rowKeySchema property

The row key schema for this table.

The schema is used to decode the raw row key bytes into a structured format. The order of field declarations in this schema is important, as it reflects how the raw row key bytes are structured. Currently, this only affects how the key is read via a GoogleSQL query from the ExecuteQuery API. For a SQL query, the _key column is still read as raw bytes. But queries can reference the key fields by name, which will be decoded from _key using provided type and encoding. Queries that reference key fields will fail if they encounter an invalid row key. For example, if _key = "some_id#2024-04-30#\x00\x13\x00\xf3" with the following schema: { fields { field_name: "id" type { string { encoding: utf8_bytes {} } } } fields { field_name: "date" type { string { encoding: utf8_bytes {} } } } fields { field_name: "product_code" type { int64 { encoding: big_endian_bytes {} } } } encoding { delimited_bytes { delimiter: "#" } } } The decoded key parts would be: id = "some_id", date = "2024-04-30", product_code = 1245427 The query "SELECT _key, product_code FROM table" will return two columns: /------------------------------------------------------\ | _key | product_code | | --------------------------------------|--------------| | "some_id#2024-04-30#\x00\x13\x00\xf3" | 1245427 | ------------------------------------------------------/ The schema has the following invariants: (1) The decoded field values are order-preserved. For read, the field values will be decoded in sorted mode from the raw bytes. (2) Every field in the schema must specify a non-empty name. (3) Every field must specify a type with an associated encoding. The type is limited to scalar types only: Array, Map, Aggregate, and Struct are not allowed. (4) The field names must not collide with existing column family names and reserved keywords "_key" and "_timestamp". The following update operations are allowed for row_key_schema: - Update from an empty schema to a new schema. - Remove the existing schema. This operation requires setting the ignore_warnings flag to true, since it might be a backward incompatible change. Without the flag, the update request will fail with an INVALID_ARGUMENT error. Any other row key schema update operation (e.g. update existing schema columns names or types) is currently unsupported.