Closed sabbyanandan closed 5 years ago
Hi, @AddaxT: The scriptable-transform
processor is limited in capabilities; besides that, you won't be able to unit/IT test any of your business logic.
There are few options for Python.
1) Use any of the available python
processors
2) Use the gRPC processor, so you can negotiate with a side-car running native Python workload via gRPC.
3) Alternatively, since you're deploying onto Kubernetes, you could build a python-native application as a Docker image, and register that in SCDF to build a streaming data pipeline with it. We have experimented that in SCDF via spring-cloud/spring-cloud-dataflow#3110. A preliminary version of the docs/samples on this PR commit here. We are actively iterating on it, so please expect changes and be prepared for ongoing updates to the PR.
Closing this due to no activity.
From @AddaxT: