hibernate 复合主键二级缓存问题

0 0

hibernate 复合主键二级缓存问题10

大家好，对于hibernate二级缓存使用，以下使用情况，想请教一下：

1. 使用复合主键

2. 通过1对多关联获取相关的数据

如“工作组” 与 “角色”的关系，一个工作组对应多个角色，而且工作组与角色的对应之间存在其它属性，如描述等：

工作组代码:

@Entity
@Cache(usage=CacheConcurrencyStrategy.READ_WRITE)
@Table(name = "sys_work_group")
public class WorkGroup implements Serializable {

	/**
	 * 
	 */
	private static final long serialVersionUID = -4016855062656988636L;
	@Id
	@Column(name = "workgroup_id", nullable = false)
	@GenericGenerator(name = "generator", strategy = "identity")
	@GeneratedValue(generator = "generator")
	private Integer workgroup_id;

	@OneToMany(cascade = { CascadeType.MERGE,CascadeType.REMOVE }, mappedBy = "workGroupRolePK.workGroup", targetEntity = WorkGroupRole.class, fetch = FetchType.LAZY)
	
	private Set<WorkGroupRole> workGroupRoles;

	@OneToMany(mappedBy = "workGroup",fetch=FetchType.LAZY)
	private Set<User> users;

	@Column(name = "name", nullable = false)
	private String name;

	@Column(name = "descri")
	private String descri;

	// 省略......
}

角色类代码:

@Entity
@Cache(usage = CacheConcurrencyStrategy.READ_WRITE)
@Table(name = "sys_role")
public class Role implements Serializable {

	/**
	 * 
	 */
	private static final long serialVersionUID = 9055661647747413322L;

	@Id
	@Column(name = "roleid", nullable = false)
	//设置ID生成方式
	@GenericGenerator(name = "generator", strategy = "identity")
	//返回自增ID
	@GeneratedValue(generator = "generator")
	private Integer roleid;

	/**
	 * 角色名称
	 */
	@Column(name = "name", nullable = false)
	private String name;

	/**
	 * 排序ID
	 */
	@Column(name = "orderid")
	private Integer orderid;

	/**
	 * 备注
	 */
	@Column(name = "descri")
	private String descri;

	@ManyToMany(targetEntity = Menu.class, fetch = FetchType.LAZY)
	@JoinTable(name = "sys_role_menu", joinColumns = { @JoinColumn(name = "roleid") }, inverseJoinColumns = { @JoinColumn(name = "mid") })
	@Cache(region="cn.cpic.po.sys.Role.menus",usage=CacheConcurrencyStrategy.READ_WRITE)
	private Set<Menu> menus = new HashSet<Menu>();

	//@OneToMany(mappedBy = "role", fetch = FetchType.EAGER)
	@Transient
	private Set<User> users = new HashSet<User>();

	/**
	 * 用来保存资源与角色之间的映射关系
	 */
	@OneToMany(mappedBy = "roleResUrlPK.role", targetEntity = RoleResUrl.class, fetch = FetchType.LAZY, cascade = { CascadeType.PERSIST, CascadeType.MERGE, CascadeType.REMOVE })
	@Cache(region="cn.cpic.po.sys.RoleResUrl",usage=CacheConcurrencyStrategy.READ_WRITE)
	private Set<RoleResUrl> roleResUrls = new HashSet<RoleResUrl>();

	public void setRoleid(Integer roleid) {
		this.roleid = roleid;
	}
        //// 省略
}

工作组与角色的对应类:

@Entity
@Cache(usage=CacheConcurrencyStrategy.READ_WRITE)
@Table(name = "sys_workgroup_role")
public class WorkGroupRole implements Serializable {

	/**
	 * 
	 */
	private static final long serialVersionUID = 5723834544769525469L;

	@Id
	@Cache(region="cn.cpic.po.sys.WorkGroupRole.workGroupRolePK",usage=CacheConcurrencyStrategy.READ_WRITE)
	private WorkGroupRolePK workGroupRolePK;

	@Column(name = "descri")
	private String descri;

	@Transient
	private boolean check;
        // getter and setter 省略

}

以及主键类@Embeddable
public class WorkGroupRolePK implements Serializable {

	/**
	 * 
	 */
	private static final long serialVersionUID = 1572066453764922900L;

	@ManyToOne(targetEntity = WorkGroup.class, cascade = { CascadeType.PERSIST, CascadeType.MERGE })
	@JoinColumn(name = "workgroup_id", nullable = false)
	private WorkGroup workGroup;

	@ManyToOne(targetEntity = Role.class, cascade = { CascadeType.PERSIST, CascadeType.MERGE })
	@JoinColumn(name = "roleid", nullable = false)
	private Role role;

	public WorkGroupRolePK() {

	}

	public WorkGroupRolePK(WorkGroup workGroup, Role role) {
		this.setWorkGroup(workGroup);
		this.setRole(role);
	}
//// getter and setter 省略
	
	@Override
	public boolean equals(Object obj) {
		if (!(obj instanceof RoleResUrlPK)) {
			return false;
		}
		WorkGroupRolePK workGroupRolePK = (WorkGroupRolePK) obj;

		if ((workGroupRolePK.getRole() == null || this.role == null) || (workGroupRolePK.getWorkGroup() == null || this.getWorkGroup() == null)) {
			return false;
		}

		return (this.getRole().getRoleid() == workGroupRolePK.getRole().getRoleid()) && (this.getWorkGroup().getWorkgroup_id() == workGroupRolePK.getWorkGroup().getWorkgroup_id());
	}

	@Override
	public int hashCode() {
		int result = 51;
		result = 25 * result + (this.role == null ? result : this.role.getRoleid());
		result = 25 * result + (this.workGroup == null ? result : this.workGroup.getWorkgroup_id());
		return result;
	}

}

代码如上配置后，当以下调用：

Set<WorkGroupRole> workGroupRoles = workGroup.getWorkGroupRoles();

if (!CollectionUtils.isEmpty(workGroupRoles)) {
	for (WorkGroupRole workGroupRole : workGroupRoles) {
              //............
        }
}

按我的理解，这里第二次调用应该是不会发出sql了的，但实际上，每一次调用都会发sql到数据库查询，不知道是我对hibernate二级缓存理解错了，还是配置，或者代码有不对，不知道哪位知道这其中的问题，谢谢~

Hibernate: select workgroupr0_.roleid as roleid29_0_, workgroupr0_.workgroup_id as workgroup3_29_0_, workgroupr0_.descri as descri29_0_ from sys_workgroup_role workgroupr0_ where workgroupr0_.roleid=? and workgroupr0_.workgroup_id=?
Hibernate: select workgroupr0_.roleid as roleid29_0_, workgroupr0_.workgroup_id as workgroup3_29_0_, workgroupr0_.descri as descri29_0_ from sys_workgroup_role workgroupr0_ where workgroupr0_.roleid=? and workgroupr0_.workgroup_id=?
.............................. 省略

Hibernate 二级缓存复合主键

2012年11月22日 15:29

cnblsp2
10
0 0 2

2个答案按时间排序按投票排序

0 0

我也遇到这个问题，有没有最终的解决方案呢？

2015年7月24日 16:29

penwei
0
0 0 6

添加评论

0 0

Set<WorkGroupRole> workGroupRoles = workGroup.getWorkGroupRoles();

if (!CollectionUtils.isEmpty(workGroupRoles)) {
for (WorkGroupRole workGroupRole : workGroupRoles) {
//............
}
}

你说的第二次调用指的是第二次调用上面的整个代码段？还是指的第二次for循环？

上面的代码段如果是第一次执行的话，肯定是要通过SQL从数据库中取数据的。
第二次执行上面的代码段的话，应该就会直接从缓存取数据了。

如果你说的是for循环的第二次的话，因为每次循环对应不同的数据，肯定要每次执行一个sql的。

2012年11月23日 08:15