gravitino.client.metadata_object_impl.MetadataObjectImpl¶
- class gravitino.client.metadata_object_impl.MetadataObjectImpl(names: List[str], metadata_object_type: Type)¶
Bases:
MetadataObjectMethods
__init__(names, metadata_object_type)The full name of the object.
name()The name of the object.
parent()The parent full name of the object.
type()The type of the object.
- class Type(value)¶
Bases:
EnumThe type of object in the Gravitino system. Every type will map one kind of the entity of the underlying system.
- CATALOG = 'catalog'¶
A catalog is a collection of metadata from a specific metadata source, like Apache Hive catalog, Apache Iceberg catalog, JDBC catalog, etc.
- COLUMN = 'column'¶
A column is a sub-collection of the table that represents a group of same type data.
- FILESET = 'fileset'¶
A fileset is mapped to a directory on a file system like HDFS, S3, ADLS, GCS, etc.
- METALAKE = 'metalake'¶
A metalake is a concept of tenant. It means an organization. A metalake contains many data sources.
- MODEL = 'model'¶
A model is mapped to the model artifact in ML.
- ROLE = 'role'¶
A role is an object contains specific securable objects with privileges.
- SCHEMA = 'schema'¶
“A schema is a sub collection of the catalog. The schema can contain filesets, tables, topics, etc.
- TABLE = 'table'¶
A table is mapped the table of relational data sources like Apache Hive, MySQL, etc.
- TOPIC = 'topic'¶
A topic is mapped the topic of messaging data sources like Apache Kafka, Apache Pulsar, etc.
- full_name() str¶
The full name of the object.
Full name will be separated by “.” to represent a string identifier of the object, like catalog, catalog.table, etc.
- Returns:
str: The name of the object.
- name() str¶
The name of the object.
- Returns:
str: The name of the object.
- abstract parent() str | None¶
The parent full name of the object.
If the object doesn’t have parent, this method will return None.
- Returns:
Optional[str]: The parent full name of the object.