Struct IndexCollector

Namespace

NumSharp.Backends.Kernels

Assembly

NumSharp.dll

A growable buffer for collecting long indices, backed by NDArray storage. Replaces LongIndexBuffer - uses NumSharp's existing unmanaged memory infrastructure.

public struct IndexCollector

Inherited Members

ValueType.Equals(object)

ValueType.GetHashCode()

ValueType.ToString()

object.Equals(object, object)

object.GetType()

object.ReferenceEquals(object, object)

Extension Methods

LinqExtensions.Yield<T>(T)

Remarks

Benefits over LongIndexBuffer:

• Uses NDArray's proven unmanaged storage
• No separate Dispose needed (NDArray handles cleanup)
• ToResult() returns the storage directly (no copy for final result)
• Integrates with NumSharp's memory management

Growth strategy:

• Doubles capacity below 1 billion elements
• 33% growth above 1 billion (matches Hashset/LongList pattern)

Constructors

IndexCollector(long)

Creates a new collector with the specified initial capacity.

public IndexCollector(long initialCapacity)

Parameters

initialCapacity long

Initial capacity (number of elements).

Properties

Count

Number of elements collected.

public long Count { get; }

Property Value

long

Data

Pointer to the underlying data.

public long* Data { get; }

Property Value

long*

this[long]

Gets the value at the specified index.

public long this[long index] { get; }

Parameters

index long

Property Value

long

Methods

Add(long)

Adds a value, growing the storage if necessary.

public void Add(long value)

Parameters

value long

ToResult()

Returns the collected indices as an NDArray, trimmed to actual count.

public NDArray<long> ToResult()

Returns

NDArray<long>

Remarks

If count equals capacity, returns storage directly (no copy). Otherwise, creates a properly-sized copy.