You need to design a data pipeline that ingests data from CSV, Avro, and Parquet...

The ETL (Extract, Transform, Load) methodology is the best approach for this scenario because it allows you to extract data from the files, transform it by applying the necessary data cleansing (including removing malicious SQL injections), and then load the sanitized data into BigQuery. By transforming the data before loading it into BigQuery, you ensure that only clean and safe data is stored, which is critical for security and data quality.