Skip to content

Reduce overhead from Slice#getInt in PlainByteArrayDecoders #24284

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 1 commit into from
Dec 2, 2024
Merged

Reduce overhead from Slice#getInt in PlainByteArrayDecoders #24284

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
32 changes: 31 additions & 1 deletion lib/trino-parquet/src/main/java/io/trino/parquet/reader/decoders/PlainByteArrayDecoders.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -159,7 +159,7 @@ public void read(BinaryBuffer values, int offset, int length)
currentInputOffset += positionLength + Integer.BYTES;
}

createOutputBuffer(values, offset, length, inputSlice, outputBufferSize);
values.addChunk(createOutputBuffer(values.getOffsets(), offset, length, inputSlice, outputBufferSize));
input.skip(currentInputOffset);
}

Expand All @@ -168,6 +168,36 @@ public void skip(int n)
{
skipPlainValues(input, n);
}

/**
* Create one big slice of data and add it to the output buffer since buffer size is known.
* Specialized for the case of strings with no truncation
*/
private static Slice createOutputBuffer(int[] offsets, int offset, int length, Slice inputSlice, int outputBufferSize)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

so offsets have double meaning? As in it's length as out it's offset? Would be good to have that commented

{
byte[] outputBuffer = new byte[outputBufferSize];
int currentInputOffset = 0;
int currentOutputOffset = 0;

byte[] inputArray;
int inputArrayOffset;
if (length != 0) {
inputArray = inputSlice.byteArray();
inputArrayOffset = inputSlice.byteArrayOffset();
}
else {
inputArray = new byte[0];
inputArrayOffset = 0;
}
for (int i = 0; i < length; i++) {
int positionLength = offsets[offset + i + 1];
arraycopy(inputArray, inputArrayOffset + currentInputOffset + Integer.BYTES, outputBuffer, currentOutputOffset, positionLength);
offsets[offset + i + 1] = offsets[offset + i] + positionLength;
currentInputOffset += positionLength + Integer.BYTES;
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why add BYTES here and in arraycopy? Just add BYTES * 2?

currentOutputOffset += positionLength;
}
return Slices.wrappedBuffer(outputBuffer);
}
}

private static void skipPlainValues(SimpleSliceInputStream input, int n)
Expand Down