Azure Cosmos DB Python SDK limitations

Question

Python SDK for Azure Cosmos DB SQL Api has limitations on support for GROUP BY and aggregate queries.

Posting this question to understand -

1. If there is a plan to fix these limitations, if so, any ETA/release timeline?
2. Meanwhile, are there any recommended workarounds specifically for the above mentioned cases

Answer 1

Hello,

ETA is September : https://github.com/Azure/azure-sdk-for-python/issues/25868#issuecomment-1522394000

Some years ago, workarounds were mentioned in this topic:
https://stackoverflow.com/questions/45270125/alternative-to-group-by-for-cosmos-db

A package called 'documentdb-lumenize' is available for .NET and Node.js, which supports Aggregations (Group-by, Pivot-table, and N-dimensional Cube) and Time Series Transformations as Stored Procedures in DocumentDB.
The examples for calling the stored procedure are only given for Node.js and .NET but the procedure can be called from any client-side library, including Python.
https://github.com/lmaccherone/documentdb-lumenize

Additionally, for Python and Scala, you could refer to azure-cosmosdb-spark.