Docs Menu
Docs Home
/
MongoDB Shell

Write Scripts

On this page

  • Compatibility
  • Execute a JavaScript File
  • Execute a Script from Within mongosh
  • Execute a Script From the Command Line
  • Execute a Script From the Command Line with Authentication
  • Terminate a Script on Error
  • Execute Code From a Configuration File
  • Execute JavaScript Code
  • Execute MongoDB Code
  • Execute JavaScript and MongoDB Code
  • Open a New Connection
  • Connect to a Local MongoDB Instance
  • Connect to an Atlas Deployment
  • Connect to a MongoDB Instance that Enforces Access Control
  • Use connect() to Connect to a MongoDB Instance
  • Connection Considerations

You can write scripts for the MongoDB Shell that modify data in MongoDB or perform administrative operations. You may also want to package your scripts as snippets for easier distribution and management.

This tutorial introduces using the MongoDB Shell with JavaScript to access MongoDB.

You can write scripts for the MongoDB Shell for deployments hosted in the following environments:

  • MongoDB Atlas: The fully managed service for MongoDB deployments in the cloud

  • MongoDB Enterprise: The subscription-based, self-managed version of MongoDB

  • MongoDB Community: The source-available, free-to-use, and self-managed version of MongoDB

To learn more about using the MongoDB Shell for deployments hosted in MongoDB Atlas, see Connect via mongosh.

You can execute a .js file from within the MongoDB Shell using the load() method.

The load() method accepts relative and absolute paths. If the current working directory of the MongoDB Shell is /data/db, and connect-and-insert.js is in the /data/db/scripts directory, then the following calls within the MongoDB Shell are equivalent:

load( "scripts/connect-and-insert.js" )
load( "/data/db/scripts/connect-and-insert.js" )

The following example creates and executes a script that:

  • Connects to a local instance running on the default port.

  • Connects to the myDatabase database.

  • Populates the movies collection with sample documents.

  1. Create a file named connect-and-insert.js with the following contents:

    db = connect( 'mongodb://localhost/myDatabase' );
    db.movies.insertMany( [
    {
    title: 'Titanic',
    year: 1997,
    genres: [ 'Drama', 'Romance' ]
    },
    {
    title: 'Spirited Away',
    year: 2001,
    genres: [ 'Animation', 'Adventure', 'Family' ]
    },
    {
    title: 'Casablanca',
    genres: [ 'Drama', 'Romance', 'War' ]
    }
    ] )
  2. To load and execute the connect-and-insert.js file, use mongosh to connect to your deployment and run the following command:

    load( "connect-and-insert.js" )
  3. To confirm that the documents loaded correctly, use the myDatabase collection and query the movies collection.

    use myDatabase
    db.movies.find()

Note

There is no search path for the load() method. If the target script is not in the current working directory or the full specified path, the MongoDB Shell cannot access the file.

You can use mongosh to execute a script from the command line without entering the mongosh console.

To specify the filename, use the --file or -f parameter to specify the filename. You may also need to specify connection information in addition to the --file or -f parameters.

Tip

If you pass a filename to mongosh without using the parameter flags the connection may fail if there are other command line arguments.

To pass filenames always use --file or -f.

The following example creates scripts and runs them from the command line.

  1. Copy this script and save it as loadMovies.js.

    db = connect( 'mongodb://localhost/films' );
    db.movies.insertMany( [
    {
    title: 'Titanic',
    year: 1997,
    genres: [ 'Drama', 'Romance' ]
    },
    {
    title: 'Spirited Away',
    year: 2001,
    genres: [ 'Animation', 'Adventure', 'Family' ]
    },
    {
    title: 'Casablanca',
    genres: [ 'Drama', 'Romance', 'War' ]
    }
    ] )

    Tip

    Verify the connection string in the highlighted line. If your MongoDB instance is not running on localhost:27017, you must edit the connection string.

    For example, the following connection string connects to localhost port 27500:

    db = connect( 'mongodb://localhost:27500/films' );
  2. Copy this script and save it as queryMovies.js.

    db = connect( 'mongodb://localhost/films' );
    printjson( db.movies.find( {} ) );
  3. Run the scripts from the command line.

    mongosh --file loadMovies.js --file queryMovies.js
  4. Verify the output.

    Loading file: loadMovies.js
    Loading file: queryMovies.js
    [
    {
    _id: ObjectId("616f1b8092dbee425b661117"),
    title: 'Titanic',
    year: 1997,
    genres: [ 'Drama', 'Romance' ]
    },
    {
    _id: ObjectId("616f1b8092dbee425b661118"),
    title: 'Spirited Away',
    year: 2001,
    genres: [ 'Animation', 'Adventure', 'Family' ]
    },
    {
    _id: ObjectId("616f1b8092dbee425b661119"),
    title: 'Casablanca',
    genres: [ 'Drama', 'Romance', 'War' ]
    }
    ]

    The output of the db.collection.find() command shows that the movies collection was updated.

    Tip

    To make the output visible, use printjson() to call db.collection.find().

    printjson( db.movies.find( {} ) ) ;

To execute a script against a remote mongod instance that requires authentication, specify the connection and authentication details in addition to the filename.

For example:

mongosh --host 172.17.0.3 --port 27500 --username filmFan --password superSecret --file loadMovies.js

You can also specify the shortened form of the options:

mongosh --host 172.17.0.3 --port 27500 -u filmFan -p superSecret -f loadMovies.js

Tip

In shells like bash and zsh, if you begin a command with a space it will not be saved in your command history. This minimizes exposure if you enter passwords on the command line.

It is often useful to terminate a running script if an exception is thrown, or in the case of unexpected results. Exiting at a specific point in the script prevents the execution of unnecessary code and potentially unexpected results.

To terminate a script, you can call the exit(<code>) method, where the <code> is any user-specified value.

As a best practice, wrap code in a try - catch, calling the exit method in the catch block. Likewise, to check the results of a query or any command, you can add an if - else statement and call the exit method if the results are not what is expected.

On startup, mongosh checks your HOME directory for a JavaScript file named .mongoshrc.js. If this file is found, mongosh reads the content of .mongoshrc.js before displaying the prompt for the first time.

Tip

See also:

To update the mongosh prompt to display line numbers, add the following code to <your-home-directory>/.mongoshrc.js

let cmdCount = 1;
prompt = function() {
return (cmdCount++) + "> ";
}

The prompt will look like this:

1> show collections
2> use test
3>

To create a log of when your mongosh client connects to a database, add the following code to <your-home-directory>/.mongoshrc.js:

db.clientConnections.insertOne( { connectTime: ISODate() } )

Each time you connect to a database, the MongoDB server adds a document like the following to the clientConnections collection.

{
_id: ObjectId("61d4bbf0fa4c85f53418070f"),
connectTime: ISODate("2022-01-04T21:28:16.367Z")
}

The current database name is part of the default mongosh prompt. To reformat the prompt to show the database and hostname, use a function like this one:

{
const hostnameSymbol = Symbol('hostname');
prompt = () => {
if (!db[hostnameSymbol])
db[hostnameSymbol] = db.serverStatus().host;
return `${db.getName()}@${db[hostnameSymbol]}> `;
};
}

The prompt will look like this:

admin@centos0722:27502>

From the MongoDB Shell or from a JavaScript file, you can instantiate database connections using the Mongo() method:

new Mongo()
new Mongo(<host>)
new Mongo(<host:port>)

Note

The MongoDB Shell does not support the ClientSideFieldLevelEncryptionOptions document with the Mongo() method.

Consider a MongoDB instance running on localhost on the default port.

The following example:

  • Instantiates a new connection to the instance, and

  • Sets the global db variable to myDatabase using the Mongo.getDB() method.

conn = Mongo();
db = conn.getDB("myDatabase");

To connect to a deployment hosted in MongoDB Atlas, run the mongosh command with your Atlas connection string. For example:

mongosh "mongodb+srv://YOUR_CLUSTER_NAME.YOUR_HASH.mongodb.net/" --apiVersion YOUR_API_VERSION --username YOUR_USERNAME

Once you've established a connection to your deployment, you can instantiate database connections directly from the MongoDB Shell. The following example:

  • Instantiates a connection to the current deployment by using the db.getMongo() method.

  • Sets the global db variable to myDatabase by using the Mongo.getDB() method.

conn = db.getMongo()
db = conn.getDB("myDatabase");

To connect to a MongoDB instance that enforces access control, you must include the credentials in the connection string.

The following command connects to a MongoDB instance that is:

  • Running on localhost on the default port, and

  • Secured using SCRAM.

conn = Mongo("mongodb://<username>:<password>@localhost:27017/<authDB>");

Note

The MongoDB Shell redacts credentials from the command history and the logs.

You can also use the connect() method to connect to the MongoDB instance.

The following command:

  • Connects to the MongoDB instance that is running on localhost with the non-default port 27020, and

  • Sets the global db variable.

db = connect("localhost:27020/myDatabase");

Consider portability and the operating environment when you write scripts.

If the connection details are included in the script:

  • You do not need to specify connection information on the command line.

  • You should use the --nodb parameter.

Consider a mongod instance running on localhost:27500.

The following script prints the number of users. Copy the code and save it as getUserCount.js.

db = connect( "localhost:27500/admin" );˘
printjson( db.system.users.countDocuments() );

Run getUserCount.js:

mongosh --nodb --file getUserCount.js
  • mongosh defaults to port 27170.

  • mongod is running on port 27500.

  • The --nodb parameter instructs mongosh to run a script without first connecting to a mongod instance.

The highlighted line is correct, but getUserCount.js will not run without --nodb because mongosh cannot connect to the local instance. With --nodb, mongosh runs getUserCount.js and uses the highlighted information to connect.

It is convenient to specify connection information in your script, but that also makes it less portable. The getUserCount.js script would have to be updated to run on a remote instance or one running on a different port.

To increase portability, use db.getSiblingDB() and specify the connection information on the command line.

The following script is more portable than getUserCount.js because it does not have specific connection details. Copy the code and save it as portableGetUserCount.js.

db = db.getSiblingDB( "admin" );
printjson( db.system.users.countDocuments() );

To run portableGetUserCount.js, specify the host and port on the command line:

mongosh --host 172.17.0.3 --port 27500 --file portableGetUserCount.js

To run portableGetUserCount.js on a different host or port, change the connection details on the command line. Unlike getUserCount.js, you do not have to edit the script to run portableGetUserCount.js.

Back

Client-Side Field Level Encryption