paulwong

在SPRING DATA MONGODB中使用聚合统计查询

在SQL语句中如要做统计一般是这种方式:
SELECT ..,SUM(1)
FROM ..
WHERE ..
GROUP BY ..
HAVING ..
SORT ..


在MONGODB中的架构图



在SPRING DATA MONGODB中是这样写的:
public class VideoRepositoryImpl implements VideoRepositoryCustom{
    
    private static Logger logger = LoggerFactory.getLogger(VideoRepositoryImpl.class);
    
    @Autowired
    private MongoTemplate mongoTemplate;
    

    public List<Cat1UpdateCount> getVideoWithUpdateFrag(List<String> importantCat1List) {
        
        logger.info(new Date().toString());
        
        /**
         * db.videos.aggregate(
            [
               { $match: { "frags.isnew" : true } },
               { $unwind: "$frags" },
               { $match: { "frags.isnew" : true } },
               { $group: { 
                           _id: {cat1:"$cat1"},
                           count: { $sum: 1 },
                           publishdate2: { $max: "$publishdate"}
                         }
               }
               
            ]
            )
         
*/
        Aggregation agg = newAggregation(
                project("frags","cat1","publishdate"),//挑选所需的字段
                match(
                        Criteria.where("frags.isnew").is(Boolean.TRUE)
                        .and("cat1").in(importantCat1List)
                     ),//筛选符合条件的记录
                unwind("frags"),//如果有MASTER-ITEM关系的表,需同时JOIN这两张表的,展开子项LIST,且是内链接,即如果父和子的关联ID没有的就不会输出
                match(Criteria.where("frags.isnew").is(Boolean.TRUE)),
                group("cat1")//设置分组字段
                    .count().as("updateCount")//增加COUNT为分组后输出的字段
                    .last("publishdate").as("publishDate"),//增加publishDate为分组后输出的字段
                project("publishDate","cat1","updateCount")//重新挑选字段
                    .and("cat1").previousOperation()//为前一操作所产生的ID FIELD建立别名
            );

            AggregationResults<Cat1UpdateCount> results = mongoTemplate.aggregate(agg, Video.COLLECTION_NAME, Cat1UpdateCount.class);
            List<Cat1UpdateCount> cat1UpdateCountList = results.getMappedResults();
        
        return cat1UpdateCountList;
    }

}

其中frags的数据类型是LIST

Cat1UpdateCount.java
import java.io.Serializable;

public class Cat1UpdateCount implements Serializable{

    private static final long serialVersionUID = 4240876746984930098L;
    
    private String cat1;
    
    private int updateCount;
    
    private String publishDate;

    public String getCat1() {
        return cat1;
    }

    public void setCat1(String cat1) {
        this.cat1 = cat1;
    }

    public int getUpdateCount() {
        return updateCount;
    }

    public void setUpdateCount(int updateCount) {
        this.updateCount = updateCount;
    }

    public String getPublishDate() {
        return publishDate;
    }

    public void setPublishDate(String publishDate) {
        this.publishDate = publishDate;
    }

    public String toString() {
        return "Cat1UpdateCount [cat1=" + cat1 + ", updateCount=" + updateCount
                + ", publishDate=" + publishDate + "]";
    }

}

参考:
http://docs.spring.io/spring-data/data-mongodb/docs/current/reference/htmlsingle/#mongo.group

http://docs.mongodb.org/manual/reference/operator/aggregation/group/
http://docs.mongodb.org/manual/reference/sql-aggregation-comparison/

posted on 2013-12-24 09:44 paulwong 阅读(11651) 评论(0)  编辑  收藏 所属分类: SPRINGMONGODB


只有注册用户登录后才能发表评论。


网站导航: