Spring Data Jpa返回自定义对象
我是坏人哦 人气:0tasks表对应的Entity
@Entity @NoArgsConstructor @AllArgsConstructor @Table(name = "tasks") @Data public class Tasks extends BaseEntity { @Id @GeneratedValue(strategy = GenerationType.IDENTITY) private int taskId; private Integer websiteId; private String status; private String lastOperator; private String lastOperationTime; private String jobId; private int retryTimes; }
websites表对应的Entity
@Entity @Table(name = "websites") @Data @NoArgsConstructor @AllArgsConstructor public class Websites extends BaseEntity{ @Id @GeneratedValue(strategy = GenerationType.IDENTITY) private int websiteId; private String country; private String websiteName; private String baseUrl; private String smartphoneUrl; private String tabletUrl; private String smartdeviceUrl; private String websiteCategory; private String webtypeRefreshRate; private String urlRefreshRate; private String configTime; private String configDesc; }
自定义对象
@Data @NoArgsConstructor @AllArgsConstructor public class CustomizedDto implements Serializable { private static final long serialVersionUID = -7242005560621561106L; private String country; private String websiteName; private String baseUrl; private String status; }
方法一、简单查询直接new对象
使用hql,将结果返回到new出来的自定义对象中,注意,自定义对象中要有对应的构造函数。
@Query(value = "select new com.bigdata.mrcrawler.dto.CustomizedDto(w.country,w.websiteName,w.baseUrl,t.status) " + "from Websites w join Tasks t on w.id=t.websiteId " + "where t.status=?1") List<CustomizedDto> getByStatus1(String status);
但是这个方法只使用于简单的查询语句,如果遇到复杂查询需要使用原生SQL(即nativeQuery=true)时, 此方法不适用,会抛出如下异常。
@Query(value = "select w.country as country,w.website_name as websiteName,w.base_url as baseUrl,t.status as status " + "from websites w join tasks t on w.website_id=t.website_id " + "where t.status=?1",nativeQuery = true) List<CustomizedDto> getByStatus2(String status);
No converter found capable of converting from type [org.springframework.data.jpa.repository.query.AbstractJpaQuery$TupleConverter$TupleBackedMap] to type [com.bigdata.mrcrawler.dto.CustomizedDto]
方法二、Service层使用EntityManager
直接在Service层使用EntityManager进行查询,可以自由组装各种复杂sql。
public List<CustomizedDto> t(String param) { String sql = "select w.country as country,w.website_name as websiteName,w.base_url as baseUrl,t.status as status " + "from websites w join tasks t on w.website_id=t.website_id " + "where t.status='" + param + "'"; List<CustomizedDto> res = entityManager.createNativeQuery(sql).getResultList(); return res; }
但是会有sql注入问题,例如:param传入Running' or 1=1 --\t 上述查询会将查询整个数据表。解决方法如下,使用预编译防止sql注入问题。
public List<CustomizedDto> t(String param) { String sql = "select w.country as country,w.website_name as websiteName,w.base_url as baseUrl,t.status as status " + "from websites w join tasks t on w.website_id=t.website_id " + "where t.status=:param" ; Query nativeQuery = entityManager.createNativeQuery(sql); nativeQuery.setParameter("param", param); List<CustomizedDto> res = nativeQuery.getResultList(); return res; }
然而,个人很不喜欢这种代码中嵌入大片大片sql的写法,排查问题的时候看得头疼(别问,问就是被坑过/捂脸.jpg/)。所以方法二即便可行,我私心还是不想推荐。
方法三、Dao层使用Map接收自定义对象
使用List<Map> 接收返回结果,无论是原生sql还是hql都支持,当然也支持分页,只需要把返回对象改为Page<Map>即可,其他操作与普通分页没有差别。
@Query(value = "select w.country as country,w.website_name as websiteName,w.base_url as baseUrl,t.status as status " + "from websites w join tasks t on w.website_id=t.website_id " + "where t.status=?1",nativeQuery = true) List<Map> getByStatus3(String status);
Service层也需要用List<Map>进行接收。
public List<Map> t(String param) { return testRepository.getByStatus3(param); }
如果后续还有其他操作,还是需要转成自定义对象怎么办,毕竟Map操作起来挺麻烦的。可以用如下解决方案,先用Object进行接收,然后强转成自定义对象List。
public List<CustomizedDto> t(String param) { Object data = testRepository.getByStatus3(param); return (List<CustomizedDto>)data; }
强转需要注意自定义对象和数据库中字段类型的强一致性,如数据库中datetime类型,自定义对象对应的字段必须是Date,不能是String,否则转换的时候会有问题,数据会丢失。
总结
加载全部内容