Docs Menu
Docs Home
/ / /
Kotlin Coroutine

Modify Documents

On this page

  • Overview
  • Update
  • Update Operation Parameters
  • Replace
  • Replace Operation Parameters

In this guide, you can learn how to modify documents in a MongoDB collection using two distinct operation types:

  • Update

  • Replace

Update operations specify the fields and values to change in one or more documents. A replace operation specifies the fields and values to replace a single document from your collection.

In the following examples, a paint store sells five different colors of paint. The paint_inventory collection represents their current inventory:

{ "_id": 1, "color": "red", "qty": 5 }
{ "_id": 2, "color": "purple", "qty": 8 }
{ "_id": 3, "color": "yellow", "qty": 0 }
{ "_id": 4, "color": "green", "qty": 6 }
{ "_id": 5, "color": "pink", "qty": 0 }

This data is modeled with the following Kotlin data class:

data class PaintOrder(
@BsonId val id: Int,
val color: String,
val qty: Int
)

Update operations can modify fields and values. They apply changes specified in an update document to one or more documents that match your query filter.

The updateOne() method changes the first document your query filter matches and the updateMany() method changes all the documents your query filter matches.

You can call the updateOne() and updateMany() methods on a MongoCollection instance as follows:

collection.updateOne(query, updateDocument)
collection.updateMany(query, updateDocument)

The updateOne() and updateMany() methods both have the following parameters:

  • query specifies a query filter with the criteria to match documents to update in your collection

  • updateDocument specifies the fields and values to modify in the matching document or documents. For this example, we use the Updates builder to create the update document.

You can create the updateDocument using an Updates builder as follows:

val updateDocument = Updates.operator(field, value)

See the MongoDB API documentation for a complete list of Updates builders and their usage.

The paint store needs to update their inventory after a customer returns a can of yellow paint.

To update the single can of paint, call the updateOne() method specifying the following:

  • A query filter that matches the yellow color

  • An update document that contains instructions to increment the qty field by "1"

val filter = Filters.eq(PaintOrder::color.name, "yellow")
val update = Updates.inc(PaintOrder::qty.name, 1)
val result = collection.updateOne(filter, update)
println("Matched document count: $result.matchedCount")
println("Modified document count: $result.modifiedCount")
Matched document count: 1
Modified document count: 1

The paint store then receives a fresh shipment and needs to update their inventory again. The shipment contains 20 cans of each paint color.

To update the inventory, call the updateMany() method specifying the following:

  • A query filter that matches all the colors

  • An update document that contains instructions to increment the qty field by "20"

val filter = Filters.empty()
val update = Updates.inc(PaintOrder::qty.name, 20)
val result = collection.updateMany(filter, update)
println("Matched document count: $result.matchedCount")
println("Modified document count: $result.modifiedCount")
Matched document count: 5
Modified document count: 5

The following shows the updated documents in the paint_inventory collection:

{ "_id": 1, "color": "red", "qty": 25 }
{ "_id": 2, "color": "purple", "qty": 28 }
{ "_id": 3, "color": "yellow", "qty": 20 }
{ "_id": 4, "color": "green", "qty": 26 }
{ "_id": 5, "color": "pink", "qty": 20 }

If zero documents match the query filter in the update operation, updateMany() makes no changes to documents in the collection. See our upsert guide to learn how to insert a new document instead of updating one if no documents match.

Important

The updateOne() and updateMany() methods cannot make changes to a document that violate unique index constraints on the collection. See the MongoDB server manual for more information on unique indexes.

A replace operation substitutes one document from your collection. The substitution occurs between a document your query filter matches and a replacement document.

The replaceOne() method removes all the existing fields and values in the matching document (except the _id field) and substitutes it with the replacement document.

You can call the replaceOne() method on a MongoCollection instance as follows:

collection.replaceOne(query, replacementDocument)

The replaceOne() method has the following parameters:

  • query specifies a query filter with the criteria to match a document to replace in your collection

  • replacementDocument specifies fields and values of a new Document object to replace in the matched document

The paint store realizes they need to update their inventory again. What they thought was 20 cans of pink paint is actually 25 cans of orange paint.

To update the inventory, call the replaceOne() method specifying the following:

  • A query filter that matches documents where the color is "pink"

  • A replacement document where the color is "orange" and the qty is "25"

val filter = Filters.eq(PaintOrder::color.name, "pink")
val update = PaintOrder(5, "orange", 25)
val result = collection.replaceOne(filter, update)
println("Matched document count: $result.matchedCount")
println("Modified document count: $result.modifiedCount")
Matched document count: 1
Modified document count: 1

The following shows the updated document:

{ "_id": 5, "color": "orange", "qty": 25 }

If zero documents match the query filter in the replace operation, replaceOne() makes no changes to documents in the collection. See our upsert guide to learn how to insert a new document instead of replacing one if no documents match.

If multiple documents match the query filter specified in the replaceOne() method, it replaces the first result.

Important

The replaceOne() method cannot make changes to a document that violate unique index constraints on the collection. See the MongoDB server manual for more information on unique indexes.

Next

MongoDB Kotlin Driver