Wrangling Data Flow (WDF) in ADF now supports Parquet format. You can have your data stored in ADLS Gen2 or Azure Blob in parquet format and use that to do agile data preparation using Wrangling Data Flow in ADF
Create a parquet format dataset in ADF and use that as an input in your wrangling data flow
You can then use the parquet format dataset as an input to your Wrangling Data Flow to do agile data preparation at cloud scale via spark execution
Learn more about using Wrangling Data Flow to do data preparation at cloud scale here.