Java Hibernate延迟加载与显式查询

我有一个数据库性能问题

假设我的数据模型如下所示

UserTable.java

@Id
@Column(name = "USER_ID", nullable = false)
private Long userId;

// other user fields

@OneToMany(mappedBy = "user", cascade = CascadeType.ALL, orphanRemoval = true, fetch = FetchType.LAZY)
private Set<ProfilePhotoTable> photos = new HashSet<>();

@OneToMany(mappedBy = "user", cascade = CascadeType.ALL, orphanRemoval = true, fetch = FetchType.LAZY)
private Set<AlbumPhotoTable> photos = new HashSet<>();

@Id
@Column(name = "PHOTO_ID", nullable = false)
private Long photoId;

// other photo fields

@ManyToOne(fetch = FetchType.LAZY)
@JoinColumn(name = "USER_ID")
private UserTable user;

@Entity
@Table(name = "PROFILEPHOTO")
@PrimaryKeyJoinColumn(name = "PROFILEPHOTO_ID", referencedColumnName = "PHOTO_ID")
public class ProfilePhotoTable extends PhotoTable

@Entity
@Table(name = "ALBUMPHOTO")
@PrimaryKeyJoinColumn(name = "ALBUMPHOTO_ID", referencedColumnName = "PHOTO_ID")
public class ProfilePhotoTable extends PhotoTable

ProfilePhotoTable.java

@Id
@Column(name = "USER_ID", nullable = false)
private Long userId;

// other user fields

@OneToMany(mappedBy = "user", cascade = CascadeType.ALL, orphanRemoval = true, fetch = FetchType.LAZY)
private Set<ProfilePhotoTable> photos = new HashSet<>();

@OneToMany(mappedBy = "user", cascade = CascadeType.ALL, orphanRemoval = true, fetch = FetchType.LAZY)
private Set<AlbumPhotoTable> photos = new HashSet<>();

@Id
@Column(name = "PHOTO_ID", nullable = false)
private Long photoId;

// other photo fields

@ManyToOne(fetch = FetchType.LAZY)
@JoinColumn(name = "USER_ID")
private UserTable user;

@Entity
@Table(name = "PROFILEPHOTO")
@PrimaryKeyJoinColumn(name = "PROFILEPHOTO_ID", referencedColumnName = "PHOTO_ID")
public class ProfilePhotoTable extends PhotoTable

@Entity
@Table(name = "ALBUMPHOTO")
@PrimaryKeyJoinColumn(name = "ALBUMPHOTO_ID", referencedColumnName = "PHOTO_ID")
public class ProfilePhotoTable extends PhotoTable

AlbumPhotoTable.java

@Id
@Column(name = "USER_ID", nullable = false)
private Long userId;

// other user fields

@OneToMany(mappedBy = "user", cascade = CascadeType.ALL, orphanRemoval = true, fetch = FetchType.LAZY)
private Set<ProfilePhotoTable> photos = new HashSet<>();

@OneToMany(mappedBy = "user", cascade = CascadeType.ALL, orphanRemoval = true, fetch = FetchType.LAZY)
private Set<AlbumPhotoTable> photos = new HashSet<>();

@Id
@Column(name = "PHOTO_ID", nullable = false)
private Long photoId;

// other photo fields

@ManyToOne(fetch = FetchType.LAZY)
@JoinColumn(name = "USER_ID")
private UserTable user;

@Entity
@Table(name = "PROFILEPHOTO")
@PrimaryKeyJoinColumn(name = "PROFILEPHOTO_ID", referencedColumnName = "PHOTO_ID")
public class ProfilePhotoTable extends PhotoTable

@Entity
@Table(name = "ALBUMPHOTO")
@PrimaryKeyJoinColumn(name = "ALBUMPHOTO_ID", referencedColumnName = "PHOTO_ID")
public class ProfilePhotoTable extends PhotoTable

现在让我们假设我想写一个查询来获取用户的所有照片——所有个人资料照片和所有相册照片

但是，我不希望每次从数据库请求用户信息时都提取照片，这就是为什么我在照片字段上指定了

fetch=FetchType.LAZY

基本上，我对这两种方法有所怀疑

两个独立的查询，第一个查询通过ID从数据库中获取

UserTable

，第二个查询获取照片，类似于

SELECT*from Photo，其中userId=：userId

一个查询，将

连接fetch

用户和相应的照片。但是，我不太确定这个查询将如何进行，因为相册和个人资料中的照片是分开的。我发现在这里面

comments=entityManager.createQuery(

“选择电脑”+ “来自后现代个人电脑”+ “加入获取pc.post”+ “where pc.review=：review”，PostComment.class） .setParameter（“审查”，审查） .getResultList（）

应该使用，但我不确定如何将其应用到我的用例中

---

问题是，哪种方法在性能方面更好，如果是第二种方法，那么查询的结构应该如何？

我认为答案将取决于用户查询中也查询照片的部分。如果答案是一小部分，那么让它保持懒惰是有意义的，这样你就不必一直拉它们。

---

无论如何，我不会将照片存储在传统的关系数据库类型的设置中。我会把它们放在某种对象存储中，只使用RDBMS保存ID

如果您使用的是JPA 2.1或更高版本，您可以使用@NamedEntityGraph来控制，如果您想获取“急切”或“懒惰”的数据

---

这篇文章解释得很好：

。你可以让

用户渴望（“不懒惰”）（实现你想要的（据我所知）.“懒散”在联想的“N:-”方面最有意义）