JPA提供了一个SqlResultSetMapping，允许您将本地查询返回的任何内容映射到实体或自定义类。

EDIT JPA 1.0不允许映射到非实体类。只有在JPA 2.1中，才添加了一个ConstructorResult来映射java类的返回值。

同样，对于OP获取计数的问题，定义一个带有单个ColumnResult的结果集映射就足够了

参见下面的示例，使用POJO作为伪实体从本机查询检索结果，而不使用复杂的SqlResultSetMapping。只需要两个注释，一个裸露的@Enity和一个假的@Id在你的POJO。@Id可以用于您选择的任何字段，@Id字段可以有重复的键，但不能为空值。

因为@Enity不映射到任何物理表，所以这个POJO被称为伪实体。

环境:eclipselink 2.5.0-RC1, jpa-2.1.0, mysql-connector-java-5.1.14

您可以在这里下载完整的maven项目

本机查询是基于mysql的样本雇员数据库 http://dev.mysql.com/doc/employee/en/employees-installation.html

persistence . xml

<?xml version="1.0" encoding="UTF-8"?><persistence xmlns="http://xmlns.jcp.org/xml/ns/persistence" 
    xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" version="2.1" 
    xsi:schemaLocation="http://xmlns.jcp.org/xml/ns/persistence http://xmlns.jcp.org/xml/ns/persistence/persistence_2_1.xsd">
<persistence-unit name="jpa-mysql" transaction-type="RESOURCE_LOCAL">
    <class>org.moonwave.jpa.model.pojo.Employee</class>
    <properties>
        <property name="javax.persistence.jdbc.url" value="jdbc:mysql://localhost:3306/employees" />
        <property name="javax.persistence.jdbc.user" value="user" />
        <property name="javax.persistence.jdbc.password" value="***" />
        <property name="javax.persistence.jdbc.driver" value="com.mysql.jdbc.Driver" />
    </properties>
</persistence-unit>

Employee.java

package org.moonwave.jpa.model.pojo;

@Entity
public class Employee {

@Id
protected Long empNo;

protected String firstName;
protected String lastName;
protected String title;

public Long getEmpNo() {
    return empNo;
}
public void setEmpNo(Long empNo) {
    this.empNo = empNo;
}
public String getFirstName() {
    return firstName;
}
public void setFirstName(String firstName) {
    this.firstName = firstName;
}
public String getLastName() {
    return lastName;
}
public void setLastName(String lastName) {
    this.lastName = lastName;
}   
public String getTitle() {
    return title;
}
public void setTitle(String title) {
    this.title = title;
}
public String toString() {
    StringBuilder sb = new StringBuilder();
    sb.append("empNo: ").append(empNo);
    sb.append(", firstName: ").append(firstName);
    sb.append(", lastName: ").append(lastName);
    sb.append(", title: ").append(title);
    return sb.toString();
}
}

EmployeeNativeQuery.java

public class EmployeeNativeQuery {
private EntityManager em;
private EntityManagerFactory emf;

public void setUp() throws Exception {
    emf=Persistence.createEntityManagerFactory("jpa-mysql");
    em=emf.createEntityManager();
}
public void tearDown()throws Exception {
    em.close();
    emf.close();
}

@SuppressWarnings("unchecked")
public void query() {
    Query query = em.createNativeQuery("select e.emp_no as empNo, e.first_name as firstName, e.last_name as lastName," + 
            "t.title from employees e join titles t on e.emp_no = t.emp_no", Employee.class);
    query.setMaxResults(30);
    List<Employee> list = (List<Employee>) query.getResultList();
    int i = 0;
    for (Object emp : list) {
        System.out.println(++i + ": " + emp.toString());
    }
}

public static void main( String[] args ) {
    EmployeeNativeQuery test = new EmployeeNativeQuery();
    try {
        test.setUp();
        test.query();
        test.tearDown();
    } catch (Exception e) {
        System.out.println(e);
    }
}
}

将SQL查询转换为POJO类集合的简单方法，

Query query = getCurrentSession().createSQLQuery(sqlQuery).addEntity(Actors.class);
List<Actors> list = (List<Actors>) query.list();
return list;

首先声明以下注释:

@Target(ElementType.TYPE)
@Retention(RetentionPolicy.RUNTIME)
public @interface NativeQueryResultEntity {
}

@Target(ElementType.FIELD)
@Retention(RetentionPolicy.RUNTIME)
public @interface NativeQueryResultColumn {
    int index();
}

然后按以下方式注释你的POJO:

@NativeQueryResultEntity
public class ClassX {
    @NativeQueryResultColumn(index=0)
    private String a;

    @NativeQueryResultColumn(index=1)
    private String b;
}

然后写注释处理器:

public class NativeQueryResultsMapper {

    private static Logger log = LoggerFactory.getLogger(NativeQueryResultsMapper.class);

    public static <T> List<T> map(List<Object[]> objectArrayList, Class<T> genericType) {
        List<T> ret = new ArrayList<T>();
        List<Field> mappingFields = getNativeQueryResultColumnAnnotatedFields(genericType);
        try {
            for (Object[] objectArr : objectArrayList) {
                T t = genericType.newInstance();
                for (int i = 0; i < objectArr.length; i++) {
                    BeanUtils.setProperty(t, mappingFields.get(i).getName(), objectArr[i]);
                }
                ret.add(t);
            }
        } catch (InstantiationException ie) {
            log.debug("Cannot instantiate: ", ie);
            ret.clear();
        } catch (IllegalAccessException iae) {
            log.debug("Illegal access: ", iae);
            ret.clear();
        } catch (InvocationTargetException ite) {
            log.debug("Cannot invoke method: ", ite);
            ret.clear();
        }
        return ret;
    }

    // Get ordered list of fields
    private static <T> List<Field> getNativeQueryResultColumnAnnotatedFields(Class<T> genericType) {
        Field[] fields = genericType.getDeclaredFields();
        List<Field> orderedFields = Arrays.asList(new Field[fields.length]);
        for (int i = 0; i < fields.length; i++) {
            if (fields[i].isAnnotationPresent(NativeQueryResultColumn.class)) {
                NativeQueryResultColumn nqrc = fields[i].getAnnotation(NativeQueryResultColumn.class);
                orderedFields.set(nqrc.index(), fields[i]);
            }
        }
        return orderedFields;
    }
}

使用上述框架如下:

String sql = "select a,b from x order by a";
Query q = entityManager.createNativeQuery(sql);

List<ClassX> results = NativeQueryResultsMapper.map(q.getResultList(), ClassX.class);

既然其他人已经提到了所有可能的解决方案，我将分享我的变通方案。

在我使用Postgres 9.4的情况下，当我和Jackson一起工作时，

//Convert it to named native query.
List<String> list = em.createNativeQuery("select cast(array_to_json(array_agg(row_to_json(a))) as text) from myschema.actors a")
                   .getResultList();

List<ActorProxy> map = new ObjectMapper().readValue(list.get(0), new TypeReference<List<ActorProxy>>() {});

我相信您可以在其他数据库中找到相同的方法。

另供参考，JPA 2.0本机查询结果为映射

如果希望将自定义查询结果直接映射到实体，而不需要编写任何映射代码，可以尝试这种方式。根据我的经验，这是最方便的方法，但缺点是失去了hibernate ddl-auto的好处:

Disable hibernate validation by removing the hibernate.ddl-auto. If not doing this, hibernate can complain about missing table in database. Create a pojo with @Entity for the custom result set without table mapping, something like: @Getter @Setter @Entity public class MyCustomeResult implements Serializable { @Id private Long id; @Column(name = "name") private String name; } In repository, use the entity to map directly from query.getResultList() public List<MyCustomeResult> findByExampleCustomQuery(Long test) { String sql = "select id, name from examples where id =:test"; Query query = entityManager.createNativeQuery(sql, MyCustomeResult.class); return query.setParameter("test", test).getResultList(); }