a9c5efb3d1ba7c1402e7ef1f2beb11369ad981a6046c8e4491f30
+  metadata.gz: 7dbf7489f53095634bb91e2a7e7fd8bb1a009ba3d81fb98312c7b8ae8f0516bc786e13b1100fdef1dd113e80c59b3d8433663ea4773d04ee517927f1bce1b560
+  data.tar.gz: b01d4ca76bb300dd4f755db2eb43758dd8146fe4e4ab1061d47cc973fe8e559b936df004054ed9ed92c9aefc6abee8f7db0ab33bf1ca812b57a9d7c192771077

data/CHANGELOG.md

@@ -1,5 +1,57 @@
 # Changelog
 
+## [v2.9.3](https://github.com/bensheldon/good_job/tree/v2.9.3) (2022-01-23)
+
+[Full Changelog](https://github.com/bensheldon/good_job/compare/v2.9.2...v2.9.3)
+
+**Closed issues:**
+
+- Assets not loaded when Rails is configured with a different hostname for assets [\#491](https://github.com/bensheldon/good_job/issues/491)
+
+**Merged pull requests:**
+
+- Use `*_url` route helpers for Dashboard assets to avoid being overridden by `config.asset_host` [\#493](https://github.com/bensheldon/good_job/pull/493) ([bensheldon](https://github.com/bensheldon))
+
+## [v2.9.2](https://github.com/bensheldon/good_job/tree/v2.9.2) (2022-01-19)
+
+[Full Changelog](https://github.com/bensheldon/good_job/compare/v2.9.1...v2.9.2)
+
+**Fixed bugs:**
+
+- Error on GJ admin UI search form [\#487](https://github.com/bensheldon/good_job/issues/487)
+- Use `websearch_to_tsquery` or \(`plainto_tsquery` for Postgres \< v11\) for Dashboard search filter [\#488](https://github.com/bensheldon/good_job/pull/488) ([bensheldon](https://github.com/bensheldon))
+
+**Merged pull requests:**
+
+- Update README to illustrate using named arguments for the unique key. [\#486](https://github.com/bensheldon/good_job/pull/486) ([phallstrom](https://github.com/phallstrom))
+- Add details about exactly where to require the engine. [\#485](https://github.com/bensheldon/good_job/pull/485) ([phallstrom](https://github.com/phallstrom))
+- $ symbol gets copied when clicking on the copy button [\#484](https://github.com/bensheldon/good_job/pull/484) ([zeevy](https://github.com/zeevy))
+
+## [v2.9.1](https://github.com/bensheldon/good_job/tree/v2.9.1) (2022-01-13)
+
+[Full Changelog](https://github.com/bensheldon/good_job/compare/v2.9.0...v2.9.1)
+
+**Fixed bugs:**
+
+- Start async adapters once `ActiveRecord` and `ActiveJob` have loaded, potentially before `Rails.application.initialized?` [\#483](https://github.com/bensheldon/good_job/pull/483) ([bensheldon](https://github.com/bensheldon))
+
+**Closed issues:**
+
+- Graceful fallback to polling when LISTEN/NOTIFY isn't available [\#482](https://github.com/bensheldon/good_job/issues/482)
+- Long running locks on latest good job [\#480](https://github.com/bensheldon/good_job/issues/480)
+
+## [v2.9.0](https://github.com/bensheldon/good_job/tree/v2.9.0) (2022-01-09)
+
+[Full Changelog](https://github.com/bensheldon/good_job/compare/v2.8.1...v2.9.0)
+
+**Implemented enhancements:**
+
+- Add JRuby / JDBC support for LISTEN  [\#479](https://github.com/bensheldon/good_job/pull/479) ([bensheldon](https://github.com/bensheldon))
+
+**Merged pull requests:**
+
+- Remove demo CleanupJob in favor of using built-in cleanup intervals [\#478](https://github.com/bensheldon/good_job/pull/478) ([bensheldon](https://github.com/bensheldon))
+
 ## [v2.8.1](https://github.com/bensheldon/good_job/tree/v2.8.1) (2022-01-03)
 
 [Full Changelog](https://github.com/bensheldon/good_job/compare/v2.8.0...v2.8.1)

data/README.md

@@ -73,19 +73,19 @@ For more of the story of GoodJob, read the [introductory blog post](https://isla
 1. Install the gem:
 
     ```bash
-    $ bundle install
+    bundle install
     ```
 
 1. Run the GoodJob install generator. This will generate a database migration to create a table for GoodJob's job records:
 
     ```bash
-    $ bin/rails g good_job:install
+    bin/rails g good_job:install
     ```
 
     Run the migration:
 
     ```bash
-    $ bin/rails db:migrate
+    bin/rails db:migrate
     ```
 
    Optional: If using Rails' multiple databases with the `migrations_paths` configuration option, use the `--database` option:
@@ -142,7 +142,7 @@ For more of the story of GoodJob, read the [introductory blog post](https://isla
 ## Compatibility
 
 - **Ruby on Rails:** 5.2+
-- **Ruby:** MRI 2.5+. JRuby 9.2.13+ (_JRuby's `activerecord-jdbcpostgresql-adapter` gem does not support Postgres LISTEN/NOTIFY)._
+- **Ruby:** MRI 2.5+. JRuby 9.2.13+
 - **Postgres:** 9.6+
 
 ## Configuration
@@ -328,7 +328,7 @@ _🚧 GoodJob's dashboard is a work in progress. Please contribute ideas and cod
 
 GoodJob includes a Dashboard as a mountable `Rails::Engine`.
 
-1. Explicitly require the Engine code at the top of your `config/application.rb` file, immediately after Rails is required. This is necessary because the mountable engine is an optional feature of GoodJob.
+1. Explicitly require the Engine code at the top of your `config/application.rb` file, immediately after Rails is required and before Bundler requires the Rails' groups. This is necessary because the mountable engine is an optional feature of GoodJob.
 
     ```ruby
     # config/application.rb
@@ -339,6 +339,8 @@ GoodJob includes a Dashboard as a mountable `Rails::Engine`.
     # ...
     ```
 
+   Note: If you find the dashboard fails to reload due to a routing error and uninitialized constant `GoodJob::ExecutionsController`, this is likely because you are not requiring the engine early enough.
+
 1. Mount the engine in your `config/routes.rb` file. The following will mount it at `http://example.com/good_job`.
 
     ```ruby
@@ -399,6 +401,9 @@ class MyJob < ApplicationJob
     # Can be String or Lambda/Proc that is invoked in the context of the job.
     # Note: Arguments passed to #perform_later must be accessed through `arguments` method.
     key: -> { "Unique-#{arguments.first}" } #  MyJob.perform_later("Alice") => "Unique-Alice"
+
+    # If the method uses named parameters, they can be accessed like so:
+    # key: -> { "Unique-#{arguments.first['name']}" } #  MyJob.perform_later(name: "Alice")
   )
 
   def perform(first_name)

data/engine/app/views/layouts/good_job/base.html.erb

@@ -5,14 +5,15 @@
   <%= csrf_meta_tags %>
   <%= csp_meta_tag %>
 
-  <%= stylesheet_link_tag bootstrap_path(format: :css, v: GoodJob::VERSION) %>
-  <%= stylesheet_link_tag style_path(format: :css, v: GoodJob::VERSION) %>
+  <%# Assets must use *_url route helpers to avoid being overriden by config.asset_host %>
+  <%= stylesheet_link_tag bootstrap_url(format: :css, v: GoodJob::VERSION), skip_pipeline: true %>
+  <%= stylesheet_link_tag style_url(format: :css, v: GoodJob::VERSION) %>
 
-  <%= javascript_include_tag bootstrap_path(format: :js, v: GoodJob::VERSION), nonce: true %>
-  <%= javascript_include_tag chartjs_path(format: :js, v: GoodJob::VERSION), nonce: true %>
-  <%= javascript_include_tag scripts_path(format: :js, v: GoodJob::VERSION), nonce: true %>
+  <%= javascript_include_tag bootstrap_url(format: :js, v: GoodJob::VERSION), nonce: true %>
+  <%= javascript_include_tag chartjs_url(format: :js, v: GoodJob::VERSION), nonce: true %>
+  <%= javascript_include_tag scripts_url(format: :js, v: GoodJob::VERSION), nonce: true %>
 
-  <%= javascript_include_tag rails_ujs_path(format: :js, v: GoodJob::VERSION), nonce: true %>
+  <%= javascript_include_tag rails_ujs_url(format: :js, v: GoodJob::VERSION), nonce: true %>
 </head>
 <body>
   <nav class="navbar navbar-expand-lg navbar-light bg-light">

data/lib/good_job/adapter.rb

@@ -27,7 +27,7 @@ module GoodJob
     # @param queues [String, nil] determines which queues to execute jobs from when +execution_mode+ is set to +:async+. See {file:README.md#optimize-queues-threads-and-processes} for more details on the format of this string. You can also set this with the environment variable +GOOD_JOB_QUEUES+. Defaults to +"*"+.
     # @param poll_interval [Integer, nil] sets the number of seconds between polls for jobs when +execution_mode+ is set to +:async+. You can also set this with the environment variable +GOOD_JOB_POLL_INTERVAL+. Defaults to +1+.
     # @param start_async_on_initialize [Boolean] whether to start the async scheduler when the adapter is initialized.
-    def initialize(execution_mode: nil, queues: nil, max_threads: nil, poll_interval: nil, start_async_on_initialize: Rails.application.initialized?)
+    def initialize(execution_mode: nil, queues: nil, max_threads: nil, poll_interval: nil, start_async_on_initialize: GoodJob.async_ready?)
       @configuration = GoodJob::Configuration.new(
         {
           execution_mode: execution_mode,

data/lib/good_job/dependencies.rb

@@ -0,0 +1,50 @@
+# frozen_string_literal: true
+
+module GoodJob # :nodoc:
+  # Extends GoodJob module to track Rails boot dependencies.
+  module Dependencies
+    extend ActiveSupport::Concern
+
+    included do
+      # @!attribute [rw] _rails_after_initialize_hook_called
+      #   @!scope class
+      #   Whether Railtie.after_initialize has been called yet (default: +false+).
+      #   This will be set on  but before +Rails.application.initialize?+ is +true+.
+      #   @return [Boolean]
+      mattr_accessor :_rails_after_initialize_hook_called, default: false
+
+      # @!attribute [rw] _active_job_loaded
+      #   @!scope class
+      #   Whether ActiveJob has loaded (default: +false+).
+      #   @return [Boolean]
+      mattr_accessor :_active_job_loaded, default: false
+
+      # @!attribute [rw] _active_record_loaded
+      #   @!scope class
+      #   Whether ActiveRecord has loaded (default: +false+).
+      #   @return [Boolean]
+      mattr_accessor :_active_record_loaded, default: false
+    end
+
+    class_methods do
+      # Whether GoodJob's  has been initialized as of the calling of +Railtie.after_initialize+.
+      # @return [Boolean]
+      def async_ready?
+        Rails.application.initialized? || (
+           _rails_after_initialize_hook_called &&
+           _active_job_loaded &&
+           _active_record_loaded
+         )
+      end
+
+      def start_async_adapters
+        return unless async_ready?
+
+        GoodJob::Adapter.instances
+                        .select(&:execute_async?)
+                        .reject(&:async_started?)
+                        .each(&:start_async)
+      end
+    end
+  end
+end

data/lib/good_job/filterable.rb

@@ -34,9 +34,18 @@ module GoodJob
         next if query.blank?
 
         tsvector = "(to_tsvector('english', serialized_params) || to_tsvector('english', id::text) || to_tsvector('english', COALESCE(error, '')::text))"
-        where("#{tsvector} @@ to_tsquery(?)", query)
-          .order(sanitize_sql_for_order([Arel.sql("ts_rank(#{tsvector}, to_tsquery(?))"), query]) => 'DESC')
+        to_tsquery_function = database_supports_websearch_to_tsquery? ? 'websearch_to_tsquery' : 'plainto_tsquery'
+        where("#{tsvector} @@ #{to_tsquery_function}(?)", query)
+          .order(sanitize_sql_for_order([Arel.sql("ts_rank(#{tsvector}, #{to_tsquery_function}(?))"), query]) => 'DESC')
       end)
     end
+
+    class_methods do
+      def database_supports_websearch_to_tsquery?
+        return @_database_supports_websearch_to_tsquery if defined?(@_database_supports_websearch_to_tsquery)
+
+        @_database_supports_websearch_to_tsquery = connection.postgresql_version >= 110000
+      end
+    end
   end
 end

data/lib/good_job/notifier.rb

@@ -16,9 +16,6 @@ module GoodJob # :nodoc:
 
     include Notifier::ProcessRegistration
 
-    # Raised if the Database adapter does not implement LISTEN.
-    AdapterCannotListenError = Class.new(StandardError)
-
     # Default Postgres channel for LISTEN/NOTIFY
     CHANNEL = 'good_job'
     # Defaults for instance of Concurrent::ThreadPoolExecutor
@@ -129,8 +126,6 @@ module GoodJob # :nodoc:
     # @!visibility private
     # @return [void]
     def listen_observer(_time, _result, thread_error)
-      return if thread_error.is_a? AdapterCannotListenError
-
       if thread_error
         GoodJob._on_thread_error(thread_error)
         ActiveSupport::Notifications.instrument("notifier_notify_error.good_job", { error: thread_error })
@@ -214,6 +209,15 @@ module GoodJob # :nodoc:
         raw_connection.wait_for_notify(WAIT_INTERVAL) do |channel, _pid, payload|
           yield(channel, payload)
         end
+      elsif raw_connection.respond_to?(:jdbc_connection)
+        raw_connection.execute_query("SELECT 1")
+        notifications = raw_connection.jdbc_connection.getNotifications
+        Array(notifications).each do |notification|
+          channel = notification.getName
+          payload = notification.getParameter
+          yield(channel, payload)
+        end
+        sleep WAIT_INTERVAL
       else
         sleep WAIT_INTERVAL
       end

data/lib/good_job/railtie.rb

@@ -34,8 +34,27 @@ module GoodJob
     end
 
     initializer "good_job.start_async" do
+      # This hooks into the hookable places during Rails boot, which is unfortunately not Rails.application.initialized?
+      # If an Adapter is initialized during boot, we want to want to start its async executors once the framework dependencies have loaded.
+      # When exactly that happens is out of our control because gems or application code may touch things earlier than expected.
+      # For example, as of Rails 6.1, if an ActiveRecord model is touched during boot, that triggers ActiveRecord to load,
+      # which touches DestroyAssociationAsyncJob, which loads ActiveJob, which may initialize a GoodJob::Adapter, all of which
+      # happens _before_ ActiveRecord finishes loading. GoodJob will deadlock if an async executor is started in the middle of
+      # ActiveRecord loading.
+
       config.after_initialize do
-        GoodJob::Adapter.instances.each(&:start_async)
+        ActiveSupport.on_load(:active_record) do
+          GoodJob._active_record_loaded = true
+          GoodJob.start_async_adapters
+        end
+
+        ActiveSupport.on_load(:active_job) do
+          GoodJob._active_job_loaded = true
+          GoodJob.start_async_adapters
+        end
+
+        GoodJob._rails_after_initialize_hook_called = true
+        GoodJob.start_async_adapters
       end
     end
   end

data/lib/good_job/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 module GoodJob
   # GoodJob gem version.
-  VERSION = '2.8.1'
+  VERSION = '2.9.3'
 end

data/lib/good_job.rb

@@ -18,6 +18,8 @@ require "good_job/railtie"
 #
 # +GoodJob+ is the top-level namespace and exposes configuration attributes.
 module GoodJob
+  include GoodJob::Dependencies
+
   DEFAULT_LOGGER = ActiveSupport::TaggedLogging.new(ActiveSupport::Logger.new($stdout))
 
   # @!attribute [rw] active_record_parent_class

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: good_job
 version: !ruby/object:Gem::Version
-  version: 2.8.1
+  version: 2.9.3
 platform: ruby
 authors:
 - Ben Sheldon
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2022-01-03 00:00:00.000000000 Z
+date: 2022-01-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activejob
@@ -421,6 +421,7 @@ files:
 - lib/good_job/cron_manager.rb
 - lib/good_job/current_thread.rb
 - lib/good_job/daemon.rb
+- lib/good_job/dependencies.rb
 - lib/good_job/execution.rb
 - lib/good_job/execution_result.rb
 - lib/good_job/filterable.rb