ruby-on-rails - Rails 的模型 Hook 会等到数据库事务完成吗？

Question

我有一个带有 after_save 的模型和 after_commit钩。 after_save钩子(Hook)在数据库中存储了after_commit所需的一些信息。钩。
假设所有数据库事务在 after_commit 之前完成是否安全？钩子(Hook)着火了？
这里有一个小例子来说明这个问题:

class TicketComment < ApplicationRecord
  after_save :extract_mentions
  after_commit :notify

  def extract_mentions
    current_mentions = mentions.map(&:user).map(&:username)
    new_mentions = description.scan(/(?<=^|(?<=[^a-zA-ZÀ-ž0-9_-]))@([a-zA-ZÀ-ž]+[a-zA-ZÀ-ž0-9_]+)/).flatten

    mentions_to_add = new_mentions - current_mentions
    mentions_to_remove = current_mentions - new_mentions

    users_to_add = User.select(:id).where("username IN (?)", mentions_to_add.flatten).map(&:id)
    users_to_remove = User.select(:id).where("username IN (?)", mentions_to_remove.flatten).map(&:id)

    users_to_add&.each do |id_to_add|
      mentions.create(user_id: id_to_add)
    end

    mentions.where(user_id: users_to_remove).destroy_all
  end

  def notify
    user_ids = ticket.participants.pluck(:id) - [Current.user.id]
    TicketCommentMailer.comment_added(id, user_ids).deliver_later
  end
end

请注意，此示例已简化。 extract_mentions -part 是跨多个模型使用的模型问题，因此我无法运行 notify extract_mentions 中的代码-钩。

Answer 1

用一些引用来扩展我的评论，是的，您可以假设所有数据库事务都在您的 #after_commit 之前完成。回调运行。这是 active_record/transaction.rb 中的评论(github):

#after_commit callbacks are called on every record saved or destroyed within atransaction immediately after the transaction is committed. #after_rollback callbacksare called on every record saved or destroyed within a transaction immediately after thetransaction or savepoint is rolled back.

These callbacks are useful for interacting with other systems since you will be guaranteedthat the callback is only executed when the database is in a permanent state.

事务代码有点复杂，因为它处理嵌套事务、保存点等。但最终，在 TransactionManager.commit_transaction 中( active_record/connect_adapters/abstract/transaction.rb )，管理器确实在运行 #after_commit 之前执行了特定于数据库的提交操作回调与父事务关联的所有唯一记录 ID。

def commit_transaction
  @connection.lock.synchronize do
    ...
    transaction.commit  # executes database-specific commit action
    transaction.commit_records # executes after_commit callbacks
  end
end

所有保存和销毁操作，包括它们的持久性回调，都隐式包装在事务 block 中。根据 Rails 中的另一条评论 active_record/transaction.rb :

#transaction calls can be nested. By default, this makes all databasestatements in the nested transaction block become part of the parenttransaction.

因此， #create 生成的数据库语句和 #destroy_all在您的 #after_save回调与原始父事务同时提交，并且可以依赖于 #after_commit打回来。一个简化的例子来说明:

class Foo < ApplicationRecord
  after_save :do_stuff
  after_commit :check

  def do_stuff
    Bar.create!(name: 'bar')
  end

  def check
    Rails.logger.info "Bar count: #{Bar.count}"
    Rails.logger.info "name: #{Bar.first.name}"
  end
end

输出日志:

> Foo.create!(name: "Jerry")
  (0.0ms)  begin transaction
  Foo Create (0.3ms)  INSERT INTO "foos" ("name", "created_at", "updated_at") VALUES (?, ?, ?)  [["name", "Jerry"], ["created_at", "2020-09-16 01:49:57.480367"], ["updated_at", "2020-09-16 01:49:57.480367"]]
  Bar Create (0.2ms)  INSERT INTO "bars" ("name", "created_at", "updated_at") VALUES (?, ?, ?)  [["name", "bar"], ["created_at", "2020-09-16 01:49:57.485953"], ["updated_at", "2020-09-16 01:49:57.485953"]]
 (2.3ms)  commit transaction
 (0.1ms)  SELECT COUNT(*) FROM "bars"
Bar count: 1
  Bar Load (0.1ms)  SELECT "bars".* FROM "bars" ORDER BY "bars"."id" ASC LIMIT ?  [["LIMIT", 1]]
bar

请注意，当然任何 #after_commit对于在事务中创建、更新或销毁的任何模型，都会调用回调。所以，如果 Bar还定义了 #after_commit回调，那些也会运行。