mongo客户端有两种方式与mongodb服务进行交互,一种是mongo shell,一种是执行javascript脚本.mongo shell,平时用的比较多,但是javascript脚本却很少用.前段时间,在一个项目中通过写js脚本批量的更新了一些数据,做一下总结.
示例: mongo localhost:27017/hr_assistant index.js 在index.js中先跟mongodb打个招呼,然后打印一下当前的数据库
print('hello mongodb') // 打印连接之后的数据库 print(db);执行结果:
MongoDB shell version v4.0.11 connecting to: mongodb://localhost:27017/hr_assistant?gssapiServiceName=mongodb Implicit session: session { "id" : UUID("3e0c8e42-57d5-429d-a00d-bad97ec95d73") } MongoDB server version: 4.0.11 hello mongodb hr_assistant 上述方式是通过mongo客户端直接执行一个js脚本,也可以在进入客户端的命令行交互中使用load函数去加载一个脚本实例:
// 进入客户端 pan@ubuntu18:~/disk/panyanan/blog/mongodb$ mongo localhost:27017/hr_assistant MongoDB shell version v4.0.11 connecting to: mongodb://127.0.0.1:27017/ // load脚本文件 > load('index.js') hello mongodb hr_assistant true注意上述两种方式都需要注意脚本的路径问题，最好是在脚本所有文件夹去执行mongo命令，省心，省事．
mongoShell是一个简化版的JavaScript Shell, 可以自由的编写javascript代码,也内置了很多属于mongodb的特有对象及方法,但是在脚本中你无法使用像 show databases、show collections、use test这种命令，需要使用客户端提供函数如db.getCollectionNames()；下面这个表格是常用命令对应的函数
commandsfunctionshow dbs, show databasesdb.adminCommand('listDatabases')use dbdb = db.getSiblingDB('db')show collectionsdb.getCollectionsNames()show usersdb.getUsers()还有两个在js脚本中常用的函数，print()/printjson()可以将参数打印到标准输出上。
示例 index.js：
// 因为是--nodb 形式启动的mongo客户端 所以新建一个连接 const conn = new Mongo('localhost:27017'); print(`连接: ${conn}`) let db = conn.getDB('hr_assistant'); print(`当前数据库：${db}`); const dbs = db.adminCommand('listDatabases'); print('显示所有的数据库:') printjson(dbs); const collections = db.getCollectionNames(); print(`${db}中的collections:`); printjson(collections); db = db.getSiblingDB('test'); print(`切换数据库为${db}`);运行结果:
pan@ubuntu18:~/disk/panyanan/blog/mongodb$ mongo --nodb index.js MongoDB shell version v4.0.11 连接: connection to localhost:27017 当前数据库：hr_assistant 显示所有的数据库: { "databases" : [ { "name" : "admin", "sizeOnDisk" : 32768, "empty" : false }, { "name" : "config", "sizeOnDisk" : 49152, "empty" : false }, { "name" : "hr_assistant", "sizeOnDisk" : 98873344, "empty" : false }, { "name" : "local", "sizeOnDisk" : 73728, "empty" : false } ], "totalSize" : 99028992, "ok" : 1 } hr_assistant中的collections: [ "hr_business_call_records", "hr_business_info_collection", "hr_business_info_follow", "hr_business_info_interviewed", "hr_business_info_meeting", "hr_business_info_planned", "hr_business_info_user", "hr_business_resume_assessment", "hr_business_resume_basic_work", "hr_business_resume_basiceducation", "hr_business_resume_basichealth", "hr_business_resume_basicinfo", "hr_business_resume_head_portrait", "hr_business_resume_jobobjective", "hr_business_resume_percentage", "hr_business_resume_workexp", "hr_business_sign_list", "hr_business_signed_result" ] 切换数据库为test了解了如何运行js脚本以及与mongoshell的区别,就可以在js文件中成为一个curdBoy了.因为在js文件中可以使用mongo客户端提供的所有增删改查方法,加上js内置对象/数组的方法,简直是如虎添翼. 下面是个简单的示例
const conn = new Mongo('localhost:27017'); const db = conn.getDB('test'); // 向emp集合中插入一些记录 let emps =[ { ename: 'Smith', deptno: 20, job: 'salesman', mgr: '', sal: 800, }, { ename: 'Peter', deptno: 30, job: 'manager', mgr: '', sal: 1000, }, { ename: 'Jack', deptno: 40, job: 'president', mgr: '', sal: 3000, }, { ename: 'Rose', deptno: 50, job: 'analyst', mgr: '', sal: 1500, }, ] // 批量插入 let result = db.emps.insert(emps); print(`批量插入一写员工`) print(result); // 获取jack let jack = db.emps.findOne({ename: 'Jack'}); print(`获取jack:`) printjson(jack) // 调整jack的薪资为5000 result = db.emps.update({_id: jack._id}, {$set: {sal: 5000}}) jack = db.emps.findOne({ename: 'Jack'}); // 更新薪资后的jack print('更新薪资后的jack:') printjson(jack) // 获取所有的员工 emps = db.emps.find({}); print('获取所有的员工:'); // 更新所有员工的领导为jordan emps.forEach(function(emp) { printjson(emp); emp.mgr = 'Jordan'; db.emps.save(emp) }); emps = db.emps.find({}); print('更新所有员工的领导为jordan'); while(emps.hasNext()) { printjson(emps.next()) } result = db.emps.remove({}); print(`删除员工:: ${result}`);运行结果
pan@ubuntu18:~/disk/panyanan/blog/mongodb$ mongo --nodb curd.js MongoDB shell version v4.0.11 批量插入一写员工 BulkWriteResult({ "writeErrors" : [ ], "writeConcernErrors" : [ ], "nInserted" : 4, "nUpserted" : 0, "nMatched" : 0, "nModified" : 0, "nRemoved" : 0, "upserted" : [ ] }) 获取jack: { "_id" : ObjectId("5d44fdd70998b36ed6983e6c"), "ename" : "Jack", "deptno" : 40, "job" : "president", "mgr" : "", "sal" : 3000 } 更新薪资后的jack: { "_id" : ObjectId("5d44fdd70998b36ed6983e6c"), "ename" : "Jack", "deptno" : 40, "job" : "president", "mgr" : "", "sal" : 5000 } 获取所有的员工: { "_id" : ObjectId("5d44fdd70998b36ed6983e6a"), "ename" : "Smith", "deptno" : 20, "job" : "salesman", "mgr" : "", "sal" : 800 } { "_id" : ObjectId("5d44fdd70998b36ed6983e6b"), "ename" : "Peter", "deptno" : 30, "job" : "manager", "mgr" : "", "sal" : 1000 } { "_id" : ObjectId("5d44fdd70998b36ed6983e6c"), "ename" : "Jack", "deptno" : 40, "job" : "president", "mgr" : "", "sal" : 5000 } { "_id" : ObjectId("5d44fdd70998b36ed6983e6d"), "ename" : "Rose", "deptno" : 50, "job" : "analyst", "mgr" : "", "sal" : 1500 } 更新所有员工的领导为jordan { "_id" : ObjectId("5d44fdd70998b36ed6983e6a"), "ename" : "Smith", "deptno" : 20, "job" : "salesman", "mgr" : "Jordan", "sal" : 800 } { "_id" : ObjectId("5d44fdd70998b36ed6983e6b"), "ename" : "Peter", "deptno" : 30, "job" : "manager", "mgr" : "Jordan", "sal" : 1000 } { "_id" : ObjectId("5d44fdd70998b36ed6983e6c"), "ename" : "Jack", "deptno" : 40, "job" : "president", "mgr" : "Jordan", "sal" : 5000 } { "_id" : ObjectId("5d44fdd70998b36ed6983e6d"), "ename" : "Rose", "deptno" : 50, "job" : "analyst", "mgr" : "Jordan", "sal" : 1500 } 删除员工:: WriteResult({ "nRemoved" : 4 })转载于:https://www.cnblogs.com/pandapeter/p/11294232.html
相关资源：垃圾分类数据集及代码