pyspark.sql.VariantVal#

class pyspark.sql.VariantVal(value, metadata)[source]#

A class to represent a Variant value in Python.

Added in version 4.0.0.

Parameters:

valuebytes

The bytes representing the value component of the Variant.

metadatabytes

The bytes representing the metadata component of the Variant.

Methods

toPython()

Convert the VariantVal to a Python data structure.

Examples

>>> from pyspark.sql import functions as sf
>>> df = spark.createDataFrame([ {'json': '''{ "a" : 1 }'''} ])
>>> v = df.select(sf.parse_json(df.json).alias("var")).head().var
>>> v.toPython()
{'a': 1}
>>> v.toJson()
'{"a":1}'

Methods

parseJson(json_str)	Parse a JSON string and construct a VariantVal.
toJson([zone_id])	Convert the VariantVal to a JSON string.
toPython()	Convert the VariantVal to a Python data structure.