在前面的文章“mongodb 查詢的語(yǔ)法”里，我介紹了Mongodb的常用查詢語(yǔ)法，Mongodb的update操作也有點(diǎn)復(fù)雜，我結(jié)合自己的使用經(jīng)驗(yàn)，在這里介紹一下，給用mongodb的朋友看看，也方便以后自己用到的時(shí)候查閱：注：在這篇文章及上篇文章內(nèi)講的語(yǔ)法介紹都是在mongodb shell環(huán)境內(nèi)的，和真正運(yùn)用語(yǔ)言編程（如java,php等）使用時(shí)，在使用方法上會(huì)有一些差別，但語(yǔ)法(如查詢條件，$in,$inc等)是一樣的。本文是參考官方文檔來(lái)介紹的，之所以有官方文檔還要在這介紹，一方面是就當(dāng)翻譯，畢竟每次要用時(shí)去看英文文檔比較累，第二是官方文檔講解比較簡(jiǎn)單，有時(shí)光看官方文檔不好理解，我在實(shí)際操作的情況下可以做些補(bǔ)充。好了，不多說(shuō)了，下面正式開始：mongodb更新有兩個(gè)命令：1).update()命令db.collection.update( criteria, objNew, upsert, multi )criteria : update的查詢條件，類似sql update查詢內(nèi)where后面的objNew : update的對(duì)象和一些更新的操作符（如$,$inc...）等，也可以理解為sql update查詢內(nèi)set后面的upsert : 這個(gè)參數(shù)的意思是，如果不存在update的記錄，是否插入objNew,true為插入，默認(rèn)是false，不插入。multi : mongodb默認(rèn)是false,只更新找到的第一條記錄，如果這個(gè)參數(shù)為true,就把按條件查出來(lái)多條記錄全部更新。例：db.test0.update( { "count" : { $gt : 1 } } , { $set : { "test2" : "OK"} } ); 只更新了第一條記錄db.test0.update( { "count" : { $gt : 3 } } , { $set : { "test2" : "OK"} },false,true ); 全更新了db.test0.update( { "count" : { $gt : 4 } } , { $set : { "test5" : "OK"} },true,false ); 只加進(jìn)去了第一條db.test0.update( { "count" : { $gt : 5 } } , { $set : { "test5" : "OK"} },true,true ); 全加進(jìn)去了db.test0.update( { "count" : { $gt : 15 } } , { $inc : { "count" : 1} },false,true );全更新了db.test0.update( { "count" : { $gt : 10 } } , { $inc : { "count" : 1} },false,false );只更新了第一條2).save()命令db.collection.save( x )x就是要更新的對(duì)象，只能是單條記錄。如果在collection內(nèi)已經(jīng)存在一個(gè)和x對(duì)象相同的"_id"的記錄。mongodb就會(huì)把x對(duì)象替換collection內(nèi)已經(jīng)存在的記錄，否則將會(huì)插入x對(duì)象，如果x內(nèi)沒(méi)有_id,系統(tǒng)會(huì)自動(dòng)生成一個(gè)再插入。相當(dāng)于上面update語(yǔ)句的upsert=true,multi=false的情況。例：db.test0.save({count:40,test1:"OK"}); #_id系統(tǒng)會(huì)生成db.test0.save({_id:40,count:40,test1:"OK"}); #如果test0內(nèi)有_id等于40的，會(huì)替換，否則插入。mongodb的更新操作符：1) $inc用法：{ $inc : { field : value } }意思對(duì)一個(gè)數(shù)字字段field增加value，例：> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 16, "test1" : "TESTTEST", "test2" : "OK", "test3" : "TESTTEST", "test4" : "OK", "test5" : "OK" }> db.test0.update( { "_id" : 15 } , { $inc : { "count" : 1 } } );> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 17, "test1" : "TESTTEST", "test2" : "OK", "test3" : "TESTTEST", "test4" : "OK", "test5" : "OK" }> db.test0.update( { "_id" : 15 } , { $inc : { "count" : 2 } } );> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 19, "test1" : "TESTTEST", "test2" : "OK", "test3" : "TESTTEST", "test4" : "OK", "test5" : "OK" }> db.test0.update( { "_id" : 15 } , { $inc : { "count" : -1 } } );> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : "TESTTEST", "test2" : "OK", "test3" : "TESTTEST", "test4" : "OK", "test5" : "OK" }2) $set用法：{ $set : { field : value } }就是相當(dāng)于sql的set field = value，全部數(shù)據(jù)類型都支持$set。例：> db.test0.update( { "_id" : 15 } , { $set : { "test1" : "testv1","test2" : "testv2","test3" : "testv3","test4" : "testv4" } } );> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : "testv1", "test2" : "testv2", "test3" : "testv3", "test4" : "testv4", "test5" : "OK" }3) $unset用法：{ $unset : { field : 1} }顧名思義，就是刪除字段了。例：> db.test0.update( { "_id" : 15 } , { $unset : { "test1":1 } } );> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test2" : "testv2", "test3" : "testv3", "test4" : "testv4", "test5" : "OK" }> db.test0.update( { "_id" : 15 } , { $unset : { "test2": 0 } } );> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test3" : "testv3", "test4" : "testv4", "test5" : "OK" }> db.test0.update( { "_id" : 15 } , { $unset : { "test3":asdfasf } } );Fri May 14 16:17:38 JS Error: ReferenceError: asdfasf is not defined (shell):0> db.test0.update( { "_id" : 15 } , { $unset : { "test3":"test" } } );> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test4" : "testv4", "test5" : "OK" }沒(méi)看出field : 1里面的1是干什么用的，反正只要有東西就行。4) $push用法：{ $push : { field : value } }
把value追加到field里面去，field一定要是數(shù)組類型才行，如果field不存在，會(huì)新增一個(gè)數(shù)組類型加進(jìn)去。例：
> db.test0.update( { "_id" : 15 } , { $set : { "test1" : ["aaa","bbb"] } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "aaa", "bbb" ], "test4" : "testv4", "test5" : "OK" }
> db.test0.update( { "_id" : 15 } , { $push : { "test1": "ccc" } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "aaa", "bbb", "ccc" ], "test4" : "testv4", "test5" : "OK" }
> db.test0.update( { "_id" : 15 } , { $push : { "test2": "ccc" } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "aaa", "bbb", "ccc" ], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }
> db.test0.update( { "_id" : 15 } , { $push : { "test1": ["ddd","eee"] } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "aaa", "bbb", "ccc", [ "ddd", "eee" ] ], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }5) $pushAll用法：{ $pushAll : { field : value_array } }同$push,只是一次可以追加多個(gè)值到一個(gè)數(shù)組字段內(nèi)。例：> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "aaa", "bbb", "ccc", [ "ddd", "eee" ] ], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }> db.test0.update( { "_id" : 15 } , { $pushAll : { "test1": ["fff","ggg"] } } );> db.test0.find( { "_id" : 15 } );{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "aaa", "bbb", "ccc", [ "ddd", "eee" ], "fff", "ggg" ], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }6) $addToSet用法：{ $addToSet : { field : value } }
增加一個(gè)值到數(shù)組內(nèi)，而且只有當(dāng)這個(gè)值不在數(shù)組內(nèi)才增加。例：
> db.test0.update( { "_id" : 15 } , { $addToSet : { "test1": {$each : ["444","555"] } } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [
"aaa",
"bbb",
"ccc",
[
"ddd",
"eee"
],
"fff",
"ggg",
[
"111",
"222"
],
"444",
"555"
], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }
> db.test0.update( { "_id" : 15 } , { $addToSet : { "test1": {$each : ["444","555"] } } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [
"aaa",
"bbb",
"ccc",
[
"ddd",
"eee"
],
"fff",
"ggg",
[
"111",
"222"
],
"444",
"555"
], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }
> db.test0.update( { "_id" : 15 } , { $addToSet : { "test1": ["444","555"] } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [
"aaa",
"bbb",
"ccc",
[
"ddd",
"eee"
],
"fff",
"ggg",
[
"111",
"222"
],
"444",
"555",
[
"444",
"555"
]
], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }
> db.test0.update( { "_id" : 15 } , { $addToSet : { "test1": ["444","555"] } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [
"aaa",
"bbb",
"ccc",
[
"ddd",
"eee"
],
"fff",
"ggg",
[
"111",
"222"
],
"444",
"555",
[
"444",
"555"
]
], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }
7) $pop
刪除數(shù)組內(nèi)的一個(gè)值
用法：
刪除最后一個(gè)值：{ $pop : { field : 1 } }刪除第一個(gè)值：{ $pop : { field : -1 } }
注意，只能刪除一個(gè)值，也就是說(shuō)只能用1或-1，而不能用2或-2來(lái)刪除兩條。mongodb 1.1及以后的版本才可以用，例：
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [
"bbb",
"ccc",
[
"ddd",
"eee"
],
"fff",
"ggg",
[
"111",
"222"
],
"444"
], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }
> db.test0.update( { "_id" : 15 } , { $pop : { "test1": -1 } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [
"ccc",
[
"ddd",
"eee"
],
"fff",
"ggg",
[
"111",
"222"
],
"444"
], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }
> db.test0.update( { "_id" : 15 } , { $pop : { "test1": 1 } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "ccc", [ "ddd", "eee" ], "fff", "ggg", [ "111", "222" ] ], "test2" : [ "ccc" ], "test4" : "testv4",
"test5" : "OK" }
8) $pull
用法：$pull : { field : value } }
從數(shù)組field內(nèi)刪除一個(gè)等于value值。例：
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "ccc", [ "ddd", "eee" ], "fff", "ggg", [ "111", "222" ] ], "test2" : [ "ccc" ], "test4" : "testv4",
"test5" : "OK" }
> db.test0.update( { "_id" : 15 } , { $pull : { "test1": "ggg" } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "ccc", [ "ddd", "eee" ], "fff", [ "111", "222" ] ], "test2" : [ "ccc" ], "test4" : "testv4", "test5"
: "OK" }
9) $pullAll
用法：{ $pullAll : { field : value_array } }
同$pull,可以一次刪除數(shù)組內(nèi)的多個(gè)值。例：
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ "ccc", [ "ddd", "eee" ], "fff", [ "111", "222" ] ], "test2" : [ "ccc" ], "test4" : "testv4", "test5"
: "OK" }
> db.test0.update( { "_id" : 15 } , { $pullAll : { "test1": [ "ccc" , "fff" ] } } );
> db.test0.find( { "_id" : 15 } );
{ "_id" : { "floatApprox" : 15 }, "count" : 18, "test1" : [ [ "ddd", "eee" ], [ "111", "222" ] ], "test2" : [ "ccc" ], "test4" : "testv4", "test5" : "OK" }
10) $ 操作符
$是他自己的意思，代表按條件找出的數(shù)組里面某項(xiàng)他自己。呵呵，比較坳口。看一下官方的例子：
> t.find()
{ "_id" : ObjectId("4b97e62bf1d8c7152c9ccb74"), "title" : "ABC", "comments" : [ { "by" : "joe", "votes" : 3 }, { "by" : "jane", "votes" : 7 } ] }
> t.update( {'comments.by':'joe'}, {$inc:{'comments.$.votes':1}}, false, true )
> t.find()
{ "_id" : ObjectId("4b97e62bf1d8c7152c9ccb74"), "title" : "ABC", "comments" : [ { "by" : "joe", "votes" : 4 }, { "by" : "jane", "votes" : 7 } ] }
需要注意的是，$只會(huì)應(yīng)用找到的第一條數(shù)組項(xiàng)，后面的就不管了。還是看例子：
> t.find();
{ "_id" : ObjectId("4b9e4a1fc583fa1c76198319"), "x" : [ 1, 2, 3, 2 ] }
> t.update({x: 2}, {$inc: {"x.$": 1}}, false, true);
> t.find();
還有注意的是$配合$unset使用的時(shí)候，會(huì)留下一個(gè)null的數(shù)組項(xiàng)，不過(guò)可以用{$pull:{x:null}}刪除全部是null的數(shù)組項(xiàng)。例：
> t.insert({x: [1,2,3,4,3,2,3,4]})
> t.find()
{ "_id" : ObjectId("4bde2ad3755d00000000710e"), "x" : [ 1, 2, 3, 4, 3, 2, 3, 4 ] }
> t.update({x:3}, {$unset:{"x.$":1}})
> t.find()
{ "_id" : ObjectId("4bde2ad3755d00000000710e"), "x" : [ 1, 2, null, 4, 3, 2, 3, 4 ] }
{ "_id" : ObjectId("4b9e4a1fc583fa1c76198319"), "x" : [ 1, 3, 3, 2 ] }