Skip to content

Avoid Boxing/Unboxing on accessing elements of VBufferDataFrameColumn (fix merge issues) #6867

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Oct 17, 2023
Merged

Avoid Boxing/Unboxing on accessing elements of VBufferDataFrameColumn (fix merge issues) #6867

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c77f5f7
Avoid Boxing/Unboxing on accessing elements of VBufferDataFrameColumn
asmirnov82 Oct 16, 2023
f0ac954
Avoid boxing for vbuffer column
asmirnov82 Oct 17, 2023
67036d0
Merge remote-tracking branch 'origin/main' into avoid_boxing_for_vbuf…
asmirnov82 Oct 17, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
38 changes: 20 additions & 18 deletions src/Microsoft.Data.Analysis/VBufferDataFrameColumn.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,6 @@ namespace Microsoft.Data.Analysis
/// </summary>
public partial class VBufferDataFrameColumn<T> : DataFrameColumn, IEnumerable<VBuffer<T>>
{

public static int MaxCapacity = ArrayUtility.ArrayMaxSize / Unsafe.SizeOf<VBuffer<T>>();

private readonly List<List<VBuffer<T>>> _vBuffers = new List<List<VBuffer<T>>>(); // To store more than intMax number of vbuffers
Expand Down Expand Up @@ -56,9 +55,7 @@ public VBufferDataFrameColumn(string name, IEnumerable<VBuffer<T>> values) : bas
}
}

private long _nullCount;

public override long NullCount => _nullCount;
public override long NullCount => 0;

protected internal override void Resize(long length)
{
Expand Down Expand Up @@ -94,6 +91,11 @@ private int GetBufferIndexContainingRowIndex(long rowIndex)
}

protected override object GetValue(long rowIndex)
{
return GetTypedValue(rowIndex);
}

protected VBuffer<T> GetTypedValue(long rowIndex)
{
int bufferIndex = GetBufferIndexContainingRowIndex(rowIndex);
return _vBuffers[bufferIndex][(int)(rowIndex % MaxCapacity)];
Expand All @@ -118,30 +120,30 @@ protected override IReadOnlyList<object> GetValues(long startIndex, int length)

protected override void SetValue(long rowIndex, object value)
{
if (value == null || value is VBuffer<T>)
if (value == null)
{
int bufferIndex = GetBufferIndexContainingRowIndex(rowIndex);
int bufferOffset = (int)(rowIndex % MaxCapacity);
var oldValue = _vBuffers[bufferIndex][bufferOffset];
_vBuffers[bufferIndex][bufferOffset] = (VBuffer<T>)value;
if (!oldValue.Equals((VBuffer<T>)value))
{
if (value == null)
_nullCount++;
if (oldValue.Length == 0 && _nullCount > 0)
_nullCount--;
}
throw new NotSupportedException("Null values are not supported by VBufferDataFrameColumn");
}
else if (value is VBuffer<T> vbuffer)
{
SetTypedValue(rowIndex, vbuffer);
}
else
{
throw new ArgumentException(string.Format(Strings.MismatchedValueType, typeof(VBuffer<T>)), nameof(value));
}
}

protected void SetTypedValue(long rowIndex, VBuffer<T> value)
{
int bufferIndex = GetBufferIndexContainingRowIndex(rowIndex);
_vBuffers[bufferIndex][(int)(rowIndex % MaxCapacity)] = value;
}

public new VBuffer<T> this[long rowIndex]
{
get => (VBuffer<T>)GetValue(rowIndex);
set => SetValue(rowIndex, value);
get => GetTypedValue(rowIndex);
set => SetTypedValue(rowIndex, value);
}

/// <summary>
Expand Down