[Bug]: Python SchemaRegistry is not properly populated when calling `add`

[mlhamel]

What happened?

In that #28169

The following change was introduced:

  def add(self, typing, schema):
    if not schema.id:
      self.by_id[schema.id] = (typing, schema)

I know the SCHEMA_REGISTRY is an internal service but we were using it in our tests and stopped working with the release of beam 2.51.

Issue Priority

Priority: 3 (minor)

Issue Components

• Component: Python SDK
• Component: Java SDK
• Component: Go SDK
• Component: Typescript SDK
• Component: IO connector
• Component: Beam YAML
• Component: Beam examples
• Component: Beam playground
• Component: Beam katas
• Component: Website
• Component: Spark Runner
• Component: Flink Runner
• Component: Samza Runner
• Component: Twister2 Runner
• Component: Hazelcast Jet Runner
• Component: Google Cloud Dataflow Runner

[damccorm]

Hey @mlhamel as mentioned in

beam/sdks/python/apache_beam/typehints/schema_registry.py

Line 18 in 0a813b9

"""This module is intended for internal use only. Nothing defined here provides

that code is intended for internal use only and does not have backwards compatibility guarantees. In general, you should not explicitly depend on its behavior unless you're ok taking on the risk of it changing