Question

我正在使用JPA EclipseLink（Java和MySQL服务器），我希望使用非独占或某些“一对一”之间的查询来执行查询。相同属性（主键）上的关系：

public class elements implements Serializable {
private static final long serialVersionUID = 1L;

@OneToOne(cascade = CascadeType.ALL)
@JoinColumn(name = "element_id", unique = true, nullable = true, insertable = false, updatable = false)
private Element eFeatures1;

@OneToOne(cascade = CascadeType.ALL)
@JoinColumn(name = "element_id", unique = true, nullable = true, insertable = false, updatable = false)
private Element eFeatures2;

@OneToOne(cascade = CascadeType.ALL)
@JoinColumn(name = "element_id", unique = true, nullable = true, insertable = false, updatable = false)
private NewCompoundsLM eFeatures3;

@Id
@Basic(optional = false)
@NotNull
@Size(min = 1, max = 10)
@Column(name = "element_id")
private int elementId;

... some more attributes
}

public class features1 implements Serializable {
 private static final long serialVersionUID = 1L;

@Id
@Column(name = "element_id")
private int elementId;

@Size(max = 20)
@NotNull
@Column(name = "kegg_id")
private String keggId;
... some more attributes
}

public class features2 implements Serializable {
 private static final long serialVersionUID = 1L;

@Id
@Column(name = "element_id")
private int elementId;

@Size(max = 20)
@NotNull
@Column(name = "hmdb_id")
private String hmdbId;
... some more attributes
}
... some more attributes
}

public class features3 implements Serializable {
 private static final long serialVersionUID = 1L;

@Id
@Column(name = "element_id")
private int elementId;
... some more attributes
}

将element_id作为表features1，features2和features3

中的外键

之后，我想要检索表eFeatures1或eFeatures2或eFeatures3中存在的元素，使用非独占或，但我不能执行它

我尝试了下一个查询：

SELECT e FROM elements e 
WHERE (e.features1.keggId!='' or e.features2.hmdbId!='') 
and (e.mass >= 236.08001738701788 and e.mass <= 236.0941826129821);

但是，当我运行时，我获得的结果是e.efetarues1.keggId！=＆＃39;＆＃39; AND e.efetarues2.hmdbId！=＆＃39;＆＃39;，结果只有一个元素。我想要检索具有e.efeatures1.keggId！=＆＃39;＆＃39;的3个元素，这3个元素具有e.efeatures2.hmdbId！=＆＃39;＆＃39;和达到两个条件的1个元素。有没有办法用JPA EclipseLink获得它？

我在SQL中做过，我想在JPA中执行此查询：

SELECT c.compound_id,ck.kegg_id,ch.hmdb_id FROM compounds c 
LEFT JOIN compounds_kegg ck 
on c.compound_id=ck.compound_id 
left join compounds_hmdb ch 
on c.compound_id=ch.compound_id 
where (ck.kegg_id is not null or ch.hmdb_id is not null) 
and (c.mass >= 236.08001738701788 and c.mass <= 236.0941826129821);

+-------------+---------+-----------+
| compound_id | kegg_id | hmdb_id   |
+-------------+---------+-----------+
|       50303 | NULL    | AAAA      |
|       68206 | NULL    | BBBB      |
|       89758 | AAAAAA  | NULL      |
|       95110 | BBBBBB  | NULL      |
|       91763 | CCCCCC  | NULL      |
|       71801 | DDDDDD  | EEEE      |
+-------------+---------+-----------+

但是，当我在JPA中运行查询时：

SELECT e FROM elements e 
WHERE (e.features1.keggId!='' or e.features2.hmdbId!='') 
and (e.mass >= 236.08001738701788 and e.mass <= 236.0941826129821);

查询仅返回71801

此外，我跑的时候

SELECT e FROM elements e 
WHERE (e.ncKegg is not null or nc.ncHMDB is not null) 
and (nc.mass >= 236.08001738701788 and nc.mass <= 236.0941826129821)

查询返回满足质量的最后一个条件的所有元素（8个元素而不是6个元素）。

已编辑：我已经使用了查询

SELECT c FROM NewCompounds c 
LEFT JOIN NewCompoundsKegg ck on c.compoundId=ck.compoundId 
LEFT JOIN NewCompoundsHMDB ch on c.compoundId=ch.compoundId 
where (ck.keggId is not null or ch.hmdbId is not null) 
and (c.mass >= 236.08001738701788 and c.mass <= 236.0941826129821)

但是我不知道为什么我必须在JPA中使用左连接，如果我已经在主元素中声明了关系。另外，我不知道它如何影响应用程序的性能以包括3个左连接

提前致谢。

Answer 1

这是一个可能的解决方案：

{{1}}

但是我不知道为什么我必须在JPA中使用左连接，如果我已经在主元素中声明了关系。另外，我不知道它如何影响应用程序的性能以包括3个左连接

JPA或条件超过oneToOne关系

1 个答案: