绝好的MongoDB学习资料之三. Schema Design

绝好的MongoDB学习资料之三. Schema Design

 

1. Document-Oriented

MongoDB 是一种面向文档(document-oriented)的数据库，其内存储的是一种 JSON-like 结构化数据。尽管拥有和关系型数据库 Database/Table 类似的的 DB/Collection 概念，但同一 Collection 内的 Document 可以拥有不同的属性。

(注: 以下 > 提示符表示 mongo JS 代码，>>> 为 Python 代码)

 

1. > use blog

2. switched to db blog

3.  

4. > db.users.insert({name:"user1", age:15})

5. > db.users.insert({name:"user2", age:20, sex:1})

6.  

7. > db.users.find()

8. { "_id" : ObjectId("4c479885089df9b53474170a"), "name" : "user1", "age" : 15 }

9. { "_id" : ObjectId("4c479896089df9b53474170b"), "name" : "user2", "age" : 20, "sex" : 1 }

 

可以通过 $exists 判断某个字段是否存在。

 

1. > db.users.find({sex:{$exists:true}})

2. { "_id" : ObjectId("4c479896089df9b53474170b"), "name" : "user2", "age" : 20, "sex" : 1 }

 

2. Embed vs. Reference

Document 采取 JSON-like 这种层级结构，因此我们可以直接用嵌入(Embed)代替传统关系型数据库的关联引用(Reference)。

 

1. > u = db.users.findOne({name:"user1"})

2. {

3. "_id" : ObjectId("4c479885089df9b53474170a"),

4. "name" : "user1",

5. "age" : 15

6. }

7.  

8. > u.address = ["address1", "address2"]

9. [ "address1", "address2" ]

10.  

11. > db.users.save(u)

12.  

13. > db.users.findOne({name:"user1"})

14. {

15. "_id" : ObjectId("4c479885089df9b53474170a"),

16. "name" : "user1",

17. "age" : 15,

18. "address" : [

19. "address1",

20. "address2"

21. ]

22. }

23.  

24. > db.users.findOne({address:"address1"})

25. {

26. "_id" : ObjectId("4c479885089df9b53474170a"),

27. "name" : "user1",

28. "age" : 15,

29. "address" : [

30. "address1",

31. "address2"

32. ]

33. }

 

MongoDB 支持以 "." 分割的 namespace 路径，但需要注意 key 不能以 "$" 开头，不能包含 "." 字符 (条件表达式中的多级路径须用引号)。

 

1. > u = db.users.findOne({address:"address1"})

2. {

3. "_id" : ObjectId("4c479885089df9b53474170a"),

4. "name" : "user1",

5. "age" : 15,

6. "address" : [

7. "address1",

8. "address2"

9. ]

10. }

11.  

12. > u.im = {msn:"user1@hotmail.com", qq:12345678}

13. { "msn" : "user1@hotmail.com", "qq" : 12345678 }

14.  

15. > db.users.save(u)

16.  

17. > u = db.users.findOne({"im.qq":12345678})

18. {

19. "_id" : ObjectId("4c479885089df9b53474170a"),

20. "name" : "user1",

21. "age" : 15,

22. "address" : [

23. "address1",

24. "address2"

25. ],

26. "im" : {

27. "msn" : "user1@hotmail.com",

28. "qq" : 12345678

29. }

30. }

31.  

32. > u.im.qq

33. 12345678

34.  

35. > u.im.msn

36. user1@hotmail.com

37.  

38. > db.users.update({"im.qq":12345678}, {$set:{"im.qq":12345}})

39.  

40. > u = db.users.findOne({"im.qq":12345})

41. {

42. "_id" : ObjectId("4c479885089df9b53474170a"),

43. "name" : "user1",

44. "age" : 15,

45. "address" : [

46. "address1",

47. "address2"

48. ],

49. "im" : {

50. "msn" : "user1@hotmail.com",

51. "qq" : 12345

52. }

53. }

54.  

55. > u = db.users.find({"im.qq":{$exists:true}}, {"im.qq":1})

56. { "_id" : ObjectId("4c479885089df9b53474170a"), "im" : { "qq" : 12345 } }

 

注意: 由于每篇文档都包含完整的 key 数据，因此使用尽可能短的 key 可以有效节省存储空间。

3. Operator

(1) $all: 判断数组属性是否包含全部条件。

 

1. > db.users.insert({name:"user3", data:[1,2,3,4,5,6,7]})

2. > db.users.insert({name:"user4", data:[1,2,3]})

3.  

4. > db.users.find({data:{$all:[2,3,4]}})

5. { "_id" : ObjectId("4c47a133b48cde79c6780df0"), "name" : "user3", "data" : [ 1, 2, 3, 4, 5, 6, 7 ] }

 

注意和 $in 的区别。$in 是检查目标属性值是条件表达式中的一员，而 $all 则要求属性值包含全部条件元素。

(2) $size: 匹配数组属性元素数量。

 

1. > db.users.insert({name:"user3", data:[1,2,3,4,5,6,7]})

2. > db.users.insert({name:"user4", data:[1,2,3]})

3.  

4. > db.users.find({data:{$size:3}})

5. { "_id" : ObjectId("4c47a13bb48cde79c6780df1"), "name" : "user4", "data" : [ 1, 2, 3 ] }

 

(3) $type: 判断属性类型。

 

1. > db.users.insert({name:"user5", t:1})

2. > db.users.insert({name:"user6", t:1.34})

3. > db.users.insert({name:"user7", t:"abc"})

4.  

5. > db.users.find({t:{$type:1}})

6. { "_id" : ObjectId("4c47a231b48cde79c6780df2"), "name" : "user5", "t" : 1 }

7. { "_id" : ObjectId("4c47a23eb48cde79c6780df3"), "name" : "user6", "t" : 1.34 }

8.  

9. > db.users.find({t:{$type:2}})

10. { "_id" : ObjectId("4c47a258b48cde79c6780df4"), "name" : "user7", "t" : "abc" }

 

类型值:

• double:1
• string: 2
• object: 3
• array: 4
• binary data: 5
• object id: 7
• boolean: 8
• date: 9
• null: 10
• regular expression: 11
• javascript code: 13
• symbol: 14
• javascript code with scope: 15
• 32-bit integer: 16
• timestamp: 17
• 64-bit integer: 18
• min key: 255
• max key: 127

(4) 使用正则表达式进行查询。

 

 

1. > db.users.find({name:/user[135]/i}, {name:1})

2. { "_id" : ObjectId("4c479885089df9b53474170a"), "name" : "user1" }

3. { "_id" : ObjectId("4c47a133b48cde79c6780df0"), "name" : "user3" }

4. { "_id" : ObjectId("4c47a231b48cde79c6780df2"), "name" : "user5" }

5.  

6. >>> users = db.users.find({"name" : {"$regex" : r"(?i)user[135]"}}, ["name"])

7. >>> for u in users: print u

8. ...

9. {u'_id': ObjectId('4c479885089df9b53474170a'), u'name': u'user1'}

10. {u'_id': ObjectId('4c47a133b48cde79c6780df0'), u'name': u'user3'}

11. {u'_id': ObjectId('4c47a231b48cde79c6780df2'), u'name': u'user5'}

 

正则表达式标记:

i: 忽略大小写。
m: 默认为单行处理，此标记表示多行。
x: 扩展。

(5) 数组属性元素值匹配。

 

1. > db.users.find({data:"abc"})

2. { "_id" : ObjectId("4c47a481b48cde79c6780df5"), "name" : "user8", "data" : [ { "a" : 1, "b" : 10 }, 3, "abc" ] }

3.  

4. > db.users.find({data:{$elemMatch:{a:1, b:{$gt:5}}}})

5. { "_id" : ObjectId("4c47a481b48cde79c6780df5"), "name" : "user8", "data" : [ { "a" : 1, "b" : 10 }, 3, "abc" ] }

 

{data:"abc"} 仅简单匹配数组属性是否包含该元素。$elemMatch 则可以处理更复杂的元素查找条件。当然也可以写成如下方式。

 

1. > db.users.find({"data.a":1, "data.b":{$gt:5}})

2. { "_id" : ObjectId("4c47a481b48cde79c6780df5"), "name" : "user8", "data" : [ { "a" : 1, "b" : 10 }, 3, "abc" ] }

 

还可以直接使用序号进行操作。

 

1. > db.users.find({"data.2":3})

2. { "_id" : ObjectId("4c47a133b48cde79c6780df0"), "name" : "user3", "data" : [ 1, 2, 3, 4, 5, 6, 7 ] }

3. { "_id" : ObjectId("4c47a13bb48cde79c6780df1"), "name" : "user4", "data" : [ 1, 2, 3 ] }

 

(6) $not: 取反，表示返回条件不成立的文档。

似乎只能跟正则和 $mod 一起使用？？？？

 

1. > u = db.users.find({name:{$not:/user3/}}, {name:1})

2. { "_id" : ObjectId("4c479885089df9b53474170a"), "name" : "user1" }

3. { "_id" : ObjectId("4c47a13bb48cde79c6780df1"), "name" : "user4" }

4. { "_id" : ObjectId("4c47a231b48cde79c6780df2"), "name" : "user5" }

5. { "_id" : ObjectId("4c47a23eb48cde79c6780df3"), "name" : "user6" }

6. { "_id" : ObjectId("4c47a258b48cde79c6780df4"), "name" : "user7" }

7. { "_id" : ObjectId("4c47a481b48cde79c6780df5"), "name" : "user8" }

8. { "_id" : ObjectId("4c479896089df9b53474170b"), "name" : "user2" }

 

(7) $unset: 和 $set 相反，表示移除文档属性。

 

1. > u = db.users.find({name:"user1"})

2. { "_id" : ObjectId("4c479885089df9b53474170a"), "name" : "user1", "age" : 15, "address" : [ "address1", "address2" ], "im" : { "msn" : "user1@hotmail.com", "qq" : 12345 } }

3.  

4. > db.users.update({name:"user1"}, {$unset:{address:1, im:1}})

5.  

6. > u = db.users.find({name:"user1"})

7. { "_id" : ObjectId("4c479885089df9b53474170a"), "age" : 15, "name" : "user1" }

 

(8) $push: 和 $ pushAll 都是向数组属性添加元素。

 

1. > u = db.users.find({name:"user1"})

2. { "_id" : ObjectId("4c479885089df9b53474170a"), "age" : 15, "name" : "user1" }

3.  

4. > db.users.update({name:"user1"}, {$push:{data:1}})

5.  

6. > u = db.users.find({name:"user1"})

7. { "_id" : ObjectId("4c479885089df9b53474170a"), "age" : 15, "data" : [ 1 ], "name" : "user1" }

8.  

9. > db.users.update({name:"user1"}, {$pushAll:{data:[2,3,4,5]}})

10.  

11. > u = db.users.find({name:"user1"})

12. { "_id" : ObjectId("4c479885089df9b53474170a"), "age" : 15, "data" : [ 1, 2, 3, 4, 5 ], "name" : "user1" }

 

(9) $addToSet: 和 $push 类似，不过仅在该元素不存在时才添加 (Set 表示不重复元素集合)。

 

1. > db.users.update({name:"user2"}, {$addToSet:{data:1}})

2. > db.users.update({name:"user2"}, {$addToSet:{data:1}})

3.  

4. > u = db.users.find({name:"user2"})

5. { "_id" : ObjectId("4c479896089df9b53474170b"), "data" : [ 1 ], "name" : "user2" }

6. > db.users.update({name:"user2"}, {$push:{data:1}})

7.  

8. > u = db.users.find({name:"user2"})

9. { "_id" : ObjectId("4c479896089df9b53474170b"), "data" : [ 1, 1 ], "name" : "user2" }

 

要添加多个元素，使用 $each。

 

1. > db.users.update({name:"user2"}, {$addToSet:{data:{$each:[1,2,3,4]}}})

2.  

3. > u = db.users.find({name:"user2"})

4. { "_id" : ObjectId("4c479896089df9b53474170b"), "data" : [ 1, 2, 3, 4 ], "name" : "user2" }

 

(10) $pop: 移除数组属性的元素，$pull 按值移除，$pullAll 移除所有符合提交的元素。

 

1. > u = db.users.find({name:"user3"})

2. { "_id" : ObjectId("4c47a133b48cde79c6780df0"), "data" : [ 1, 2, 3, 4, 5, 6, 7, 2, 3 ], "name" : "user3" }

3.  

4. > db.users.update({name:"user3"}, {$pop:{data:1}}) // 移除最后一个元素

5.  

6. > u = db.users.find({name:"user3"})

7. { "_id" : ObjectId("4c47a133b48cde79c6780df0"), "data" : [ 1, 2, 3, 4, 5, 6, 7, 2 ], "name" : "user3" }

8.  

9. > db.users.update({name:"user3"}, {$pop:{data:-1}}) // 移除第一个元素

10.  

11. > u = db.users.find({name:"user3"})

12. { "_id" : ObjectId("4c47a133b48cde79c6780df0"), "data" : [ 2, 3, 4, 5, 6, 7, 2 ], "name" : "user3" }

13.  

14. > db.users.update({name:"user3"}, {$pull:{data:2}}) // 移除全部 2

15.  

16. > u = db.users.find({name:"user3"})

17. { "_id" : ObjectId("4c47a133b48cde79c6780df0"), "data" : [ 3, 4, 5, 6, 7 ], "name" : "user3" }

18.  

19. > db.users.update({name:"user3"}, {$pullAll:{data:[4,5,6]}}) // 移除 4,5,6

20.  

21. > u = db.users.find({name:"user3"})

22. { "_id" : ObjectId("4c47a133b48cde79c6780df0"), "data" : [ 3, 7 ], "name" : "user3" }

 

(11) $where: 用 JS 代码来代替有些丑陋的 $lt、$gt。

MongoDB 内置了 Javascript Engine (SpiderMonkey)。可直接使用 JS Expression，甚至使用 JS Function 写更复杂的 Code Block。

 

1. > for (var i = 0; i < 10; i++) db.users.insert({name:"user"+i, age:i})

2.  

3. > db.users.find()

4. { "_id" : ObjectId("4c47b3372a9b2be866da226e"), "name" : "user0", "age" : 0 }

5. { "_id" : ObjectId("4c47b3372a9b2be866da226f"), "name" : "user1", "age" : 1 }

6. { "_id" : ObjectId("4c47b3372a9b2be866da2270"), "name" : "user2", "age" : 2 }

7. { "_id" : ObjectId("4c47b3372a9b2be866da2271"), "name" : "user3", "age" : 3 }

8. { "_id" : ObjectId("4c47b3372a9b2be866da2272"), "name" : "user4", "age" : 4 }

9. { "_id" : ObjectId("4c47b3372a9b2be866da2273"), "name" : "user5", "age" : 5 }

10. { "_id" : ObjectId("4c47b3372a9b2be866da2274"), "name" : "user6", "age" : 6 }

11. { "_id" : ObjectId("4c47b3372a9b2be866da2275"), "name" : "user7", "age" : 7 }

12. { "_id" : ObjectId("4c47b3372a9b2be866da2276"), "name" : "user8", "age" : 8 }

13. { "_id" : ObjectId("4c47b3372a9b2be866da2277"), "name" : "user9", "age" : 9 }

14.  

15. > db.users.find({$where:"this.age > 7 || this.age < 3"})

16. { "_id" : ObjectId("4c47b3372a9b2be866da226e"), "name" : "user0", "age" : 0 }

17. { "_id" : ObjectId("4c47b3372a9b2be866da226f"), "name" : "user1", "age" : 1 }

18. { "_id" : ObjectId("4c47b3372a9b2be866da2270"), "name" : "user2", "age" : 2 }

19. { "_id" : ObjectId("4c47b3372a9b2be866da2276"), "name" : "user8", "age" : 8 }

20. { "_id" : ObjectId("4c47b3372a9b2be866da2277"), "name" : "user9", "age" : 9 }

21.  

22. > db.users.find("this.age > 7 || this.age < 3")

23. { "_id" : ObjectId("4c47b3372a9b2be866da226e"), "name" : "user0", "age" : 0 }

24. { "_id" : ObjectId("4c47b3372a9b2be866da226f"), "name" : "user1", "age" : 1 }

25. { "_id" : ObjectId("4c47b3372a9b2be866da2270"), "name" : "user2", "age" : 2 }

26. { "_id" : ObjectId("4c47b3372a9b2be866da2276"), "name" : "user8", "age" : 8 }

27. { "_id" : ObjectId("4c47b3372a9b2be866da2277"), "name" : "user9", "age" : 9 }

28.  

29. > db.users.find({$where: function(){ return this.age > 7 || this.age < 3;}})

30. { "_id" : ObjectId("4c47b3372a9b2be866da226e"), "name" : "user0", "age" : 0 }

31. { "_id" : ObjectId("4c47b3372a9b2be866da226f"), "name" : "user1", "age" : 1 }

32. { "_id" : ObjectId("4c47b3372a9b2be866da2270"), "name" : "user2", "age" : 2 }

33. { "_id" : ObjectId("4c47b3372a9b2be866da2276"), "name" : "user8", "age" : 8 }

34. { "_id" : ObjectId("4c47b3372a9b2be866da2277"), "name" : "user9", "age" : 9 }

35.  

36. >>> for u in db.users.find({"$where":"this.age > 7 || this.age < 3"}): print u

37. ...

38. {u'age': 0.0, u'_id': ObjectId('4c47b3372a9b2be866da226e'), u'name': u'user0'}

39. {u'age': 1.0, u'_id': ObjectId('4c47b3372a9b2be866da226f'), u'name': u'user1'}

40. {u'age': 2.0, u'_id': ObjectId('4c47b3372a9b2be866da2270'), u'name': u'user2'}

41. {u'age': 8.0, u'_id': ObjectId('4c47b3372a9b2be866da2276'), u'name': u'user8'}

42. {u'age': 9.0, u'_id': ObjectId('4c47b3372a9b2be866da2277'), u'name': u'user9'}

43.  

44. >>> for u in db.users.find().where("this.age > 7 || this.age < 3"): print u

45. ...

46. {u'age': 0.0, u'_id': ObjectId('4c47b3372a9b2be866da226e'), u'name': u'user0'}

47. {u'age': 1.0, u'_id': ObjectId('4c47b3372a9b2be866da226f'), u'name': u'user1'}

48. {u'age': 2.0, u'_id': ObjectId('4c47b3372a9b2be866da2270'), u'name': u'user2'}

49. {u'age': 8.0, u'_id': ObjectId('4c47b3372a9b2be866da2276'), u'name': u'user8'}

50. {u'age': 9.0, u'_id': ObjectId('4c47b3372a9b2be866da2277'), u'name': u'user9'}

51.  

52. >>> for u in db.users.find().where("function() { return this.age > 7 || this.age < 3;}"): print u

53. ...

54. {u'age': 0.0, u'_id': ObjectId('4c47b3372a9b2be866da226e'), u'name': u'user0'}

55. {u'age': 1.0, u'_id': ObjectId('4c47b3372a9b2be866da226f'), u'name': u'user1'}

56. {u'age': 2.0, u'_id': ObjectId('4c47b3372a9b2be866da2270'), u'name': u'user2'}

57. {u'age': 8.0, u'_id': ObjectId('4c47b3372a9b2be866da2276'), u'name': u'user8'}

58. {u'age': 9.0, u'_id': ObjectId('4c47b3372a9b2be866da2277'), u'name': u'user9'}