@@ -29,6 +29,8 @@ class SparkSourceFormat(Enum):
2929
3030
3131class SparkSource (DataSource ):
32+ """A SparkSource object defines a data source that a Spark offline store can use"""
33+
3234 def __init__ (
3335 self ,
3436 * ,
@@ -44,6 +46,25 @@ def __init__(
4446 owner : Optional [str ] = "" ,
4547 timestamp_field : Optional [str ] = None ,
4648 ):
49+ """Creates a SparkSource object.
50+
51+ Args:
52+ name: The name of the data source, which should be unique within a project.
53+ table: The name of a Spark table.
54+ query: The query to be executed in Spark.
55+ path: The path to file data.
56+ file_format: The format of the file data.
57+ created_timestamp_column: Timestamp column indicating when the row
58+ was created, used for deduplicating rows.
59+ field_mapping: A dictionary mapping of column names in this data
60+ source to feature names in a feature table or view.
61+ description: A human-readable description.
62+ tags: A dictionary of key-value pairs to store arbitrary metadata.
63+ owner: The owner of the DataSource, typically the email of the primary
64+ maintainer.
65+ timestamp_field: Event timestamp field used for point-in-time joins of
66+ feature values.
67+ """
4768 # If no name, use the table as the default name.
4869 if name is None and table is None :
4970 raise DataSourceNoNameException ()
0 commit comments