[Bug]: Beam YAML 'Create' transform crashes on mixed primitive and dictionary elements due to schema inconsistency

[atheendre130505]

What happened?

What happened? Description: The Create transform in Beam YAML crashes with an AttributeError when the input elements list contains a mix of primitive types (int, str) and dictionary mappings.

YamlProviders.create wraps primitives in a Row with a field named element, but it converts dictionaries into Row objects using the dict keys. Because Beam's
Create transform infers the PCollection schema from the very first element, any subsequent element with a different structure causing a serialization crash because the coder cannot find the expected attributes.

Beam Version: 2.61.0 (Python SDK)

python

from apache_beam.yaml.yaml_provider import YamlProviders
import apache_beam as beam
with beam.Pipeline() as p:
    # A mix of a primitive (Row(element=1)) and a dict (Row(a=2))
    p | YamlProviders.create([1, {"a": 2}])

Resulting Error: AttributeError: 'Row' object has no attribute 'element'

The pipeline fails immediately upon start-up or during construction, rather than unifying the elements into a consistent schema or providing a diagnostic error message.

Issue Priority

Priority: 2 (default / most bugs should be filed as P2)

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: Infrastructure
• Component: Spark Runner
• Component: Flink Runner
• Component: Samza Runner
• Component: Twister2 Runner
• Component: Hazelcast Jet Runner
• Component: Google Cloud Dataflow Runner

[dheeraj12347]

Hi @atheendre130505,
I'd like to take this issue! I've worked with Python SDKs and data serialization in other Apache projects (like Fory), so I'm familiar with the challenges of schema inference and Row object construction.

The description provides a very clear reproduction case. I'll start by investigating apache_beam/yaml/yaml_provider.py to see how we can better unify the schema when YamlProviders.create encounters mixed types, perhaps by enforcing a consistent wrapper or improving the diagnostic error.

Please assign this to me, and I'll work on a fix.