在之前的两篇博文《MongoDB基础之原生方法增删改查》和《MongoDB基础之查询表达式》中,介绍了 MongoDB 的基本查询功能,这篇博文我们来介绍一个比较复杂的场景,也就是聚合查询。
3.1 测试数据录入
db.post.insert([
{id:1, title:'PHP从入门到精通', author:'张无忌', category_id:1, tags:['php','后端'], hits:20, created_at:ISODate('2019-03-01 10:20:32')},
{id:2, title:'Linux从学童到出家', author:'张翠山', category_id:1, tags:['linux','服务器'], hits:234, created_at:ISODate('2019-03-12 16:22:50')},
{id:3, title:'MySQL从删库到跑路', author:'张三丰', category_id:2, tags:['mysql','database','数据库'], hits:73, created_at:ISODate('2019-04-01 19:43:20')},
{id:4, title:'利用Python求心理阴影面积', author:'张无忌', category_id:1, tags:['python','后端'], hits:30, created_at:ISODate('2019-04-23 16:20:09')},
{id:5, title:'Java从基础到脸黑', author:'殷素素', category_id:1, tags:['java','后端','性能'], hits:145, created_at:ISODate('2019-05-01 11:43:22')},
{id:6, title:'Mongodb从幼儿园到博士后', author:'赵敏', category_id:2, tags:['mongodb','database','query','数据库'], hits:76, created_at:ISODate('2019-05-01 16:11:07')},
{id:7, title:'用Nodejs来填补心灵空白', author:'张翠山', category_id:3, tags:['nodejs','前端','后端'], hits:203, created_at:ISODate('2019-05-02 18:34:23')},
{id:8, title:'沉迷C++无法自拔', author:'张无忌', category_id:1, tags:['c++','后端','服务器'], hits:54, created_at:ISODate('2019-05-06 13:54:23')},
{id:9, title:'Swift脱单指南', author:'张无忌', category_id:3, tags:['swift','前端'], hits:183, created_at:ISODate('2019-05-10 10:21:12')}
{id:10, title:'Swift脱单指南', author:'宋青书', category_id:3, tags:['swift','前端'], hits:42, created_at:ISODate('2019-05-18 21:32:54')}
]);
db.category.insert([
{id:1, name:'后端编程语言'},
{id:2, name:'数据库'},
{id:3, name:'前端编程语言'}
]);
3.2 管道
$project
db.post.aggregate([
{$project:{title:1, author:1}}
])
db.post.aggregate([
{$project:{title:1, writer:'$author'}}
])
3.2.2 $match
db.post.aggregate([
{$match:{category_id:2}}
])
db.post.aggregate([
{$match:{hits:{$gt:100}}}
])
3.2.3 $group
db.post.aggregate([
{$group:{
_id:'$category_id',
total:{$sum:1}
}}
])
db.post.aggregate([
{$project:{
y_m:{$dateToString:{format:"%Y-%m", date:"$created_at"}}
}},
{$group:{
_id:'$y_m',
total:{$sum:1}
}}
])
3.2.4 $sort
db.post.aggregate([
{$group:{
_id:'$category_id',
total:{$sum:1}
}},
{$sort:{total:-1}}
])
3.2.5 $limit
db.post.aggregate([
{$sort:{hits:-1}},
{$limit:1}
])
3.2.6 $skip
db.post.aggregate([
{$sort:{hits:-1}},
{$skip:2},
{$limit:2}
])
3.2.7 $unwind
db.post.aggregate([
{$match:{id:3}},
{$unwind:'$tags'}
])
3.2.8 $lookup
db.post.aggregate([
{
$lookup:{
from:'category',
localField:'category_id',
foreignField:'id',
as:'category_info'
}
}
])
3.3 聚合表达式
3.3.1 $sum
db.post.aggregate([
{$group:{
_id:'$category_id',
hits_total:{$sum:'$hits'}
}}
])
3.3.2 $avg
db.post.aggregate([
{$group:{
_id:'$category_id',
hits_avg:{$avg:'$hits'}
}}
])
3.3.3 $min
db.post.aggregate([
{$group:{
_id:'$category_id',
hits_min:{$avg:'$hits'}
}}
])
3.3.4 $max
db.post.aggregate([
{$group:{
_id:'$category_id',
hits_max:{$max:'$hits'}
}}
])
3.3.5 $push
db.post.aggregate([
{$group:{
_id:'$category_id',
title_arr:{$push:'$title'}
}}
])
3.3.6 $addToSet
db.post.aggregate([
{$group:{
_id:'$category_id',
title_arr:{$addToSet:'$title'}
}}
])
3.3.7 $first
db.post.aggregate([
{$group:{
_id:'$category_id',
title_first:{$first:'$title'}
}}
])
3.3.8 $last
db.post.aggregate([
{$group:{
_id:'$category_id',
title_last:{$last:'$title'}
}}
])
本文为 陈华 原创,欢迎转载,但请注明出处:http://edu.ichenhua.cn/read/40
- 上一篇:
- MongoDB基础之查询表达式
- 下一篇:
- MongoDB基础之游标和游标遍历