birschick-bq commented on code in PR #2018:
URL: https://github.com/apache/arrow-adbc/pull/2018#discussion_r1739234738
##########
csharp/src/Drivers/Apache/Thrift/Service/Rpc/Thrift/TDoubleColumn.cs:
##########
@@ -78,14 +79,19 @@ public TDoubleColumn DeepCopy()
var _list178 = await
iprot.ReadListBeginAsync(cancellationToken);
length = _list178.Count;
- buffer = new byte[length * 8];
+ buffer = new byte[length * DoubleSize];
var memory = buffer.AsMemory();
var typedMemory = Unsafe.As<Memory<byte>, Memory<long>>(ref
memory).Slice(0, length);
iprot.Transport.CheckReadBytesAvailable(buffer.Length);
await transport.ReadExactlyAsync(memory, cancellationToken);
for (int _i179 = 0; _i179 < length; ++_i179)
{
+#if NET6_0_OR_GREATER
typedMemory.Span[_i179] =
BinaryPrimitives.ReverseEndianness(typedMemory.Span[_i179]);
+#else
+ long source =
BinaryPrimitives.ReverseEndianness(BitConverter.ToInt64(buffer, _i179 *
DoubleSize));
+ BitConverter.GetBytes(source).CopyTo(buffer, _i179 *
DoubleSize);
Review Comment:
@CurtHagenlocher
Just an explanation. It seems the behavior of the Unsafe implementation is
different before NET 6 - hence the conditional compilation. I'll attempt a more
efficient implementation for this case.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
