update 函数似乎比insert,remove跟find难用一点,原因参数多了。
成都创新互联公司基于成都重庆香港及美国等地区分布式IDC机房数据中心构建的电信大带宽,联通大带宽,移动大带宽,多线BGP大带宽租用,是为众多客户提供专业服务器托管报价,主机托管价格性价比高,为金融证券行业成都多线机房,ai人工智能服务器托管提供bgp线路100M独享,G口带宽及机柜租用的专业成都idc公司。
不过简单的写一些比较少用的技巧,普通更新一般比较简单。
关于$inc操作:
属于原子操作,可以整型,长整型或浮点数进行增加操作,如下:
> test={ ... "name":"Mike", ... "friends":32, ... "enemies":2, ... "Balance":155.6 ... } { "name" : "Mike", "friends" : 32, "enemies" : 2, "Balance" : 155.6 } > db.updatetest.insert(test) WriteResult({ "nInserted" : 1 }) > db.updatetest.update({"name":"Mike"},{"$inc":{"Balance":5.4}}) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) > db.updatetest.find() { "_id" : ObjectId("5625ab9b16a7b138cf6bf46f"), "name" : "Mike", "friends" : 32, "enemies" : 2, "Balance" : 161 }
关于$set操作:
对存在的文档指定一个字段值,如果该字段不存在则创建它,如下:
test={ "name":"Mike", "age":32, "sex":"male", "location":"China"} > db.updatetest.insert(test) WriteResult({ "nInserted" : 1 }) > db.updatetest.find() { "_id" : ObjectId("5625ae7616a7b138cf6bf471"), "name" : "Mike", "age" : 32, "sex" : "male", "location" : "China" }
增加亲属关系:
> relatives={"Wife":"May","Father":"Joe","Mother":"Jojo"} { "Wife" : "May", "Father" : "Joe", "Mother" : "Jojo" } > db.updatetest.update({"_id":ObjectId("5625ae7616a7b138cf6bf471")}, {"$set":{"relatives":relatives}}) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) > db.updatetest.find() { "_id" : ObjectId("5625ae7616a7b138cf6bf471"), "name" : "Mike", "age" : 32, "sex" : "male", "location" : "China", "relatives" : { "Wife" : "May", "Father" : "Joe", "Mother" : "Jojo" } }
利用update操作数组:
$push可以对已有的文档中的数组尾部加入一个元素,如果没有就直接创建一个新数组,如下:
people={ "title":"mail list", "Phone":"010-123456"} > db.updatetest.insert(people) WriteResult({ "nInserted" : 1 })
db.updatetest.update({"title":"mail list"}, {"$push":{"emails":"test@123.com"}}) db.updatetest.update({"title":"mail list"}, {"$push":{"emails":"test@123.com"}}) > db.updatetest.find() { "_id" : ObjectId("5625ae7616a7b138cf6bf471"), "name" : "Mike", "age" : 32, "sex" : "male", "location" : "China", "relatives" : { "Wife" : "May", "Father" : "Joe", "Mother" : "Jojo" } } { "_id" : ObjectId("5625b11216a7b138cf6bf472"), "title" : "mail list", "Phone" : "010-123456", "emails" : [ "test@123.com", "test@123.com" ] }
$push是允许重复插入记录的。
$push还允许与slice,sort,each一起使用,如下:
db.updatetest.update({"title":"mail list"}, {"$push":{"emails":{$each:["test1@123.com","test2@124.com","test3@125.com"]}}}) > db.updatetest.find() { "_id" : ObjectId("5625b2ba16a7b138cf6bf473"), "title" : "mail list", "Phone" : "010-123456", "emails" : [ "test@123.com", "test@123.com", "test1@123.com", "test2@124.com", "test3@125.com" ] }
与$slice一起使用:
> db.updatetest.update({"title":"mail list"},{$push:{"top10":{"$each":["KFC","mcdonalds","pizzahut"],"$slice":-3}}}) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
没有top10这个键的时候会自动创建一个
> db.updatetest.find() { "_id" : ObjectId("5625b44d16a7b138cf6bf474"), "title" : "mail list", "Phone" : "010-123456", "emails" : [ "test1@123.com", "test2@124.com", "test3@125.com" ], "top10" : [ "KFC", "mcdonalds", "pizzahut" ] }
现在top10这个数组刚刚好有3个元素,再次插入两个,则只保留最后三个
> db.updatetest.update({"title":"mail list"},{$push:{"top10":{"$each":["starbucks","illy"],"$slice":-3}}}) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })> db.updatetest.find() { "_id" : ObjectId("5625b44d16a7b138cf6bf474"), "title" : "mail list", "Phone" : "010-123456", "emails" : [ "test1@123.com", "test2@124.com", "test3@125.com" ], "top10" : [ "pizzahut", "starbucks", "illy" ] }
对于$push是可以插入重复数据,如果邮件地址已经存在不想再重复插入,则可以使用$addToSet操作,如下:
> db.updatetest.update({"title":"mail list"},... {"$addToSet":{"emails":"test1@123.com"}... }) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 0 })> db.updatetest.find() { "_id" : ObjectId("5625b44d16a7b138cf6bf474"), "title" : "mail list", "Phone" : "010-123456", "emails" : [ "test1@123.com", "test2@124.com", "test3@125.com" ], "top10" : [ "HP", "DELL", "MicroSoft" ] }> db.updatetest.update({"title":"mail list"}, {"$addToSet":{"emails":{$each:["test1@123.com","abc@321.com","new@newaddress.com","abc@321.com"]}} }) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })> db.updatetest.find() { "_id" : ObjectId("5625b44d16a7b138cf6bf474"), "title" : "mail list", "Phone" : "010-123456", "emails" : [ "test1@123.com", "test2@124.com", "test3@125.com", "abc@321.com", "new@newaddress.com" ], "top10" : [ "HP", "DELL", "MicroSoft" ] }
基于位置的数组修改:
由一个内嵌文档组成的数组,内容如下:
> db.updatetest.insert({
...
... "content":"MongoDB Update",
... "comments":[
... {
... "comment":"Good post",
... "author":"John",
... "votes":0
... },
... {
... "comment":"too short",
... "author":"mike",
... "votes":3
... },
... {
... "comment":"free watches",
... "author":"Alice",
... "votes":-1
... }
... ]
...
... })
WriteResult({ "nInserted" : 1 })
可以通过$符号来作为定位符,修改comments里面的键,例如修改comments.comment.author 这个键想把Alice改为May操作如下:
> db.updatetest.update({"comments.author":"Alice"},{"$set":{"comments.$.author":"May"}})
WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
> db.updatetest.find()
{ "_id" : ObjectId("5625b44d16a7b138cf6bf474"), "title" : "mail list", "Phone" : "010-123456", "emails" : [ "test1@123.com", "test2@124.com", "test3@125.com", "abc@321.com", "new@newaddress.com" ], "top10" : [ "HP", "DELL", "MicroSoft" ] }
{ "_id" : ObjectId("5625bbef16a7b138cf6bf475"), "content" : "Mongodb Update", "comments" : [ { "comment" : "Good post", "author" : "John", "votes" : 0 }, { "comment" : "too short", "author" : "mike", "votes" : 3 }, { "comment" : "free watches", "author" : "May", "votes" : -1 } ] }
最后一个是findAndModify操作:
该操作可以返回更新前的文档
如下:
> dc=db.runCommand({"findAndModify":"updatetest",
... "query":{"title":"mail list"},
... "sort":{"_id":-1},
... "update":{"$set":{"company":"QQ"}}})
> dc.value
{
"_id" : ObjectId("5625b44d16a7b138cf6bf474"),
"title" : "mail list",
"Phone" : "010-123456",
"emails" : [
"test1@123.com",
"test2@124.com",
"test3@125.com",
"abc@321.com",
"new@newaddress.com"
],
"top10" : [
"HP",
"DELL",
"MicroSoft"
]
}
关于findAndModify操作在update位置的参数可以使用remove,表示匹配query则删除该文档。
还有new,表示返回更新前的文档还是更新后的文档,默认是更新前的。
fields 文档中需要返回的字段(可选)
upsert 布尔值,true表示是一个upsert,默认为false
或者可以这样操作:
> db.updatetest.findAndModify({query:{"title":"mail list"}, update:{$set:{"company":"BBC"}}, new:true, sort:{"_id":-1}, fields:{"emails":1}, upsert:true})
{
"_id" : ObjectId("5625b44d16a7b138cf6bf474"),
"emails" : [
"test1@123.com",
"test2@124.com",
"test3@125.com",
"abc@321.com",
"new@newaddress.com"
]
}
该条语句修改了company为bbc,然后返回字段是emails,new是表示返回更新后的数据.