MongoDB 是一个基于分布式文件存储的开源数据库系统。
MongoDB 将数据存储为一个文档,数据结构由键值(key=>value)对组成。MongoDB 文档类似于 JSON 对象。字段值可以包含其他文档,数组及文档数组。
brew install mongodb
在要启动的目录下新建一个目录(如:data)
mkdir data
命令行中输入(--dbpath 参数指定数据库路径)
mongod --dbpath='./data'
如果出现 waiting for connections on port 27017 就表示启动成功。
注意:这个命令窗体绝对不能关,关闭这个窗口就相当于停止了 mongodb 服务。
| 选项 | 含义 |
|---|---|
| --port | 指定服务端口号,默认端口27017 |
| --logpath | 指定MongoDB日志文件,注意是指定文件不是目录 |
| --logappend | 使用追加的方式写日志 |
| --dbpath | 指定数据库路径 |
| --directoryperdb | 设置每个数据库将被保存在一个单独的目录 |
命令行输入
mongo
也可以设置 host
mongo --host 127.0.0.1
对应关系如下图:

show dbs
返回如下:
admin 0.000GB
book 0.000GB
leave 0.000GB
local 0.000GB
page 0.000GB
school 0.000GB
students 0.000GB
实例切换到 school 数据库下:
use school
返回如下:
switched to db school
注:如果此数据库存在,则切换到此数据库下,如果此数据库还不存在也可以切过来,我们刚创建的数据库 school 如果不在列表内,要显示它,我们需要向 school 数据库插入一些数据
db.school.insert({name:'为民小学',age:10});
db 或 db.getName()
db.dropDatabase()
返回如下:
{ "dropped" : "school", "ok" : 1 }
db.school.help()
返回如下:
BCollection help
db.school.find().help() - show DBCursor help
db.school.bulkWrite( operations, <optional params> ) - bulk execute write ...
show collections
返回如下:
grade1
grade2
db.createCollection('grade3')
返回如下:
{ "ok" : 1 }
db.grade1.insert({name: 'Lily', age: 8})
返回如下:
WriteResult({ "nInserted" : 1 })
db.collection_name.insert(document)
db.grade1.insert({name: 'Tom', age: 9})
每当插入一条新文档的时候 mongodb 会自动为此文档生成一个 _id 属性,_id一定是唯一的,用来唯一标识一个文档 _id 也可以直接指定,但如果数据库中此集合下已经有此 _id 的话插入会失败。
{
"_id" : ObjectId("5addbfbb163098017a6a72ed"),
"name" : "Tom",
"age" : 9.0
}
db.collection_name.save(document)
如果不指定 _id 字段 save() 方法类似于 insert() 方法。如果指定 _id 字段,则会更新该 _id 的数据。
// insert
db.grade1.insert({_id: '1',name: 'Han Meimei', age: 8})// WriteResult({ "nInserted" : 1 })
// 存在{_id:1},则更新 _id为1的document
db.grade1.save({_id: '1',name: 'Han Meimei', age: 9})// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
// 不存在{_id:2},则插入一条新文档
db.grade1.save({_id: '2',name: 'Han Meimei', age: 9})// WriteResult({ "nMatched" : 0, "nUpserted" : 1, "nModified" : 0, "_id" : "2" })
执行脚本插入
mongo exc_js/1.js
> load exc_js/1.js
db.collection.update(
<query>,
<updateObj>,
{
upsert: <boolean>,
multi: <boolean>
}
)
{ $inc: { <field1>: <amount1>, <field2>: <amount2>, ... } }
在原基础上累加(increment)
// 给 {name: 'Tom'} 的文档的age累加 10
db.grade1.update({name: 'Tom'}, {$inc: {age:10}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
{ $push: { <field1>: <value1>, ... } }
向数组中添加元素
db.grade1.update({name:'Tom'}, {$push: {'hobby':'reading'} })
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
// { "_id" : ObjectId("5addbfbb163098017a6a72ed"), "name" : "Tom", "hobby" : [ "reading" ] }
// 不会覆盖已有的
db.grade1.update({name:'Tom'}, {$push: {'hobby':'reading'} })
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
// { "_id" : ObjectId("5addbfbb163098017a6a72ed"), "name" : "Tom", "hobby" : [ "reading", "reading" ] }
{ $addToSet: { <field1>: <value1>, ... } }
// /第一次没有 huge
db.grade1.update({_id:3}, {$addToSet: {friends:'huge'}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
// 第二次 有 huge
db.grade1.update({_id:3}, {$addToSet: {friends:'huge'}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 0 })
{ $pop: { <field>: <-1 | 1>, ... } }
db.grade1.update({_id:3}, {$pop:{friends: 1}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
db.grade1.update({_id:3}, {$pop:{friends: -1}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
{ $addToSet: { <field>: { $each: [ <value1>, <value2> ... ] } } }
db.grade1.update({_id:3}, {$addToSet:{friends:{$each: ['huangbo','zhangyixing']}}})
//WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
// 已经有的时候就不会再添加了
db.grade1.update({_id:3}, {$addToSet:{friends:{$each: ['huangbo','zhangyixing']}}})
//WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 0 })
{ $push: { <field>: { $each: [ <value1>, <value2> ... ] } } }
Use with the $push operator to append multiple values to an array .
db.grade1.update({_id:3}, {$push:{friends:{$each: ['huangbo','zhangyixing']}}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
在 $addToSet 中使用时,若有则忽略,若没有则添加。在 $push 中使用时,不管有没有都会添加。
{field: {$ne: value} }
not equal
// 给 name为'Han Meimei' && hobby中不等于'reading' && _id不等于'2'的文档 的hobby 属性 添加一个 'drinking'
db.grade1.update({name: 'Han Meimei', hobby:{$ne:'reading'}, _id: {$ne:'2'}}, {$push: {hobby: 'drinking'}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
{ $set: { <field1>: <value1>, ... } }
/*
原来的数据:
{_id:3, info:{id: '11'}, friends:['liudehua', 'zhourunfa']}
*/
/*设置字段的第一层的值(Set Top-Level Fields)*/
db.grade1.update({_id:3}, {$set:{"info11":{id:'11'}}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
/*设置嵌套字段的值 (Set Fields in Embedded Documents)*/
db.grade1.update({_id:3}, {$set:{"info.id":'22'}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
/*修改指定索引元素*/
db.grade1.update({_id:3}, {$set:{"friends.1":'zhangmanyu'}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 0 })
{ $unset: { <field1>: "", ... } }
删除指定的键
// 把 {name: 'Tom'} 的文档中的 age 键给删除掉
db.grade1.update({name: 'Tom'}, {$unset:{'age':''}})
// WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
/* {
"_id" : ObjectId("5addbfbb163098017a6a72ed"),
"name" : "Tom"
}*/
remove方法是用来移除集合中的数据
语法
db.collection.remove(
<query>,
{
justOne: <boolean>
}
)
参数说明
query :(可选)删除的文档的条件。
justOne : (可选)如果设为 true 或 1,则只删除匹配到的多个文档中的第一个。默认为 true
/*{justOne:true} 值删除匹配到的第一条文档*/
db.grade1.remove({'name': 'Han Meimei'}, {justOne: true})
// WriteResult({ "nRemoved" : 1 })
/*删除匹配到的所有文档*/
db.grade1.remove({'name': 'Han Meimei'})
// WriteResult({ "nRemoved" : 2 })
语法
db.collection_name.find(query, projection);
参数
query - 使用查询操作符指定选择过滤器
projection - 指定配到到的文档中的返回的字段。
/*projection*/
{ field1: <value>, field2: <value> ... }
/*value:*/
1 or true: 在返回的文档中包含这个字段
0 or false:在返回的文档中排除这个字段
_id 字段默认一直返回,除非手动将 _id 字段设置为 0 或 false
举个栗子
//查询grade1下所有的文档
db.grade1.find()
//原始数据():
{ "_id" : 1, "name" : "Tom1", "age" : 9 }
{ "_id" : 2, "name" : "Tom2", "age" : 15 }
{ "_id" : 3, "name" : "Tom3", "age" : 11 }
db.grade1.find({age:{$in:[9,11]}})
// { "_id" : 1, "name" : "Tom1", "age" : 9 }
// { "_id" : 3, "name" : "Tom3", "age" : 11 }
db.grade1.find({age:{$nin:[9,11]}})
// { "_id" : 2, "name" : "Tom2", "age" : 15 }
db.grade1.find({age:{$not:{$lt:11}}})
//{ "_id" : 2, "name" : "Tom2", "age" : 15 }
//{ "_id" : 3, "name" : "Tom3", "age" : 11 }
db.grade1.find({age:{$ne:9}})
// { "_id" : 2, "name" : "Tom2", "age" : 15 }
// { "_id" : 3, "name" : "Tom3", "age" : 11 }
// 原始数据
{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
{ "_id" : 2, "name" : "Tom2", "age" : 15, "friends" : [ "Zhange San", "Li Si" ] }
{ "_id" : 3, "name" : "Tom3", "age" : 11, "friends" : [ "Zhange San", "Lily" ] }
db.grade1.find({"friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ]})
// { "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
db.grade1.find({"friends" : [ "Lily" ]})
// 空
// $all
db.grade1.find({"friends" :{$all: ["Zhang San"]}})
// { "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
// $in
db.grade1.find({"friends" :{$in: ["Zhang San"]}})
{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
// $size
db.grade1.find({"friends" :{$size:4}})
//{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
// $slice
db.collection.find( { field: value }, { array: {$slice: count } } );
> db.grade1.find({"friends" :{$size:4}}, {"friends":{$slice:2}})
//{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs" ] }
// 数据库数据
{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
{ "_id" : 2, "name" : "Tom2", "age" : 15, "friends" : [ "Zhange San", "Li Si" ] }
{ "_id" : 3, "name" : "Tom3", "age" : 11, "friends" : [ "Zhange San", "Lily" ] }
// JS表达式的字符串
> db.grade1.find({$where:'this.name == "Tom1"'})
//{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
// 函数
> db.grade1.find({$where: function(){return this.age == 9}})
// { "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
这些方法改变了执行基础查询方式。
包括 cursor.forEach()、cursor.map()、cursor.limit()、cursor.size()、cursor.count() 等。
// forEach举例
> var result = db.grade1.find({$where: function(){return this.age >= 9}});
> result.forEach(elem => printjson(elem))
/*{
"_id" : 1,
"name" : "Tom1",
"age" : 9,
"friends" : [
"Lily",
"Jobs",
"Lucy",
"Zhang San"
]
}
{
"_id" : 2,
"name" : "Tom2",
"age" : 15,
"friends" : [
"Zhange San",
"Li Si"
]
}
{
"_id" : 3,
"name" : "Tom3",
"age" : 11,
"friends" : [
"Zhange San",
"Lily"
]
}*/
// 大于等于
db.grade1.find({age:{$gte:9}})
/*{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
{ "_id" : 2, "name" : "Tom2", "age" : 15, "friends" : [ "Zhange San", "Li Si" ] }
{ "_id" : 3, "name" : "Tom3", "age" : 11, "friends" : [ "Zhange San", "Lily" ] }*/
// 大于等于9 并且 小于等于13
db.grade1.find({age:{$gte:9}, age: {$lte:13}})
/*{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
{ "_id" : 3, "name" : "Tom3", "age" : 11, "friends" : [ "Zhange San", "Lily" ] }*/
//原始数据
{ "_id" : ObjectId("5ae1b6e3e4366d57f3307239"), "name" : "Tom4" }
> db.grade1.find({_id: '5ae1b6e3e4366d57f3307239'}).count()
// 0
> db.grade1.find({_id:ObjectId('5ae1b6e3e4366d57f3307239')}).count()
// 1
count() 查询结果的条数
db.collection.find({key:/value/})
// name是以`T`开头的数据
db.grade1.find({name: /^T/})
/*{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
{ "_id" : 2, "name" : "Tom2", "age" : 15, "friends" : [ "Zhange San", "Li Si" ] }
{ "_id" : 3, "name" : "Tom3", "age" : 11, "friends" : [ "Zhange San", "Lily" ] }
{ "_id" : ObjectId("5ae1b6e3e4366d57f3307239"), "name" : "Tom4" }*/
db.collection_name.find({field1: value1, field2:value2})
//原始数据
{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
{ "_id" : 2, "name" : "Tom2", "age" : 15, "friends" : [ "Zhange San", "Li Si" ] }
{ "_id" : 3, "name" : "Tom3", "age" : 11, "friends" : [ "Zhange San", "Lily" ] }
// and name是以‘T’开头 并且 age是9 的数据
> db.grade1.find({name: /^T/, age: 9})
// { "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
db.collection_name.find({ $or: [{key1: value1}, {key2:value2} ] })
// name 是Tom1 或者 age是11 的数据
> db.grade1.find({$or:[{name: 'Tom1'}, {age: 11}]})
/*{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }
{ "_id" : 3, "name" : "Tom3", "age" : 11, "friends" : [ "Zhange San", "Lily" ] }*/
> db.grade1.find({age: 9,$or:[{name: 'Tom1'}, {age: 11}]})
/*{ "_id" : 1, "name" : "Tom1", "age" : 9, "friends" : [ "Lily", "Jobs", "Lucy", "Zhang San" ] }*/
读取指定数量的数据记录 语法
db.collectoin_name.find().limit(number)
跳过指定数量的数据
db.collectoin_name.find().skip(number)
通过参数指定排序的字段,并使用 1 和 -1 来指定排序的方式,其中 1 为升序排列,而 -1 是用于降序排列。
db.collectoin_name.find().sort({field:1})
db.collectoin_name.find().sort({field:-1})
// 原始数据为 1 2 3 4 5 6 7 8 9
> var pageIndex = 3;
> var pageSize = 3;
> var res = db.grade1.find({}).skip((pageIndex - 1) * pageSize).limit(pageSize).sort({username: 1});
> res
/*{ "_id" : ObjectId("5ae1cbc609f3ac9a41442546"), "username" : "Lily_7", "password" : 7 }
{ "_id" : ObjectId("5ae1cbc609f3ac9a41442547"), "username" : "Lily_8", "password" : 8 }
{ "_id" : ObjectId("5ae1cbc609f3ac9a41442548"), "username" : "Lily_9", "password" : 9 }*/
var res1 = db.grade1.find().skip((pageIndex - 1) * pageSize).limit(pageSize).sort({username: -1});
/*{ "_id" : ObjectId("5ae1cbc609f3ac9a41442542"), "username" : "Lily_3", "password" : 3 }
{ "_id" : ObjectId("5ae1cbc609f3ac9a41442541"), "username" : "Lily_2", "password" : 2 }
{ "_id" : ObjectId("5ae1cbc609f3ac9a41442540"), "username" : "Lily_1", "password" : 1 }*/
没有先后顺序
之前我们使用 MySQL 等关系型数据库时,主键都是设置成自增的。但在分布式环境下,这种方法就不可行了,会产生冲突。为此,MongoDB 采用了一个称之为 ObjectId 的类型来做主键。ObjectId 是一个12字节的 BSON 类型字符串。按照字节顺序,一次代表:

