Which of the following code blocks writes DataFrame itemsDf to disk at storage location filePath, making sure to substitute any existing data at that…

itemsDf.write.mode("overwrite").parquet(filePath)

Correct! itemsDf.write returns a pyspark.sql.DataFrameWriter instance whose overwriting behavior

can be modified via the mode setting or by passing mode="overwrite" to the parquet() command.

Although the parquet format is not prescribed for solving this question, parquet() is a valid operator

to initiate Spark to write the data to disk.

itemsDf.write.mode("overwrite").path(filePath)

No. A pyspark.sql.DataFrameWriter instance does not have a path() method.

itemsDf.write.option("parquet").mode("overwrite").path(filePath)

Incorrect, see above. In addition, a file format cannot be passed via the option() method.

itemsDf.write(filePath, mode="overwrite")

Wrong. Unfortunately, this is too simple. You need to obtain access to a DataFrameWriter for the

DataFrame through calling itemsDf.write upon which you can apply further methods to control how

Spark data should be written to disk. You cannot, however, pass arguments to itemsDf.write directly.

itemsDf.write().parquet(filePath, mode="overwrite")

False. See above.

More info: pyspark.sql.DataFrameWriter.parquet — PySpark 3.1.2 documentation

Static notebook | Dynamic notebook: See test 3,