Quick Tour

The following code snippets come from the QuickTour.scala example code that can be found with the driver source.


See the installation guide for instructions on how to install the MongoDB Scala Driver.

This guide uses the Helper implicits as covered in the Quick Tour Primer.

Make a Connection

The following example shows multiple ways to connect to the database mydb on the local machine, using the MongoClient companion object.

import org.mongodb.scala._

// To directly connect to the default server localhost on port 27017
val mongoClient: MongoClient = MongoClient()

// Use a Connection String
val mongoClient: MongoClient = MongoClient("mongodb://localhost")

// or provide custom MongoClientSettings
val settings: MongoClientSettings = MongoClientSettings.builder()
    .applyToClusterSettings(b => b.hosts(List(new ServerAddress("localhost")).asJava).
val mongoClient: MongoClient = MongoClient(settings)

val database: MongoDatabase = mongoClient.getDatabase("mydb")

At this point, the database object will be a connection to a MongoDB server for the specified database.


The API only returns implementations of Observable[T] when network IO required for the operation. For getDatabase("mydb") there is no network IO required. A MongoDatabase instance provides methods to interact with a database but the database might not actually exist and will only be created on the insertion of data via some means; e.g. the creation of a collection or the insertion of documents.


The MongoClient instance actually represents a pool of connections for a given MongoDB server deployment; you will only need one instance of class MongoClient even with multiple concurrently executing asynchronous operations.


Typically you only create one MongoClient instance for a given database cluster and use it across your application. When creating multiple instances:

  • All resource usage limits (max connections, etc) apply per MongoClient instance
  • To dispose of an instance, make sure you call MongoClient.close() to clean up resources

Get a Collection

To get a collection to operate upon, specify the name of the collection to the getCollection(String collectionName) method:

The following example gets the collection test:

val collection: MongoCollection[Document] = database.getCollection("test");

Insert a Document

Once you have the collection object, you can insert documents into the collection. For example, consider the following JSON document; the document contains a field info which is an embedded document:

   "name" : "MongoDB",
   "type" : "database",
   "count" : 1,
   "info" : {
               x : 203,
               y : 102

To create the document using the Scala driver, use the Document class. You can use this class to create the embedded document as well.


The Scala driver provides two document types - an immutable Document and a mutable Document.

When using an immutable document then you should explicitly add an _id value, if you need to know that _id value in the future.

If an _id is not present on insertion then driver will add one automatically create one and pass it to the server, but that _id will not be passed back to the user.

val doc: Document = Document("_id" -> 0, "name" -> "MongoDB", "type" -> "database",
                             "count" -> 1, "info" -> Document("x" -> 203, "y" -> 102))

To insert the document into the collection, use the insertOne() method. Using the results() implicit we block until the observer is completed:



In the API all methods returning a Observables are “cold” streams meaning that nothing happens until they are Subscribed to.

The example below does nothing:

val observable: Observable[Completed] = collection.insertOne(doc)

Only when an Observable is subscribed to and data requested will the operation happen:

// Explictly subscribe:
observable.subscribe(new Observer[Completed] {

  override def onNext(result: Completed): Unit = println("Inserted")

  override def onError(e: Throwable): Unit = println("Failed")

  override def onComplete(): Unit = println("Completed")

Once the document has been inserted the onNext method will be called and it will print “Inserted!” followed by the onCompleted method which will print “Completed”. If there was an error for any reason the onError method would print “Failed”.

Add Multiple Documents

To add multiple documents, you can use the insertMany() method.

The following example will add multiple documents of the form:

{ "i" : value }

Create the documents in a loop.

val documents = (1 to 100) map { i: Int => Document("i" -> i) }

To insert these documents to the collection, pass the list of documents to the insertMany() method.

val insertObservable = collection.insertMany(documents)

As we haven’t subscribed yet no documents have been inserted, lets chain together two operations, inserting and counting.

Count Documents in A Collection

Once we’ve inserted the documents list we should, have a total of 101 documents in the collection (the 100 we did in the loop, plus the first one). We can check to see if we have them all using the count() method.

Lets chain the two operations together using a for comprehension. The following code should insert the documents then count the number of documents and print the results:

val insertAndCount = for {
  insertResult <- insertObservable
  countResult <- collection.countDocuments()
} yield countResult

println(s"total # of documents after inserting 100 small ones (should be 101):  ${insertAndCount.headResult()}")

Query the Collection

Use the find() method to query the collection.

Find the First Document in a Collection

To get the first document in the collection, call the first() method on the FindObservable operation. collection.find().first() returns the first document or if no document is found the Observable just completes. This is useful for queries that should only match a single document, or if you are interested in the first document only.

Using the printHeadResult() implicit we block until the observer is completed and then print the first result:


The example will print the following document:

{ "_id" : 0, "name" : "MongoDB", "type" : "database", "count" : 1, "info" : { "x" : 203, "y" : 102 } }


The _id of the document is “0” as we explicitly set one before we inserted this document. MongoDB reserves field names that start with “_” and “$” for internal use.

Find All Documents in a Collection

To retrieve all the documents in the collection, we will use the find() method. The find() method returns a FindObservable instance that provides a fluent interface for chaining or controlling find operations. The following code retrieves all documents in the collection and prints them out (101 documents). Using the printResults() implicit we block until the observer is completed and then print each result:


Get A Single Document with a Query Filter

We can create a filter to pass to the find() method to get a subset of the documents in our collection. For example, if we wanted to find the document for which the value of the “i” field is 71, we would do the following:

import org.mongodb.scala.model.Filters._

collection.find(equal("i", 71)).first().printHeadResult()

will eventually print just one document:

{ "_id" : { "$oid" : "5515836e58c7b4fbc756320b" }, "i" : 71 }


Use the Filters, Sorts, Projections and Updates helpers for simple and concise ways of building up queries.

Get a Set of Documents with a Query

We can use the query to get a set of documents from our collection. For example, if we wanted to get all documents where "i" > 50, we could write:

collection.find(gt("i", 50)).printResults()

which should print the documents where i > 50.

We could also get a range, say 50 < i <= 100:

collection.find(and(gt("i", 50), lte("i", 100))).printResults()

Sorting documents

We can also use the Sorts helpers to sort documents. We add a sort to a find query by calling the sort() method on a FindObservable. Below we use the exists() helper and use the sort descending("i") helper to sort our documents:

import org.mongodb.scala.model.Sorts._


Projecting fields

Sometimes we don’t need all the data contained in a document. The Projections helpers can be used to build the projection parameter for the find operation and limit the fields returned.
Below we’ll sort the collection, exclude the _id field and output the first matching document:

import org.mongodb.scala.model.Projections._



Sometimes we need to aggregate the data stored in MongoDB. The Aggregates helper provides builders for each of type of aggregation stage.

Below we’ll do a simple two step transformation that will calculate the value of i * 10. First we find all Documents where i > 0 by using the Aggregates.filter helper. Then we reshape the document by using Aggregates.project in conjunction with the $multiply operator to calculate the “ITimes10” value:

import org.mongodb.scala.model.Aggregates._

collection.aggregate(Seq(filter(gt("i", 0)),
  project(Document("""{ITimes10: {$multiply: ["$i", 10]}}""")))

For $group operations use the Accumulators helper for any accumulator operations. Below we sum up all the values of i by using the Aggregates.group helper in conjunction with the Accumulators.sum helper:

collection.aggregate(List(group(null, sum("total", "$i")))).printHeadResult()


Currently, there are no helpers for aggregation expressions. Use the Document.parse() helper to quickly build aggregation expressions from extended JSON.

Updating documents

There are numerous update operators supported by MongoDB. We can use the Updates helpers to help update documents in the database.

To update at most a single document (may be 0 if none match the filter), use the updateOne method to specify the filter and the update document. Here we update the first document that meets the filter i equals 10 and set the value of i to 110:

import org.mongodb.scala.model.Updates._

collection.updateOne(equal("i", 10), set("i", 110)).printHeadResult("Update Result: ")

To update all documents matching the filter use the updateMany method. Here we increment the value of i by 100 where i is less than 100.

collection.updateMany(lt("i", 100), inc("i", 100)).printHeadResult("Update Result: ")

The update methods return an UpdateResult, which provides information about the operation including the number of documents modified by the update.

Deleting documents

To delete at most a single document (may be 0 if none match the filter) use the deleteOne method:

collection.deleteOne(equal("i", 110)).printHeadResult("Delete Result: ")

To delete all documents matching the filter use the deleteMany method. Here we delete all documents where i is greater or equal to 100:

collection.deleteMany(gte("i", 100)).printHeadResult("Delete Result: ")

The delete methods return a DeleteResult, which provides information about the operation including the number of documents deleted.

Bulk operations

These commands allow for the execution of bulk insert/update/delete operations. There are two types of bulk operations:

  1. Ordered bulk operations.

    Executes all the operation in order and error out on the first write error.

  2. Unordered bulk operations.

    Executes all the operations and reports any the errors.

    Unordered bulk operations do not guarantee order of execution.

Let’s look at two simple examples using ordered and unordered operations:

val writes: List[WriteModel[_ <: Document]] = List(
      InsertOneModel(Document("_id" -> 4)),
      InsertOneModel(Document("_id"-> 5)),
      InsertOneModel(Document("_id" -> 6)),
      UpdateOneModel(Document("_id" -> 1), set("x", 2)),
      DeleteOneModel(Document("_id" -> 2)),
      ReplaceOneModel(Document("_id" -> 3), Document("_id" -> 3, "x" -> 4))

// 1. Ordered bulk operation - order is guaranteed
collection.bulkWrite(writes).printHeadResult("Bulk write results: ")

// 2. Unordered bulk operation - no guarantee of order of operation
collection.bulkWrite(writes, BulkWriteOptions().ordered(false)).printHeadResult("Bulk write results (unordered): ")


Use of the bulkWrite methods is not recommended when connected to pre-2.6 MongoDB servers, as this was the first server version to support bulk write commands for insert, update, and delete in a way that allows the driver to implement the correct semantics for BulkWriteResult and BulkWriteException. The methods will still work for pre-2.6 servers, but performance will suffer, as each write operation has to be executed one at a time.