Function serialize_entity_key

sdk/python/feast/infra/key_encoding_utils.py:123–190

Serialize entity key to a bytestring so it can be used as a lookup key in a hash table. We need this encoding to be stable; therefore we cannot just use protobuf serialization here since it does not guarantee that two proto messages containing the same data will serialize to the sa

(
    entity_key: EntityKeyProto, entity_key_serialization_version=3
)

def serialize_entity_key(
    entity_key: EntityKeyProto, entity_key_serialization_version=3
) -> bytes:
    """
    Serialize entity key to a bytestring so it can be used as a lookup key in a hash table.

    We need this encoding to be stable; therefore we cannot just use protobuf serialization
    here since it does not guarantee that two proto messages containing the same data will
    serialize to the same byte string[1].

    [1] https://developers.google.com/protocol-buffers/docs/encoding

    Args:
        entity_key_serialization_version: version of the entity key serialization
        Versions:
        version 3: entity_key size is added to the serialization for deserialization purposes
        entity_key: EntityKeyProto

    Returns: bytes of the serialized entity key
    """
    if entity_key_serialization_version < 3:
        # Not raising the error, keeping it in warning state for reserialization purpose
        # We should remove this after few releases
        warnings.warn(
            "Serialization of entity key with version < 3 is removed. Please use version 3 by setting entity_key_serialization_version=3."
            "To reserializa your online store featrues refer -  https://github.com/feast-dev/feast/blob/master/docs/how-to-guides/entity-reserialization-of-from-v2-to-v3.md"
        )

    sorted_keys: List[str]
    sorted_values: List[ValueProto]
    if not entity_key.join_keys:
        sorted_keys = []
        sorted_values = []
    elif len(entity_key.join_keys) == 1:
        # Fast path: single entity, no sorting needed
        sorted_keys = [entity_key.join_keys[0]]
        sorted_values = [entity_key.entity_values[0]]
    else:
        # Multi-entity: use sorting
        pairs = sorted(zip(entity_key.join_keys, entity_key.entity_values))
        sorted_keys = [k for k, _ in pairs]
        sorted_values = [v for _, v in pairs]

    output: List[bytes] = []

    if entity_key_serialization_version > 2:
        output.append(struct.pack("<I", len(sorted_keys)))

    # Optimize key encoding by pre-encoding all strings
    if sorted_keys:
        encoded_keys = [k.encode("utf8") for k in sorted_keys]
        for i, k_encoded in enumerate(encoded_keys):
            output.append(struct.pack("<I", ValueType.STRING))
            if entity_key_serialization_version > 2:
                output.append(struct.pack("<I", len(k_encoded)))
            output.append(k_encoded)

    for v in sorted_values: