Factory functions

Use these to create pre-configured connections before passing them to the class constructors.

SQLAlchemy engine

eurocontrolpy.build_sqlalchemy_oracle_engine(schema='PRU_DEV')

Build a SQLAlchemy Engine for Oracle using oracledb (thin mode).

Reads credentials from environment variables:

{schema}_USR, {schema}_PWD, {schema}_DBNAME

Parameters:

schema (str, default 'PRU_DEV') – Prefix for the environment variable names.

Return type:

sqlalchemy.engine.Engine

Spark session

eurocontrolpy.build_spark_oracle_session(executor_memory='5g', driver_memory='2g', executor_cores='1', executor_instances='10', shuffle_partitions='100', default_parallelism='100', max_records_per_batch='10000', jar_path='jars/ojdbc8.jar', oracle_fetch_size=1000)

Create (or reuse) a SparkSession configured for Oracle JDBC.

Returns:

The Spark session, JDBC URL, and JDBC connection properties.

Return type:

tuple[SparkSession, str, dict]

Parameters:

• executor_memory (str)

• driver_memory (str)

• executor_cores (str)

• executor_instances (str)

• shuffle_partitions (str)

• default_parallelism (str)

• max_records_per_batch (str)

• jar_path (str)

• oracle_fetch_size (int)

Notes

Requires the environment variables PRU_DEV_USR (username), PRU_DEV_PWD (password), and PRU_DEV_DBNAME (host:port/service).

Download ojdbc8.jar from the Oracle JDBC downloads page and place it at jar_path.