Struct ObjectIdEncoding 

pub struct ObjectIdEncoding { /* private fields */ }

The ObjectIdEncoding represents each distinct term in the database with a single unique id. We call such an id object id. Here is an example of the encoding:

?variable

 ┌─────┐
 │   1 │ ────►  <#MyEntity>
 ├─────┤
 │   2 │ ────►  120^^xsd:integer
 ├─────┤
 │ ... │
 └─────┘

Object ID Mapping

The mapping implementation depends on the storage layer that is being used. For example, an in-memory RDF store will use a different implementation as an on-disk RDF store. The ObjectIdMapping trait defines the contract.

Strengths and Weaknesses

The object id encoding is very well suited for evaluating joins, as instead of joining variable-length RDF terms, we can directly join the object ids. While we do not have recent numbers for the performance gains, the original pull request quadrupled the performance of some queries (with relatively small datasets!).

However, this also introduces the necessity of decoding the object ids back to RDF terms. For example, by converting it to the PlainTermEncoding. For queries that spend little time on join operations, the cost of decoding the object ids can outweigh the benefits of using the object id encoding.

Furthermore, the encoding introduces the necessity of maintaining the ObjectIdMapping, which can be non-trivial.

Current Limitation

Currently, this id is fixed to being a 32-bit integer. However, we have an issue that tracks the progress on limiting this limitation.

Implementations

impl ObjectIdEncoding

pub fn new(object_id_size: u8) -> Self

Creates a new ObjectIdEncoding.

pub fn object_id_size(&self) -> u8

Returns the size of the object id.

Trait Implementations

impl Clone for ObjectIdEncoding

fn clone(&self) -> ObjectIdEncoding

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ObjectIdEncoding

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Hash for ObjectIdEncoding

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for ObjectIdEncoding

fn eq(&self, other: &ObjectIdEncoding) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl TermDecoder<ObjectIdEncoding> for DefaultObjectIdDecoder

type Term<'data> = ObjectId

The resulting value type of decoding an RDF term.

fn decode_terms( array: &<ObjectIdEncoding as TermEncoding>::Array, ) -> impl Iterator<Item = ThinResult<Self::Term<'_>>>

Allows extracting an iterator over all RDF terms in array that are compatible with this decoder (see TermDecoder for more information).

fn decode_term( scalar: &<ObjectIdEncoding as TermEncoding>::Scalar, ) -> ThinResult<Self::Term<'_>>

Allows extracting an iterator over all RDF terms in array that are compatible with this decoder (see TermDecoder for more information).

impl TermEncoding for ObjectIdEncoding

type Array = ObjectIdArray

Represents a wrapper for Arrow arrays of this encoding. This can be used in conjunction with TermDecoder to obtain the values from an Arrow array.

type Scalar = ObjectIdScalar

Represents a wrapper for Arrow scalars of this encoding. This can be used in conjunction with TermDecoder to obtain the values from an Arrow scalar.

fn name(&self) -> EncodingName

Returns the name of the encoding.

fn data_type(&self) -> DataType

Returns the DataType that is used for this encoding.

fn try_new_array(&self, array: ArrayRef) -> DFResult<Self::Array>

Checks whether array contains a value with the correct encoding (i.e., type and possibly metadata checks). If yes, returns an instance of Self::Array. Otherwise, an error is returned.

fn try_new_scalar(&self, scalar: ScalarValue) -> DFResult<Self::Scalar>

Checks whether scalar contains a value with the correct encoding (i.e., type and possibly metadata checks). If yes, returns an instance of Self::Scalar. Otherwise, an error is returned.

fn try_new_datum( &self, value: ColumnarValue, number_rows: usize, ) -> DFResult<EncodingDatum<Self>>

Checks whether value contains a value with the correct encoding (i.e., type and possibly metadata checks). If yes, returns a datum that either wraps an array or a scalar. Otherwise, an error is returned.

impl Eq for ObjectIdEncoding

impl StructuralPartialEq for ObjectIdEncoding