CODEWITHSUNDEEP
amazon-dynamodb
Sridhar Sarnobat
Sridhar Sarnobat

Reputation: 25296

DynamoDB - scan one-by-one (like a JDBC cursor)

I'd like to get ANY row in my DynamoDB table, but the table could be very big. I don't want to have the DB server return all rows to me and then I just get the first item in the set.

With JDBC I'd use a cursor which is just a "pointer" to row in the database. In DynamoDB I've been looking at the scan() operation (https://docs.aws.amazon.com/amazondynamodb/latest/APIReference/API_Scan.html) but there are so many similar methods, and they vary between versions, and parameters that don't do what I was expecting that I'm really confused. I really don't want to submit JSON queries if there is a simple programmatic statement to do it.

What method should I be looking for to just get items one-by-one from client to server, rather than the server batching them up and returning multiple?

My attempt:

new DynamoDB( AmazonDynamoDBClientBuilder.standard() .withRegion(Regions.US_EAST_1.name()) .build()) .getTable("my_large_table") // MaxResultSize just blindly returns rows, not matching rows. Since we have no filter criteria this is fine. .scan(new ScanSpec().withMaxResultSize(1).withConsistentRead(true)) .firstPage() .getLowLevelResult() .getScanResult() .getItems();

Version info

For now I'm using these but if it's changed in later versions I'd like to know the solution that minimizes future migration pain:

<dependency> <groupId>com.amazonaws</groupId> <artifactId>aws-java-sdk-core</artifactId> <version>1.11.127</version> </dependency> <dependency> <groupId>com.amazonaws</groupId> <artifactId>aws-java-sdk</artifactId> <version>1.11.127</version> </dependency> <dependency> <groupId>com.amazonaws</groupId> <artifactId>aws-java-sdk-dynamodb</artifactId> <version>1.11.127</version> </dependency>

Upvotes: 1

Views: 392

Answers (1)

F_SO_K
F_SO_K

Reputation: 14859

Just use the Limit Scan argument.

By default Scan will evaluate every item in a table. Using the limit operator will set "the maximum number of items to evaluate (not necessarily the number of matching items)".

So if you Scan a table with no filters, and Limit set to 1, you will always get one item returned (assuming there is at least one item).

Upvotes: 1

Related Questions