Common query indexes of mongoDB (III)

Common query indexes of mongoDB (III)
1. _ id index is the default index created by the vast majority of sets. For each inserted data, MongoDB automatically generates a unique _ id field. > Db. jerome_2.collection.insert ({x: 2}) WriteResult ({"nInserted": 1})> db. jerome_2.collection.getIndexes () [{"v": 1, "key": {"_ id": 1}, "name": "_ id _", "ns ": "jerome. jerome_2.collection "}]> db. jerome_2.collection.findOne () {"_ id": ObjectId ("557004f1f2824fa15224e20b"), "x": 2}> 2. single-Key Index 1. single index is the most common index 2. unlike _ id indexes, single-room indexes are not automatically created.

> Db. jerome_2.collection.ensureIndex ({x: 1}) # create an index {"createdcollectionautomatautomatautomatically": false, "numIndexesBefore": 1, "numIndexesAfter": 2, "OK": 1}> db. jerome_2.collection.getIndexes () [{"v": 1, "key": {"_ id": 1}, "name": "_ id _", "ns ": "jerome. jerome_2.collection "},{" v ": 1," key ": {" x ": 1}," name ":" x_1 "," ns ":" jerome. jerome_2.collection "}]> db. jerome_2.collection.find ({x: 1}) # Use the created index to query {"_ id": ObjectId ("557005a5f2824fa15224e20c"), "x": 1, "y": 2, "z": 3}>

 

3. multi-index multi-key index and single-key index are created in the same form. The difference lies in the field value. Single-key index: the value is a single value, such as a string, number, or date. Multi-key index: the value has multiple records, such as arrays.

> Db. jerome_2.collection.getIndexes () [{"v": 1, "key": {"_ id": 1}, "name": "_ id _", "ns ": "jerome. jerome_2.collection "},{" v ": 1," key ": {" x ": 1}," name ":" x_1 "," ns ":" jerome. jerome_2.collection "}]> db. jerome_2.collection.find () {"_ id": ObjectId ("557004f1f2824fa15224e20b"), "x": 2 }{ "_ id": ObjectId ("partition"), "x ": 1, "y": 2, "z": 3}> db. jeroem_2.collection.insert ({x: [1, 2, 3, 4, 5]}) # insert an array of data. For this data, mongodb creates a multi-index WriteResult ({"nInserted ": 1 })

 

4. Composite Index

> Db. jerome_2.collection.ensureIndex ({x: 1, y: 1}) # create {"createdcollectionautomatautomatautomatically": false, "numIndexesBefore": 2, "numIndexesAfter": 3, "OK ": 1}> db. jerome_2.collection.find ({x: 1, y: 2}) # Use {"_ id": ObjectId ("557005a5f2824fa15224e20c"), "x": 1, "y": 2, "z": 3}>

 

5. Expired Index 1. Expired index: an index that will expire after a period of time. 2. After the index expires, the corresponding data will be deleted. 3. This is suitable for storing data that will expire after a period of time, such as user login information and stored logs.

> Db. expire ({time: 1 },{ expireAfterSeconds: 30}) # create an expired index. The expiration time is 30 seconds {"createdcollectionautomatautomatically": false, "numIndexesBefore": 3, "numIndexesAfter ": 4, "OK": 1}> db. jerome_2.collection.insert ({time: new Date ()}) # insert data test WriteResult ({"nInserted": 1})> db. jerome_2.collection.find () {"_ id": ObjectId ("557004f1f2824fa15224e20b"), "x": 2 }{ "_ id": ObjectId ("partition"), "x ": 1, "y": 2, "z": 3} {"_ id": ObjectId ("55700b17f2824fa15224e20e"), "time": ISODate ("2015-06-04T08: 23: 51.531Z ")}> db. jerome_2.collection.find () {"_ id": ObjectId ("557004f1f2824fa15224e20b"), "x": 2 }{ "_ id": ObjectId ("partition"), "x ": 1, "y": 2, "z": 3} {"_ id": ObjectId ("55700b17f2824fa15224e20e"), "time": ISODate ("2015-06-04T08: 23: 51.531Z ")}> db. jerome_2.collection.find () # {"_ id": ObjectId ("557004f1f2824fa15224e20b"), "x": 2} {"_ id ": objectId ("557005a5f2824fa15224e20c"), "x": 1, "y": 2, "z": 3}>

 

Use Limit 1. The value stored in the expired index field must be of the specified time type. (It must be an ISODate or ISODate array and cannot use timestamp; otherwise, it cannot be automatically deleted) 2. If the ISODate array is specified, it will be deleted at the minimum time. 3. the expired index cannot be a composite index. 4. The deletion time is not accurate. (The deletion process is not run once every 60 s by the background program, and the deletion takes some time, so there is an error) 6. full-text index: {author: "", titile; "", article: ""} create method 1. db. articles. ensureIndex ({key: "text"}) 2. db. articles. ensureIndex ({key_1: "text", key_2: "text"}) 3. db. articles. ensureIndex ({"$ **": "text"}) use 1. db. articles. find ({$ text: {$ search: "aa"}) 2. db. articles. find ({$ text: {$ search: "aa bb cc"}) # space or 3. db. articles. find ({$ text: {$ search: "aa bb-cc"}) #-cc indicates that cc 4 is not included. db. articles. find ({$ text :{$ search: "\" aa \ "bb cc" }}) # ", enclosed by quotation marks, indicating the following example:

> Db. jerome_2.ensureIndex ({"article": "text"}) # create a full-text index {"createdcollectionautomatautomatically": true, "numIndexesBefore": 1, "numIndexesAfter": 2, "OK ": 1}> db. jerome_2.insert ({"article": "aa bb cc dd ee"}) # insert test data WriteResult ({"nInserted": 1})> db. jerome_2.insert ({"article": "aa bb rr gg zz"}) WriteResult ({"nInserted": 1})> db. jerome_2.insert ({"article": "aa bb"}) WriteResult ({"nInserted": 1})> db. jerome_2.insert ({"article": "aa bb cc zz ff ww"}) WriteResult ({"nInserted": 1})> db. jerome_2.find ({$ text: {$ search: "aa" }}) # Find {"_ id": ObjectId ("5572904271c0bbd90f4ce0e2"), "article ": "aa bb rr gg zz"} {"_ id": ObjectId ("5572903371c0bbd90f4ce0e1"), "article": "aa bb cc dd ee"} {"_ id ": objectId ("5572905671c0bbd90f4ce0e4"), "article": "aa bb cc zz ff ww"} {"_ id": ObjectId ("5572904771c0bbd90f4ce0e3"), "article ": "aa bb"}> db. jerome_2.find ({$ text: {$ search: "ff" }}) {"_ id": ObjectId ("5572905671c0bbd90f4ce0e4"), "article ": "aa bb cc zz ff ww"}> db. jerome_2.find ({$ text: {$ search: "aa bb cc"}) {"_ id": ObjectId ("5572904271c0bbd90f4ce0e2"), "article ": "aa bb rr gg zz"} {"_ id": ObjectId ("5572903371c0bbd90f4ce0e1"), "article": "aa bb cc dd ee"} {"_ id ": objectId ("5572905671c0bbd90f4ce0e4"), "article": "aa bb cc zz ff ww"} {"_ id": ObjectId ("5572904771c0bbd90f4ce0e3"), "article ": "aa bb"}> db. jerome_2.find ({$ text: {$ search: "aa bb-cc"}) {"_ id": ObjectId ("5572904271c0bbd90f4ce0e2"), "article ": "aa bb rr gg zz"} {"_ id": ObjectId ("5572904771c0bbd90f4ce0e3"), "article": "aa bb"}> db. jerome_2.find ({$ text: {$ search: "\" aa \ "\" bb \ "\" cc \ ""}) {"_ id ": objectId ("5572903371c0bbd90f4ce0e1"), "article": "aa bb cc dd ee"} {"_ id": ObjectId ("5572905671c0bbd90f4ce0e4"), "article ": "aa bb cc zz ff ww"}>

 

Full-text index similarity $ meta OPERATOR: {score: {$ meta: "textScore"}. After uninstalling the query condition, you can regret the similarity of the returned results, it can be used with sort to achieve good practical results.

> Db. jerome_2.find ({$ text: {$ search: "aa bb" }}, {score: {$ meta: "textScore"}) # The higher the score, the higher the similarity. {"_ Id": ObjectId ("5572904271c0bbd90f4ce0e2"), "article": "aa bb rr gg zz", "score": 1.2} {"_ id ": objectId ("5572903371c0bbd90f4ce0e1"), "article": "aa bb cc dd ee", "score": 1.2} {"_ id": ObjectId ("5572905671c0bbd90f4ce0e4 "), "article": "aa bb cc zz ff ww", "score": 1.1666666666666667} {"_ id": ObjectId ("5572904771c0bbd90f4ce0e3"), "article ": "aa bb", "score": 1.5}> db. jerome_2.find ({$ text: {$ search: "aa bb" }}, {score: {$ meta: "textScore "}}). sort ({score: {$ meta: "textScore"}) # sort by score {"_ id": ObjectId ("5572904771c0bbd90f4ce0e3"), "article ": "aa bb", "score": 1.5} {"_ id": ObjectId ("5572903371c0bbd90f4ce0e1"), "article": "aa bb cc dd ee", "score ": 1.2} {"_ id": ObjectId ("5572904271c0bbd90f4ce0e2"), "article": "aa bb rr gg zz", "score": 1.2} {"_ id ": objectId ("5572905671c0bbd90f4ce0e4"), "article": "aa bb cc zz ff ww", "score": 1.1666666666666667}>

 

Limits 1. only one $ text query can be specified for each query. 2. $ text query cannot appear in $ nor query 3. if $ text is included in the query, hint no longer works. mongoDB does not support Chinese now. geographic location index concept: stores the locations of some points in MongoDB. After creating an index, you can search for other points by location. Category: 1. 2d index, used to store and find points on the plane 2. 2dsphere index, used to store and find vertices on the sphere: 1. find a point within a certain distance from a point. 2. Search for points contained in a region. 2d index creation: db. location. ensureIndex ({"w": "2d"}) Position representation: longitude and latitude [longitude, latitude] value range: longitude [-180,180] Latitude [-90,90] query method 1. $ near query: query the closest vertex 2. $ geoWithin query: query the vertex 1 in a certain shape. $ near Query

> Db. location. ensureIndex ({"w": "2d"}) # create a 2d index> db. location. insert ({w: [1, 1]}) # insert test data WriteResult ({"nInserted": 1})> db. location. insert ({w: [1, 2]}) WriteResult ({"nInserted": 1})> db. location. insert ({w: [2, 3]}) WriteResult ({"nInserted": 1})> db. location. insert ({w: [100,80]}) WriteResult ({"nInserted": 1})> db. location. find ({w :{$ near: [100]}) # Will be returned. Your latest vertex {"_ id": ObjectId ("5572a961aba41684d6e8826c "), "w": [1, 1]} {"_ id": ObjectId ("5572a965aba41684d6e8826d"), "w": [1, 2]} {"_ id ": objectId ("5572a970aba41684d6e8826e"), "w": [2, 3]} {"_ id": ObjectId ("5572a97aaba41684d6e8826f"), "w": [100, 80]}> db. location. find ({w :{$ near: [], $ maxDistance: 2 }}) # maxDistance restrictions can be used (near cannot use minDistance) {"_ id ": objectId ("dimensions"), "w": [1, 1]} {"_ id": ObjectId ("5572a965aba41684d6e8826d"), "w": [1, 2]}

 

2. $ geoWithin: 1. $ box: rectangle. Use {$ box: [[<x1>, <y1>], [<x2>, <y2>]} to represent 2. $ center: circle. Use {$ center: [<x1>, <y1>], r]} to indicate 3. $ polygon: polygon. Use {$ polygon: [[<x1>, <y1>], [<x2>, <y2>], [<x3>, <y3>]} indicates an example:

> Db. location. find ({w: {$ geoWithin: {$ box: [[0, 0], [3, 3] }}) # rectangle {"_ id ": objectId ("5572a965aba41684d6e8826d"), "w": [1, 2]} {"_ id": ObjectId ("5572a970aba41684d6e8826e"), "w": [2, 3]} {"_ id": ObjectId ("5572a961aba41684d6e8826c"), "w": [1, 1]}> db. location. find ({w: {$ geoWithin :{$ box: [[1, 2], [2, 3] }}) {"_ id": ObjectId ("5572a965aba41684d6e8826d "), "w": [1, 2]} {"_ id": ObjectId ("5572a970aba41684d6e8826e"), "w": [2, 3]}> db. location. find ({w: {$ geoWithin: {$ center: [[100], 100] }}) # Circle, is the radius {"_ id ": objectId ("5572a961aba41684d6e8826c"), "w": [1, 1]} {"_ id": ObjectId ("5572a970aba41684d6e8826e"), "w": [2, 3]} {"_ id": ObjectId ("5572a965aba41684d6e8826d"), "w": [1, 2]}> db. location. find ({w: {$ geoWithin :{$ center: [[1000],] }}) {"_ id": ObjectId ("5572a961aba41684d6e8826c "), "w": [1, 1]} {"_ id": ObjectId ("5572a97aaba41684d6e8826f"), "w": [100, 80]} {"_ id ": objectId ("5572a970aba41684d6e8826e"), "w": [2, 3]} {"_ id": ObjectId ("5572a965aba41684d6e8826d"), "w": [1, 2]}> db. location. find ({w: {$ geoWithin: {$ polygon: [[], [], [], []}) # polygon query (the range of polygon enclosed by each vertex) {"_ id": ObjectId ("5572a970aba41684d6e8826e"), "w": [2, 3]} {"_ id": ObjectId ("5572a961aba41684d6e8826c"), "w": [1, 1]} {"_ id": ObjectId ("watermark "), "w": [1, 2]}> db. location. find ({w: {$ geoWithin: {$ polygon: [[], [], [], [], []}) {"_ id": ObjectId ("5572a970aba41684d6e8826e"), "w": [2, 3]} {"_ id": ObjectId ("identifier"), "w ": [100, 80]} {"_ id": ObjectId ("5572a961aba41684d6e8826c"), "w": [1, 1]} {"_ id ": objectId ("5572a965aba41684d6e8826d"), "w": [1, 2]}

 

3. Use the runCommand command to query geoNear.

Db. runCommand ({getNear: <collection>, near: [x, y], minDistance :( invalid for 2d indexes) maxDistance: num:})> db. runCommand ({geoNear: "location", near: [1, 2], maxDistance: 10, num: 1}) {"results": [{"dis": 0, "obj": {"_ id": ObjectId ("5572a965aba41684d6e8826d"), "w": [1, 2]}], "stats": {"nscanned ": numberLong (1), "objectsLoaded": NumberLong (1), "avgDistance": 0, "maxDistance": 0, "time": 2}, "OK": 1}>

 

Spherical geographic location index concept: spherical geographic location index creation method: db. collection. ensureIndex ({w: "2 dsphere"}) Location representation: GeoJson: describes a point, a straight line, polygon, and other shape formats: {type: "", coordinates: [<coordinates>]} supports $ minDistance and $ maxDistance for index creation. Description of important attributes in db format. collection. ensureIndex ({param}, {param}) # The second parameter indicates the index attributes. Important attributes include name, uniqueness, sparsity, and whether to delete the index regularly. 1. name: db. collection. the default naming format of ensureIndex ({}, {name: ""}) is as follows:

> db.jerome_2.ensureIndex({x:1})> db.jerome_2.ensureIndex({y:-1})> db.jerome_2.getIndexes()[    {        "v" : 1,        "key" : {            "_id" : 1        },        "name" : "_id_",        "ns" : "jerome.jerome_2"    },    {        "v" : 1,        "key" : {            "x" : 1        },        "name" : "x_1",        "ns" : "jerome.jerome_2"    },    {        "v" : 1,        "key" : {            "y" : -1        },        "name" : "y_-1",        "ns" : "jerome.jerome_2"    }

 

Composite Index

> db.jerome_2.ensureIndex({x:1,y:-1})> db.jerome_2.ensureIndex({x:1,y:-1,z:1})> db.jerome_2.getIndexes()[    {        "v" : 1,        "key" : {            "x" : 1,            "y" : -1        },        "name" : "x_1_y_-1",        "ns" : "jerome.jerome_2"    },    {        "v" : 1,        "key" : {            "x" : 1,            "y" : -1,            "z" : 1        },        "name" : "x_1_y_-1_z_1",        "ns" : "jerome.jerome_2"    }]

Mongodb limits the index to 125 bytes, so we need to customize the index name.

> db.jerome_2.ensureIndex({x:1,y:1,z:1,m:1},{name:"normal_index"})> db.jerome_2.getIndexes()[    {        "v" : 1,        "key" : {            "x" : 1,            "y" : 1,            "z" : 1,            "m" : 1        },        "name" : "normal_index",        "ns" : "jerome.jerome_2"    }]> db.jerome_2.dropIndex("normal_index"){ "nIndexesWas" : 7, "ok" : 1 }> 

 

2. uniqueness, specified by unique: db. collection. ensureIndex ({}, {unique: true/false })

> db.jerome.ensureIndex({m:1,n:1},{unique:true}){    "createdCollectionAutomatically" : true,    "numIndexesBefore" : 1,    "numIndexesAfter" : 2,    "ok" : 1}> db.jerome.insert({m:1,n:2})WriteResult({ "nInserted" : 1 })> db.jerome.insert({m:1,n:2})WriteResult({    "nInserted" : 0,    "writeError" : {        "code" : 11000,        "errmsg" : "insertDocument :: caused by :: 11000 E11000 duplicate key error index: jerome.jerome.$m_1_n_1  dup key: { : 1.0, : 2.0 }"    }})> 

 

3. sparse: Specify db. collection. ensureIndex ({},{ sparse: true/false}). Indexes created by default are not sparse. Because MongoDB does not have a fixed format, non-existent fields may be inserted during insertion, such as x: 1. MongoDB will create an index for this non-existent field, if you do not want to find such a thing, you can specify the sparse index to true, and no index will be created for non-existent fields. This can reduce the disk space temporarily and increase the insertion speed.

> Db. jerome. insert ({"m": 1}) WriteResult ({"nInserted": 1})> db. jerome. insert ({"n": 1}) WriteResult ({"nInserted": 1})> db. jerome. find ({m: {$ exists: true}) # exists searches for records with the existence or absence of a field in the dataset {"_ id": ObjectId ("55729ec1aba41684d6e8826a "), "m": 1} {"_ id": ObjectId ("55729d5caba41684d6e88268"), "m": 1, "n": 2}> db. jerome. ensureIndex ({m: 1}, {sparse: true}) # create a sparse index {"createdcollectionautomatautomatically": false, "numIndexesBefore": 2, "numIndexesAfter": 3, "OK": 1}> db. jerome. find ({m: {$ exists: false}) # find an internal MongoDB problem. Use the following method to forcibly specify the index {"_ id": ObjectId ("55729ec7aba41684d6e8826b "), "n": 1}> db. jerome. find ({m :{$ exists: false }}). hint ("m_1") # The following record does not contain m fields, so no index is created, so no record is found>

 

Note: you cannot search for records that do not exist in this field on the sparse index.