java PersistentObjectException：分离的实体传递给持久化
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/7668804/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
PersistentObjectException: detached entity passed to persist
提问by siamii
In my app I'm connected to a websocket. The socket is sending JSON data. As it is becoming available I process it in a different Job. In each job, I convert the JSON to a model object with GSON. There are 3 types of models, all independent, and with one common timestamp field in a MappedSuperclass. They don't have an id explicit field. I call save on the model and in 1 out of 1000 times a "PersistentObjectException: detached entity passed to persist" is thrown.
在我的应用程序中，我连接到了一个 websocket。套接字正在发送 JSON 数据。当它变得可用时，我会在不同的工作中处理它。在每项工作中，我都使用 GSON 将 JSON 转换为模型对象。有 3 种类型的模型，它们都是独立的，并且在 MappedSuperclass 中有一个公共时间戳字段。他们没有 id 显式字段。我在模型上调用 save 并且在 1000 次中有 1 次抛出“PersistentObjectException：分离的实体传递给持久性”。
The multiple jobs are called from a start up Job that connects to the websocket. The library I'm using creates a different thread for each incoming message. I then convert the thread into a job before saving to the database. I do this, because otherwise another PersistentObjectException is thrown about changing the id from 1 to 2 or something similar, if I allow the original thread to call save.
从连接到 websocket 的启动 Job 调用多个作业。我正在使用的库为每个传入的消息创建一个不同的线程。然后在保存到数据库之前将线程转换为作业。我这样做是因为，如果我允许原始线程调用 save，则会抛出另一个 PersistentObjectException，将 id 从 1 更改为 2 或类似的东西。
I also have another job running that accesses the database at the same time. What could be wrong?
我还有另一个正在运行的同时访问数据库的作业。可能有什么问题？
@Override
public void onMessage(final WebSocketMessage message){
new Job() {
@Override
public void doJob() {
processMessage(message.getText());
}
}.now();
}
public void processMessage(String message) {
Appointment appointment = new Gson().fromJson(message, Appointment.class);
appointment.save();
}
@Entity
public class Appointment extends CalendarEvent {
private String owner;
}
@MappedSuperclass
public abstract class CalendarEvent extends Model {
private long timestamp;
}
EDIT:Added some code sample
编辑：添加了一些代码示例
采纳答案by siamii
I've found this workaround to work for now. Maybe it's to do with how Gson and Jpa interact. Not sure...
我发现这个解决方法现在可以工作。也许这与 Gson 和 Jpa 的交互方式有关。不确定...
public void processMessage(String message) {
Appointment appointment = new Gson().fromJson(message, Appointment.class);
//recreate appointment
appointment = new Appointment(appointment);
appointment.save(); }
回答by millhouse
Basically it means the EntityManagerwon't track them any more: This is a good overview. Howthey come to be considered detached when you didn't explicitly ask for it is odd. But the workaround mightbe to use a "defensive" approach in your DAO method and mergethese odd cases; i.e.:
基本上这意味着EntityManager不会再跟踪它们：这是一个很好的概述。当您没有明确要求时，它们如何被认为是超脱的，这很奇怪。但是解决方法可能是在您的 DAO 方法和merge这些奇怪的情况下使用“防御性”方法；IE：
public void save (Model possiblyDetachedModel) {
if (entityManager.contains(possiblyDetachedModel)) {
entityManager.merge(possiblyDetachedModel);
} else {
entityManager.persist(possiblyDetachedModel);
}
}
I don't particularly like it, because it shouldn't need to be done when you're persisting brand-new objects. It might be worth putting some logging in (or debugging if at all possible) the "merge" branch and really inspecting those rogue objects - I'm pretty sure the EntityManagercan only use the @Idfield as the detection mechanism...
我不是特别喜欢它，因为当你正在persist处理全新的对象时，它不应该需要完成。可能值得在“ merge”分支中进行一些登录（或调试，如果可能的话）并真正检查那些流氓对象 - 我很确定EntityManager只能使用该@Id字段作为检测机制......
