[Clarifications] Difference between @PartitionKey (Spring annotation) and partition_key_path (Container parameter)

Question

Hi all,
I don't understand the difference between partition_key_path that I should set during container creation and the Spring annotation @PartitionKey used when I define an entity in my app.

Which is the relationship between the two modes?

In my case, I don't have nested keys, but I need to partition on a base field (ie. firstName), so I set it (ie /firstName) as partition_key_path. Then do I annotate firstName field with @PartitionKey in my code too? or is it unnecessary?

I read in this article (https://learn.microsoft.com/en-us/java/api/overview/azure/spring-data-cosmos-readme?view=azure-java-stable#nested-partition-key-support) that @PartitionKey annotation takes precedence versus partitionKeyPath defined in the @Container. Is it the same case of partition_key_path defined at Container definition context? If yes, in this case I could leave /id as partition_key_path and manage partitioning at applicative level.

Thank you for the support.

Answer 1

Hi @Francesco Cesareo , welcome to Microsoft Q&A forum.

You are right in mentioning that @partitionkey annotation takes precedence when defined. Also the same article mentions that partitionKeyPath should only be used to support nested partition key path. For general partition key support, use the @PartitionKey annotation.

So if you dont have the nested key path, you can choose to use @partitionkey annotation.

Please let us know if this helps or else we can discuss more on this.