本文介绍了MongoDB的进阶操作，包括如何删除和查看数据库中的集合，创建和转换固定集合，以及如何使用gridFS进行文件存储。详细讲解了gridFS的常用命令，如上传、下载、查询和删除文件，并通过实际操作展示了mongofiles工具的用法。
第七章 进阶指南
一数据库命令
命令的工作原理
#删除集合的操作
db.dropTest.drop()
#也可以使用如下操作,同上面的效果一样
#显示数据库中的所有集合
> show collections
blog
blog.posts
dropTest
foo
prices
Users
#删除集合
>db.runCommand({"drop":"dropTest"});
{ "ns" :"test.dropTest", "nIndexesWas" : 1, "ok" : 1 }
> show collections
blog
blog.posts
foo
prices
users
命令参考
#取得命令的最新参考
| Command | Attributes | Help |
| R S A | Report status of a replica set from the POV of this server | |
| killOp | R S A |
|
| rolesInfo | R | Returns information about roles. |
| saslContinue | R S | Subsequent steps in a SASL authentication conversation. |
| dropIndexes | W | drop indexes for a collection 删除集合里name的索引 |
| drop | W | drop a collection
|
| cloneCollectionAsCapped | W |
|
| dbHash | R S |
|
| planCacheSetFilter | R | Sets index filter for a query shape. Overrides existing filter. |
| _recvChunkStart | R A | internal |
| writebacklisten | R S A | v3.0 disallowed internal command, present for compatibility only |
| reIndex | W S | re-index a collection |
| revokeRolesFromUser | W | Revokes roles from a user. |
| parallelCollectionScan | R S |
|
| getnonce | R S | internal |
| geoNear | R S | |
| touch | R S | touch collection |
| getMore | R S | retrieve more results from an existing cursor |
| profile | W S | enable or disable performance profiling |
| handshake | R S | internal |
| compact | R S | compact collection |
| group | R | http://dochub.mongodb.org/core/aggregation 分组 |
| _recvChunkCommit | R A | internal |
| splitChunk | R A | internal command usage only |
| planCacheListPlans | R | Displays the cached plans for a query shape. |
| dropDatabase | W | drop (delete) this database 删除数据库 |
| replSetGetRBID | R S A |
|
| shardingState | W A |
|
| filemd5 | R S | example: |
| authenticate | R S | internal |
| killCursors | R S | kill a list of cursor ids |
| connectionStatus | R S | Returns connection-specific information such as logged-in users and their roles 联接状态 |
| mergeChunks | R A | Merge Chunks command |
| logout | R S | de-authenticate |
| driverOIDTest | R S |
|
| replSetSyncFrom | R S A |
|
| unsetSharding | R S A | internal |
| R S | Check if this server is primary for a replica pair/set; also if it is --master or --slave in simple master/slave setups.
| |
| _transferMods | R A | internal |
| planCacheClearFilters | R | Clears index filter for a single query shape or, if the query shape is omitted, all filters for the collection. |
| update | R | update documents 更新文档 |
| ping | R S | a way to check that the server is alive. responds immediately even if server is in a db lock. 检查服务器联接是否正常 |
| grantPrivilegesToRole | W | Grants privileges to a role |
| createIndexes | R |
|
| connPoolStats | R S | stats about connections between servers in a replica set or sharded cluster. 联接池状态 |
| invalidateUserCache | R S A | Invalidates the in-memory cache of user information |
| R A | list databases on this server 列出服务器上的所有数据库 | |
| dropAllRolesFromDatabase | W | Drops all roles from the given database. Before deleting the roles completely it must remove them from any users or other roles that reference them. If any errors occur in the middle of that process it's possible to be left in a state where the roles have been removed from some user/roles but otherwise still exist. |
| shutdown | R S A | shutdown the database. must be ran against admin db and either (1) ran from localhost or (2) authenticated. If this is a primary in a replica set and there is no member within 10 seconds of its optime, it will not shutdown without force : true. You can also specify timeoutSecs : N to wait N seconds for other members to catch up. |
| connPoolSync | R S | internal |
| setShardVersion | R S A | internal |
| fsync | R S A | |
| getCmdLineOpts | R S A | get argv |
| usersInfo | R | Returns information about users. |
| _getUserCacheGeneration | R S A | internal |
| authSchemaUpgrade | W A | Upgrades the auth data storage schema |
| convertToCapped | W |
|
| planCacheListQueryShapes | R | Displays all query shapes in a collection. |
| _mergeAuthzCollections | W A | Internal command used by mongorestore for updating user/role data |
| count | R | count objects in collection |
| logRotate | R S A |
|
| listIndexes | R | list indexes for a collection |
| replSetDeclareElectionWinner | R S A |
|
| insert | R | insert documents 插入文档 |
| geoSearch | R S |
|
| getParameter | R S A | get administrative option(s) |
| delete | R | delete documents 删除文档 |
| distinct | R |
|
| replSetStepDown | R S A |
|
| find | R | query for documents 查询文档 |
| validate | R S | Validate contents of a namespace by scanning its data structures for correctness. Slow. |
| setParameter | R S A | set administrative option(s) |
| R S | return build level feature settings | |
| repairDatabase | W S | repair database. also compacts. note: slow. 修复并压缩当前数据库 |
| collStats | R S |
|
| getShardVersion | R A | example: |
| currentOp | R S A |
|
| listCollections | R | list collections for this db 显不当前数据库中的集合 |
| whatsmyuri | R S |
|
| replSetInitiate | R S A | Initiate/christen a replica set. |
| grantRolesToUser | W | Grants roles to a user. |
| shardConnPoolStats | R S | stats about the shard connection pool |
| R S | returns lots of administrative server statistics 服务器的管理统计信息 | |
| checkShardingIndex | R | Internal command. |
| resetError | R S | reset error state (used with getpreverror) |
| revokeRolesFromRole | W | Revokes roles from another role. |
| fsyncUnlock | R S A |
|
| replSetHeartbeat | R S A |
|
| W |
返回已更新的文档 | |
| dbStats | R S | Get stats on a database. Not instantaneous. Slower for databases with large .ns files. |
| renameCollection | W A | example:
|
| forceerror | R S | for testing purposes only. forces a user assertion exception |
| clone | W | clone this database from an instance of the db on another host |
| cleanupOrphaned | R A |
|
| dropUser | W | Drops a single user. |
| getPrevError | R S | check for errors since last reseterror commandcal |
| dropAllUsersFromDatabase | W | Drops all users for a single database. |
| appendOplogNote | R A | Adds a no-op entry to the oplog |
| applyOps | W | internal (sharding) |
| availableQueryOptions | R S |
|
| replSetFreeze | R S A |
|
| updateRole | W | Used to update a role |
| replSetMaintenance | R S A |
|
| updateUser | W | Used to update a user, for example to change its password |
| _recvChunkAbort | R A | internal |
| replSetReconfig | R S A | Adjust configuration of a replica set |
| splitVector | R | Internal command. |
| dropRole | W | Drops a single role. Before deleting the role completely it must remove it from any users or roles that reference it. If any errors occur in the middle of that process it's possible to be left in a state where the role has been removed from some user/roles but otherwise still exists. 删除角色 |
| R S | returns information about the daemon's host | |
| revokePrivilegesFromRole | W | Revokes privileges from a role |
| createUser | W | Adds a user to the system 添加系统用户 |
| getLog | R S A |
|
| diagLogging | W S A | |
| copydbgetnonce | R A | get a nonce for subsequent copy db request from secure server |
| explain | R | explain database reads and writes |
| replSetUpdatePosition | R S A |
|
| replSetFresh | R S A |
|
| _migrateClone | R A | internal |
| getShardMap | R S A | lists the set of shards known to this instance |
| mapReduce | R S | Run a map/reduce operation on the server. |
| _recvChunkStatus | R A | internal |
| aggregate | R |
|
| planCacheClear | R | Drops one or all cached queries in a collection. |
| moveChunk | R A | should not be calling this directly |
| copydbsaslstart | R A | Initialize a SASL auth session for subsequent copy db request from secure server |
| R S A | Returns the current replica set configuration | |
| createRole | W | Adds a role to the system |
| replSetRequestVotes | R S A |
|
| grantRolesToRole | W | Grants roles to another role. |
| cloneCollection | R |
|
| getLastError | R S | return error status of the last operation on this connection 查看对本集合执行的最后一次操作的错误信息 |
| create | W | create a collection explicitly
|
| repairCursor | R S |
|
| resync | W S A | resync (from scratch) a stale slave or replica set secondary node. |
| planCacheListFilters | R | Displays index filters for all query shapes in a collection. |
| collMod | W | Sets collection options. |
| saslStart | R S | First step in a SASL authentication conversation. |
| currentOpCtx | R S |
|
| R S A | usage by collection, in micros | |
| eval | R | DEPRECATED |
| copydb | R A | copy a database from another host to this host |
| listCommands | R S | get a list of all db commands 返回数据库命令 |
| replSetElect | R S A |
|
| dataSize | R S | determine data size for a set of data in a certain range |
| _isSelf | R S |
|
| mapreduce.shardedfinish | R S | internal |
| R S | get version #, etc. |
二固定集合
事先创建大小固定
固定集合像环形队列,队尾插入,空间满时,队首插入,删除原有
固定集合没有索引
创建固定集合
#创建容量为10000字节,文档上限为100的my_collections集合
>db.createCollection("my_collections",{capped:true,size:10000,max:100})
{ "ok" : 1 }
> show collections
blog
blog.posts
foo
my_collections
prices
users
将普通集合转为固定集合
db.runCommand({convertToCapped:”test,size:10000”})
自然排序
#按反向插入的顺序查询
Db.my_collections.find().sort({“natural”:-1})
尾部游标
三gridFS存储文件
用来存储大于进制文件的机制
Mongofiles 可以用来在gridFS中上传 下载 列示 查找和删除文件
查看mongofile的可用选项
C:\Users\Administrator>mongofiles --help
Usage:
mongofiles <options> <command> <filename or _id>
Manipulate gridfs files using the commandline.
Possible commands include:
list - list all files;'filename' is an optional prefix which listed filenames must begin with
列表显示 gridFS中的所有文件
search - search all files;'filename' is a substring which listed filenames must contain
查询文件
put - add a file withfilename 'filename'
上传文件到gridFS
get - get a file withfilename 'filename'
将gridFS中的文件写入到文件系统中(下载)
get_id - get a file with the given '_id'
根据id查询文件
delete - delete all files withfilename 'filename'
删除gridFS中指定文件
delete_id - delete a file with the given '_id'
根据id删除文件
Seehttp://docs.mongodb.org/manual/reference/program/mongofiles/ for moreinformation.
general options:
/help printusage
/version print thetool version and
exit
verbosity options:
/v,/verbose:<level> more detailed logoutput
(include multiple times for
more verbosity, e.g. -vvvvv,
or specify anumeric value,
e.g. --verbose=N)
/quiet hide alllog output
不显示输出日志
connection options:
/h,/host:<hostname> mongodb host toconnect to
(setname/host1,host2 for
replica sets)
指定服务器名
/port:<port> server port(can also use
--host hostname:port)
指定服务器端口号
ssl options:
/ssl connectto a mongod or mongos
that has ssl enabled
/sslCAFile:<filename> the .pem file containingthe
root certificate chain from
the certificate authority
/sslPEMKeyFile:<filename> the .pem file containing the
certificate and key
/sslPEMKeyPassword:<password> the password to decrypt the
sslPEMKeyFile, if necessary
/sslCRLFile:<filename> the .pem file containingthe
certificate revocation list
/sslAllowInvalidCertificates bypass the validation for
server certificates
/sslAllowInvalidHostnames bypass the validation for
server name
/sslFIPSMode use FIPS modeof the
installed openssl library
authentication options:
/u,/username:<username> username forauthentication
/p,/password:<password> password forauthentication
/authenticationDatabase:<database-name> database that holds the
user's credentials
/authenticationMechanism:<mechanism> authentication mechanism to
use
storage options:
/d,/db:<database-name> database to use(default is
'test')
指定数据库.默认为test
/l,/local:<filename> local filename forput|get
/t,/type: content/MIME type for put
(optional)
指定上传文件的类型
/r,/replace remove other files with same
name after put
/prefix:<prefix> GridFS prefix touse (default
is 'fs')
/writeConcern:<write-concern> write concern options e.g.
--writeConcern majority,
--writeConcern '{w: 3,
wtimeout: 500, fsync: true,
j: true}' (defaults to
'majority')
query options:
/readPreference:<string>|<json> specify either a preference
name or a preference json
object
Mongofiles基本操作
#创建foo.txt文件,添加内容helloworld
D:\workspace_python>echo "helloworld!" >foo.txt
#查看当前目录下的所有文件,也可以使用dir命令.本机安装cygwin所以可以使用linux命令
D:\workspace_python>ls
chat exception.py hello.py mysite net readTextFile.py spam.py unittest wordcount.py
database.py expand internet namedata.py opencv setup sqlite3 urlib
dictTest.py foo.txt makeTextFile.py namedata_one.py property.py spam swig winGUI.py
#查询gridFS中的所有文件
D:\workspace_python>mongofiles list
2017-01-11T16:27:50.872+0800 connected to: localhost
#上传文件foo.txt到gridFS
D:\workspace_python>mongofiles putfoo.txt
2017-01-11T16:27:55.705+0800 connected to: localhost
added file: foo.txt
#删除本地磁盘文件
D:\workspace_python>rm foo.txt
#查询当前目录的所有文件
D:\workspace_python>ls
chat exception.py internet namedata.py opencv setup sqlite3 urlib
database.py expand makeTextFile.py namedata_one.py property.py spam swig winGUI.py
dictTest.py hello.py mysite net readTextFile.py spam.py unittest wordcount.py
#从gridFS 下载文件到本地磁盘
D:\workspace_python>mongofiles getfoo.txt
2017-01-11T16:28:32.492+0800 connected to: localhost
finished writing to foo.txt
#查看当前目录中的所有文件
D:\workspace_python>ls
chat exception.py hello.py mysite net readTextFile.py spam.py unittest wordcount.py
database.py expand internet namedata.py opencv setup sqlite3 urlib
dictTest.py foo.txt makeTextFile.py namedata_one.py property.py spam swig winGUI.py
#查看foo.txt文件内容
D:\workspace_python>cat foo.txt
"hello world!"
#也可以使用记事本打开查看
D:\workspace_python>notepad foo.txt
#创建foo2.txt文件
D:\workspace_python>echo "helloworld2!" >foo2.txt
#上传文件到gridFS
D:\workspace_python>mongofiles putfoo2.txt
2017-01-11T16:35:28.767+0800 connected to: localhost
added file: foo2.txt
#列表显示所有文件内容
D:\workspace_python>mongofiles list
2017-01-11T16:35:33.935+0800 connected to: localhost
foo.txt 17
foo2.txt 18
#按照指定文件名查询文件
D:\workspace_python>mongofiles searchfoo2.txt
2017-01-11T16:35:49.086+0800 connected to: localhost
foo2.txt 18
#从gridFS中删除指定文件
D:\workspace_python>mongofiles deletefoo2.txt
2017-01-11T16:36:07.107+0800 connected to: localhost
successfully deleted all instances of'foo2.txt' from GridFS
#列表显示所有文件
D:\workspace_python>mongofiles list
2017-01-11T16:36:10.579+0800 connected to: localhost
foo.txt 17
使用python操作gridFS
示例代码如下:
#!/usr/bin/env python # -*- coding: utf-8 -*- # @Time : 2017/1/11 17:17 # @Author : Retacn # @Site : gridFS基本操作 # @File : gridfs.py # @Software: PyCharm __author__ = "retacn" __copyright__ = "property of mankind." __license__ = "CN" __version__ = "0.0.1" __maintainer__ = "retacn" __email__ = "zhenhuayue@sina.com" __status__ = "Development" from pymongo import MongoClient import gridfs client=MongoClient() db=client.test fs=gridfs.GridFS(db) #列表显示gridFS 所有文件 print(fs.list()) #上传文件 file_id=fs.put("hello world!2".encode(),filename="foo2.txt") print(fs.list()) #下载文件并读取内容 print(fs.get(file_id).read().decode())
运行结果如下:
['foo.txt', 'foo2.txt']
['foo.txt', 'foo2.txt']
hello world!2
内部原理
gridFS可以将大文件分成多块,每块可以作为单独文档存储
#取得不重复的文件名列表
服务器端角本
服务器端可以通过db.eval执行javascript脚本
> db.eval("return 1;")
WARNING: db.eval is deprecated
#3.*以上好像不建议使用
数据库的引用
DBRef,是一个内嵌文档
一个简单的例子:
#使用以下语句可以标识数据库内的任何一个文档
{“$ref”:collection,”$id”:id_value}
#如果是引用另一个数据库中的文档
{“$ref”:collection,”$id”:id_value,”$db”:database}
简单示列,代码如下:
#所有用户信息
> db.users.find()
{ "_id" :ObjectId("586b40507dfc144b68de4071"), "relationships" : {"enemies" : 2, "friend" : 32 }, "username" :"retacn yue", "age" : 33, "sex" : "ma
le", "location" :"ZiBo", "emails" : [ "zhenhuayue@hotmail.com","zhenhuayue@163.com", "zhenhuayue@soho.com","zhenhuayue@126.com" ], "state" : null,
"languages" : [ "java","python", "c", "c++" ] }
{ "_id" :ObjectId("58708a439193775a5f6b5f14"), "username" :"ann", "age" : 30, "sex" : "female","location" : "BoShan", "relationships" : {"enemies"
: 1,"friend" : 20 }, "emails" : [ "znn@163.com","ann@sina.com" ], "state" : "T","languages" : [ "java", "python", "c++"] }
{ "_id" :ObjectId("58708aeb9193775a5f6b5f16"), "username" :"mike", "age" : 15, "sex" : "male","location" : "HuanTai", "relationships" : {"enemies"
: 3,"friend" : 18 }, "emails" : [ "mike@163.com","mike@sina.com" ], "state" : "F","languages" : [ "java", "c++" ] }
{ "_id" : ObjectId("58708e5e9193775a5f6b5f18"),"username" : "andy", "age" : 18, "sex": "female", "location" : "ZhangDian","relationships" : { "enem
ies" : 2, "friend" : 15 },"emails" : [ "andy@163.com", "andy@sina.com" ],"state" : "T", "languages" : [ "java","c", "c++" ] }
{ "_id" :ObjectId("5871d6cbc1be409ee45a7899"), "username" :"retacn yue" }
#添加日记信息
>db.notes.insert({"author":"mike","text":"MongoDBis fun!"})
>db.notes.insert({"author":"andy","text":"and DBRef areeasy,too","references":[{"$ref":"users","$id":"58708aeb9193775a5f6b5f16"},{"$ref":"notes","$
id":"5876fc45ca7f92d67a4482b6"}]})
#查询所有日记信息
> db.notes.find()
{ "_id" :ObjectId("5876fc45ca7f92d67a4482b6"), "author" :"mike", "text" : "MongoDB is fun!" }
{ "_id" :ObjectId("5876fdc0ca7f92d67a4482b8"), "author" :"andy", "text" : " and DBRef are easy,too","references" : [ DBRef("users", "58708aeb919377
5a5f6b5f16"), DBRef("notes","5876fc45ca7f92d67a4482b6") ] }
#查询第二条笔记对其他文档的引用,以下可以得到用户mike和笔记”MongoDBis fun”
#$ref就是两个文档的集合,$id可以取得”_id”的值
> var note=db.notes.findOne({"_id":ObjectId("5876ff4fca7f92d67a4482b9")});
> note.references.forEach(function(ref){printjson(db[ref.$ref].findOne({"_id":ref.$id})); });
{
"_id" : ObjectId("58708aeb9193775a5f6b5f16"),
"username" : "mike",
"age" : 15,
"sex" : "male",
"location" : "HuanTai",
"relationships" : {
"enemies" : 3,
"friend" : 18
},
"emails" : [
"mike@163.com",
"mike@sina.com"
],
"state" : "F",
"languages" : [
"java",
"c++"
]
}
{
"_id" : ObjectId("5876fc45ca7f92d67a4482b6"),
"author" : "mike",
"text" : "MongoDB is fun!"
}
使用python实现以上功能
#!/usr/bin/env python # -*- coding: utf-8 -*- # @Time : 2017/1/12 12:13 # @Author : Retacn # @Site : 数据库引用DBRef # @File : mongodb_DBRef.py # @Software: PyCharm __author__ = "retacn" __copyright__ = "property of mankind." __license__ = "CN" __version__ = "0.0.1" __maintainer__ = "retacn" __email__ = "zhenhuayue@sina.com" __status__ = "Development" from pymongo import MongoClient from bson.dbref import DBRef from bson.objectid import ObjectId client = MongoClient() db = client.test note = {"author": "andy", "text": "and DBRefs are easy,too", "references": [DBRef("users", ObjectId("58708aeb9193775a5f6b5f16")), DBRef("notes", ObjectId("5876fc45ca7f92d67a4482b6"))]} # db.notes.insert(note) # 查询添加信息 db_note = db.notes.find_one({"_id": ObjectId("587708852e29db1d60e6ad8d")}) #查询记录中引用的文档内容 for dbf in db_note["references"]: print(db[dbf.collection].find_one({"_id": dbf.id}))
运行结果如下:
{'relationships': {'friend': 18.0,'enemies': 3.0}, 'emails': ['mike@163.com', 'mike@sina.com'], 'languages':['java', 'c++'], 'username': 'mike', 'sex': 'male', '_id':ObjectId('58708aeb9193775a5f6b5f16'), 'age': 15.0, 'location': 'HuanTai','state': 'F'}
{'_id':ObjectId('5876fc45ca7f92d67a4482b6'), 'text': 'MongoDB is fun!', 'author':'mike'}
文档中不推荐使用dbrefs
可以直接保存对象的_id
扫一扫
682
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
