Trait Array

pub trait Array:
    Send
    + Sync
    + Debug
    + ArrayStatistics
    + ArrayVariants
    + ArrayVisitor {
    // Required methods
    fn as_any(&self) -> &dyn Any;
    fn as_any_arc(self: Arc<Self>) -> Arc<dyn Any + Send + Sync>;
    fn to_array(&self) -> ArrayRef;
    fn into_array(self) -> ArrayRef
       where Self: Sized;
    fn len(&self) -> usize;
    fn dtype(&self) -> &DType;
    fn encoding(&self) -> EncodingId;
    fn vtable(&self) -> VTableRef;
    fn is_valid(&self, index: usize) -> VortexResult<bool>;
    fn is_invalid(&self, index: usize) -> VortexResult<bool>;
    fn all_valid(&self) -> VortexResult<bool>;
    fn all_invalid(&self) -> VortexResult<bool>;
    fn valid_count(&self) -> VortexResult<usize>;
    fn invalid_count(&self) -> VortexResult<usize>;
    fn validity_mask(&self) -> VortexResult<Mask>;
    fn to_canonical(&self) -> VortexResult<Canonical>;
    fn append_to_builder(
        &self,
        builder: &mut dyn ArrayBuilder,
    ) -> VortexResult<()>;
    fn statistics(&self) -> StatsSetRef<'_>;
    fn with_children(&self, children: &[ArrayRef]) -> VortexResult<ArrayRef>;
    fn invoke(
        &self,
        compute_fn: &ComputeFn,
        args: &InvocationArgs<'_>,
    ) -> VortexResult<Option<Output>>;

    // Provided methods
    fn is_empty(&self) -> bool { ... }
    fn is_encoding(&self, encoding: EncodingId) -> bool { ... }
    fn is_arrow(&self) -> bool { ... }
    fn is_canonical(&self) -> bool { ... }
}

The base trait for all Vortex arrays.

Users should invoke functions on this trait. Implementations should implement the corresponding function on the _Impl traits, e.g. ArrayValidityImpl. The functions here dispatch to the implementations, while validating pre- and post-conditions.

Required Methods

fn as_any(&self) -> &dyn Any

Returns the array as a reference to a generic Any trait object.

fn as_any_arc(self: Arc<Self>) -> Arc<dyn Any + Send + Sync>

Returns the array as an Arc reference to a generic Any trait object.

fn to_array(&self) -> ArrayRef

Returns the array as an ArrayRef.

fn into_array(self) -> ArrayRef

where Self: Sized,

Converts the array into an ArrayRef.

fn len(&self) -> usize

Returns the length of the array.

fn dtype(&self) -> &DType

Returns the logical Vortex DType of the array.

fn encoding(&self) -> EncodingId

Returns the encoding of the array.

fn vtable(&self) -> VTableRef

Returns the encoding VTable.

fn is_valid(&self, index: usize) -> VortexResult<bool>

Returns whether the item at index is valid.

fn is_invalid(&self, index: usize) -> VortexResult<bool>

Returns whether the item at index is invalid.

fn all_valid(&self) -> VortexResult<bool>

Returns whether all items in the array are valid.

This is usually cheaper than computing a precise valid_count.

fn all_invalid(&self) -> VortexResult<bool>

Returns whether the array is all invalid.

This is usually cheaper than computing a precise invalid_count.

fn valid_count(&self) -> VortexResult<usize>

Returns the number of valid elements in the array.

fn invalid_count(&self) -> VortexResult<usize>

Returns the number of invalid elements in the array.

fn validity_mask(&self) -> VortexResult<Mask>

Returns the canonical validity mask for the array.

fn to_canonical(&self) -> VortexResult<Canonical>

Returns the canonical representation of the array.

fn append_to_builder(&self, builder: &mut dyn ArrayBuilder) -> VortexResult<()>

Writes the array into the canonical builder.

The DType of the builder must match that of the array.

fn statistics(&self) -> StatsSetRef<'_>

Returns the statistics of the array.

fn with_children(&self, children: &[ArrayRef]) -> VortexResult<ArrayRef>

Replaces the children of the array with the given array references.

fn invoke( &self, compute_fn: &ComputeFn, args: &InvocationArgs<'_>, ) -> VortexResult<Option<Output>>

Optionally invoke a kernel for the given compute function.

These encoding-specific kernels are independent of kernels registered directly with compute functions using ComputeFn::register_kernel, and are attempted only if none of the function-specific kernels returns a result.

This allows encodings the opportunity to generically implement many compute functions that share some property, for example ComputeFn::is_elementwise, without prior knowledge of the function itself, while still allowing users to override the implementation of compute functions for built-in encodings. For an example, see the implementation for chunked arrays.

The first input in the InvocationArgs is always the array itself.

Warning: do not call compute_fn.invoke(args) directly, as this will result in a recursive call.

Provided Methods

fn is_empty(&self) -> bool

Returns whether the array is empty (has zero rows).

fn is_encoding(&self, encoding: EncodingId) -> bool

Returns whether the array is of the given encoding.

fn is_arrow(&self) -> bool

Returns whether this array is an arrow encoding.

fn is_canonical(&self) -> bool

Whether the array is of a canonical encoding.

Implementations

impl dyn Array + '_

pub fn serialize( &self, ctx: &ArrayContext, options: &SerializeOptions, ) -> Vec<ByteBuffer>

Serialize the array into a sequence of byte buffers that should be written contiguously. This function returns a vec to avoid copying data buffers.

Optionally, padding can be included to guarantee buffer alignment and ensure zero-copy reads within the context of an external file or stream. In this case, the alignment of the first byte buffer should be respected when writing the buffers to the stream or file.

The format of this blob is a sequence of data buffers, possible with prefixed padding, followed by a flatbuffer containing an fba::Array message, and ending with a little-endian u32 describing the length of the flatbuffer message.

impl dyn Array + '_

pub fn tree_display(&self) -> impl Display

Trait Implementations

impl AsRef<dyn Array> for Canonical

fn as_ref(&self) -> &(dyn Array + 'static)

Converts this type into a shared reference of the (usually inferred) input type.

impl Display for dyn Array

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

Formats the value using the given formatter.

impl<'a> From<&'a dyn Array> for Input<'a>

fn from(value: &'a dyn Array) -> Self

Converts to this type from the input type.

impl IndexOrd<Scalar> for dyn Array + '_

fn index_cmp(&self, idx: usize, elem: &Scalar) -> Option<Ordering>

PartialOrd of the value at index idx with elem. For example, if self[idx] > elem, return Some(Greater).

fn len(&self) -> usize

Get the length of the underlying ordered collection

fn index_lt(&self, idx: usize, elem: &V) -> bool

fn index_le(&self, idx: usize, elem: &V) -> bool

fn index_gt(&self, idx: usize, elem: &V) -> bool

fn index_ge(&self, idx: usize, elem: &V) -> bool

impl<E> LikeFn<&dyn Array> for E

where E: for<'a> LikeFn<&'a E::Array> + Encoding,

fn like( &self, array: &dyn Array, pattern: &dyn Array, options: LikeOptions, ) -> VortexResult<Option<ArrayRef>>

impl<E> TakeFromFn<&dyn Array> for E

where E: for<'a> TakeFromFn<&'a E::Array> + Encoding,

fn take_from( &self, indices: &dyn Array, array: &dyn Array, ) -> VortexResult<Option<ArrayRef>>

impl ToOwned for dyn Array

type Owned = Arc<dyn Array>

The resulting type after obtaining ownership.

fn to_owned(&self) -> Self::Owned

Creates owned data from borrowed data, usually by cloning.

fn clone_into(&self, target: &mut Self::Owned)

Uses borrowed data to replace owned data, usually by cloning.

impl TryFrom<&dyn Array> for Mask

fn try_from(array: &dyn Array) -> Result<Self, Self::Error>

Converts from a possible nullable boolean array. Null values are treated as false.

type Error = VortexError

The type returned in the event of a conversion error.

impl TryFrom<&dyn Array> for RecordBatch

type Error = VortexError

The type returned in the event of a conversion error.

fn try_from(value: &dyn Array) -> VortexResult<Self>

Performs the conversion.

Implementations on Foreign Types

impl Array for Arc<dyn Array>

fn as_any(&self) -> &dyn Any

fn as_any_arc(self: Arc<Self>) -> Arc<dyn Any + Send + Sync>

fn to_array(&self) -> ArrayRef

fn into_array(self) -> ArrayRef

fn len(&self) -> usize

fn dtype(&self) -> &DType

fn encoding(&self) -> EncodingId

fn vtable(&self) -> VTableRef

fn is_valid(&self, index: usize) -> VortexResult<bool>

fn is_invalid(&self, index: usize) -> VortexResult<bool>

fn all_valid(&self) -> VortexResult<bool>

fn all_invalid(&self) -> VortexResult<bool>

fn valid_count(&self) -> VortexResult<usize>

fn invalid_count(&self) -> VortexResult<usize>

fn validity_mask(&self) -> VortexResult<Mask>

fn to_canonical(&self) -> VortexResult<Canonical>

fn append_to_builder(&self, builder: &mut dyn ArrayBuilder) -> VortexResult<()>

fn statistics(&self) -> StatsSetRef<'_>

fn with_children(&self, children: &[ArrayRef]) -> VortexResult<ArrayRef>

fn invoke( &self, compute_fn: &ComputeFn, args: &InvocationArgs<'_>, ) -> VortexResult<Option<Output>>

Implementors

impl<A: ArrayImpl + 'static> Array for A