Customer Data Platform: pass schema as option to PTransform

[iht]

Schema is a local file that is not part of the Python package, so this is going to fail in Dataflow probably:

dataflow-solution-guides/pipelines/cdp/cdp_pipeline/customer_data_platform.py

Lines 25 to 27 in 9fb0d38

	# Get the schema of bigquery output table
	with open('./schema/unified_table.json') as schema_file:
	output_schema = json.load(schema_file)

The schema should be read when the pipeline is launched and passed as a string to the PTransform.