Enum BlobTraceUpdates

pub enum BlobTraceUpdates {
    Row(ColumnarRecords),
    Both(ColumnarRecords, ColumnarRecordsStructuredExt),
    Structured {
        key_values: ColumnarRecordsStructuredExt,
        timestamps: Int64Array,
        diffs: Int64Array,
    },
}

The set of updates that are part of a BlobTraceBatchPart.

Variants

Row(ColumnarRecords)

Legacy format. Keys and Values are encoded into bytes via Codec, then stored in our own columnar-esque struct.

Both(ColumnarRecords, ColumnarRecordsStructuredExt)

Migration format. Keys and Values are encoded into bytes via Codec and structured into an Apache Arrow columnar format.

Structured

New-style structured format, including structured representations of the K/V columns and the usual timestamp / diff encoding.

Fields

key_values: ColumnarRecordsStructuredExt

Key-value data.

timestamps: Int64Array

Timestamp data.

diffs: Int64Array

Diffs.

Implementations

impl BlobTraceUpdates

pub fn from_part(part: Part) -> Self

Convert from a Part.

pub fn len(&self) -> usize

The number of updates.

pub fn timestamps(&self) -> &Int64Array

The updates’ timestamps as an integer array.

pub fn diffs(&self) -> &Int64Array

The updates’ diffs as an integer array.

pub fn diffs_sum<D: Codec64 + Semigroup>(&self) -> Option<D>

Return the sum of the diffs in the blob.

pub fn records(&self) -> Option<&ColumnarRecords>

Return the ColumnarRecords of the blob.

pub fn structured(&self) -> Option<&ColumnarRecordsStructuredExt>

Return the ColumnarRecordsStructuredExt of the blob.

pub fn goodbytes(&self) -> usize

Return the estimated memory usage of the raw data.

pub fn get_or_make_codec<K: Codec, V: Codec>( &mut self, key_schema: &K::Schema, val_schema: &V::Schema, ) -> &ColumnarRecords

Return the ColumnarRecords of the blob, generating it if it does not exist.

pub fn get_or_make_structured<K: Codec, V: Codec>( &mut self, key_schema: &K::Schema, val_schema: &V::Schema, ) -> &ColumnarRecordsStructuredExt

Return the ColumnarRecordsStructuredExt of the blob.

pub fn as_part(&self) -> Option<Part>

If we have structured data, cast this data as a structured part.

pub fn into_part<K: Codec, V: Codec>( &mut self, key_schema: &K::Schema, val_schema: &V::Schema, ) -> Part

Convert this blob into a structured part, transforming the codec data if necessary.

pub fn concat<K: Codec, V: Codec>( updates: Vec<BlobTraceUpdates>, key_schema: &K::Schema, val_schema: &V::Schema, metrics: &ColumnarMetrics, ) -> Result<BlobTraceUpdates>

Concatenate the given records together, column-by-column.

If ensure_codec is true, then we’ll ensure the returned BlobTraceUpdates includes Codec data, re-encoding structured data if necessary.

pub fn from_proto( lgbytes: &ColumnarMetrics, proto: ProtoColumnarRecords, ) -> Result<Self, TryFromProtoError>

See RustType::from_proto.

pub fn into_proto(&self) -> ProtoColumnarRecords

See RustType::into_proto.

pub fn as_structured<K: Codec, V: Codec>( &self, key_schema: &K::Schema, val_schema: &V::Schema, ) -> Self

Convert these updates into the specified batch format, re-encoding or discarding key-value data as necessary.

Trait Implementations

impl Clone for BlobTraceUpdates

fn clone(&self) -> BlobTraceUpdates

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for BlobTraceUpdates

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

Formats the value using the given formatter.

impl PartialEq for BlobTraceUpdates

fn eq(&self, other: &BlobTraceUpdates) -> 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 StructuralPartialEq for BlobTraceUpdates