How can i save a parquet file using pandas with a data factory orchestrated notebook?

Data Engineering

Join discussions on data engineering best practices, architectures, and optimization strategies within the Databricks Community. Exchange insights and solutions with fellow data engineers.

Hi guys,

this is my first question, feel free to correct me if i'm doing something wrong.

Anyway, i'm facing a really strange problem, i have a notebook in which i'm performing some pandas analysis, after that i save the resulting dataframe in a parquet file, and everything is working fine.

But the music change when i run the notebook using azure data factory.

Basically, the saved file is empty, 0 byte contained in it.

This is the function that i'm using to save the file:

def save_file(obj, filename, kind="csv"):
    # Backing up
    outdir = '/dbfs/FileStore/'
    outname = outdir+filename+'.'+kind
    if kind=="csv":
        obj.to_csv(outname, index=False, encoding="utf-8")
    elif kind=="xlsx":
        obj.to_excel(outname, index=False)  # encoding="utf-8"
    elif kind=="parquet":
        obj.to_parquet(outname)
    elif kind=="pkl":
        pass  # Not Implemented
    print(filename, " saved")
    return

One more note that may be useful;

even the "normal execution", without azure data factory, saves the file empty, but if i display the pandas data frame before the saving, it work correctly.

This is weird.

Thank you 🙂