Other

Collections

• array(*cols)
• array_contains(col, value)
• size(col)
• sort_array(col, asc=True)
• struct(*cols)

Sorting

• asc(col)
• desc(col)

Binary

• bitwiseNOT(col)
• shiftLeft(col, numBits)
• shiftRight(col, numBits)
• shiftRightUnsigned(col, numBits)

Dealing with null values

• coalesce(*cols)
• isnan(col)
• isnull(col)

Columns

• col(col) or column(col)
• create_map(*cols)
• explode(col)
• expr(str)
• hash(*cols)
• input_file_name()
• posexplode(col)
• sha1(col)
• sha2(col, numBits)
• soundex(col)
• spark_partition_id()

JSON

• from_json(col, schema, options={})
• get_json_object(col, path)
• json_tuple(col, *fields)
• to_json(col, options={})

Checkpoints

• checkpoint(eager=True)
  • You can set a custom checkpoint directory by using the setCheckpointDir(dir) function on the Spark context, which is accessible through ctx.spark_session.sparkContext. Make sure to include ctx as an input parameter to the compute() function of your transform.
  • Keep in mind that you only need to set the checkpoint directory once. Any subsequent attempt at setting the checkpoint to the same directory will result in RDD errors.
• localCheckpoint(eager=True)