MongoDB Collection

Collection in MongoDB is an analog of a table in a relational database. It is a group of documents. Each document in a collection can have different structures. A collection can contain up to 64 indexes.

MongoDB Create Collection

MongoDB createCollection() method is used to create a collection in mongodb.

Syntax of createCollection()

db.createCollection(name, options)

Here, name is the name of the collection and of a string type and options is an optional parameter.

The options parameter can be the following -

Option	Type	Description
capped	boolean	Specify true to create a capped collection.
autoIndexId	boolean	Specify false to disable the automatic creation of an index.
size	number	Specify a maximum size in bytes for a capped collection.
max	number	The maximum number of documents allowed in the capped collection.
storageEngine	document	Available for the WiredTiger storage engine only.
validator	document	Allow users to specify validation rules or expressions for the collection.
viewOn	string	The name of the source collection or view from which to create the view.
pipeline	array	An array that consists of the aggregation pipeline stage.
collation	document	Specifies the default collation for the collection.
writeConcern	document	A document that expresses the write-concern for the operation.

Example

db.createCollection("students", { capped : true, size : 5242880, max : 3000 } )

We can also create a collection when documents are first inserted. This is the dynamic behaviour of MongoDB. Like, in this example, we need to store a document in a collection 'employee'. So before creating this, we can directly insert the document in the 'employee' collection.

> use company;
switched to db company
> db.employee.insert({name:"Priska",dept:"IT",role:"developer"});
WriteResult({ "nInserted" : 1 })
> db.employee.find();
{ "_id" : ObjectId("5d414003092ef252dc155a6f"), "name" : "Priska", "dept" : "IT", "role" : "developer" }
>

MongoDB List Collections

MongoDB show collections display a list of all defined collections on the current selected database.
In the below command, we have selected 'school' database and listed all its collections.

> use school;
switched to db school
> show collections
attendence
marks
students
>

MongoDB Collection Statistics

We can know about individual collection statistics using stats() method. This is helpful in complicated debugging.

> db.attendence.stats()
{
	"ns" : "school.attendence",
	"size" : 53,
	"count" : 1,
	"avgObjSize" : 53,
	"storageSize" : 16384,
	"capped" : false,
	............
	...........

MongoDB Count Document

To count the number of documents in a collection, simply write the following command -

> db.students.count();
5

Size of Collection

The dataSize() method returns the size of the given collection.

> db.students.dataSize();
345

Size of Document in Collection

The storageSize() method returns the total size of documents stored in the collection.

> db.students.storageSize();
36864

Size of all indexes in a Collection

The totalIndexSize() method returns the total size of all indexes in the given collection.

> db.students.totalIndexSize();
36864

Collection Latency Statistics

MongoDB latencyStats() method returns latency statistics for read requests, write requests and database commands for the given collection.

db.attendence.latencyStats().pretty()
{
	"ns" : "school.students",
	"host" : "DESKTOP-NAME",
	"localTime" : ISODate("2019-08-02T11:03:48.966Z"),
	"latencyStats" : {
		"reads" : {
			"latency" : NumberLong(193999),
			"ops" : NumberLong(3)
		},
		"writes" : {
			"latency" : NumberLong(0),
			"ops" : NumberLong(0)
		},
		"commands" : {
			"latency" : NumberLong(242811),
			"ops" : NumberLong(3)
		},
		"transactions" : {
			"latency" : NumberLong(0),
			"ops" : NumberLong(0)
		}
	}
}

MongoDB Rename Collection

MongoDB provides renameCollection() method to rename the existing collection.

db.students.renameCollection("undergrad");
{ "ok" : 1 }

MongoDB Delete Collection

MongoDB drop() method is used to delete the existing collection. It returns true on success.

> db.marks.drop();
true