Overview
A table is a collection of related data organized in tabular format; consisting of columns and rows.
Object Definition Schema
Schema ID: 3
Field Name | Protobuf Type | Description | Required? | Default |
---|---|---|---|---|
name | string | A user-friendly name of this table | Yes | |
schema | Schema | Schema of the table, see Table Schema | Yes | |
distribution_keys | repeated uint32 | The list of IDs for columns that are used as the distribution key | No | |
sort_keys | repeated uint32 | The list of IDs for columns that are used as sort key | No | |
primary_key | repeated uint32 | The list of IDs for columns that are used as primary key | No | |
unique_columns | repeated uint32 | The list of IDs for columns that are not used as primary key but are unique | No | |
watermark_key | repeated uint32 | The list of IDs for columns that are used as watermark key, see Upsert | No | |
tombstone_key | uint32 | The column ID that is used as the tombstone key, see Upsert | ||
tombstone_string_key_value | string | If tombstone key column is of type char or varchar, the string value that represents the delete marker | ||
table_type | string | Table type, see Table Type | No | MANAGED |
table_format | string | The format of the table, which decides the usage of format_properties . Currently ICEBERG is the only option. |
Yes | |
format_properties | map |
Free form format-specific key-value string properties, e.g. Apache Iceberg | No | |
properties | map |
Free form user-defined key-value string properties | No |