liukun4515 commented on code in PR #1855:
URL: https://github.com/apache/arrow-rs/pull/1855#discussion_r903715450
##########
arrow/src/ipc/reader.rs:
##########
@@ -32,15 +33,74 @@ use crate::error::{ArrowError, Result};
use crate::ipc;
use crate::record_batch::{RecordBatch, RecordBatchOptions, RecordBatchReader};
+use crate::ipc::compression::ipc_compression::CompressionCodecType;
+use crate::ipc::compression::{
+ LENGTH_EMPTY_COMPRESSED_DATA, LENGTH_NO_COMPRESSED_DATA,
LENGTH_OF_PREFIX_DATA,
+};
+use crate::ipc::CompressionType;
use ipc::CONTINUATION_MARKER;
use DataType::*;
/// Read a buffer based on offset and length
-fn read_buffer(buf: &ipc::Buffer, a_data: &[u8]) -> Buffer {
+/// From
<https://github.com/apache/arrow/blob/6a936c4ff5007045e86f65f1a6b6c3c955ad5103/format/Message.fbs#L58>
+/// Each constituent buffer is first compressed with the indicated
+/// compressor, and then written with the uncompressed length in the first 8
+/// bytes as a 64-bit little-endian signed integer followed by the compressed
+/// buffer bytes (and then padding as required by the protocol). The
+/// uncompressed length may be set to -1 to indicate that the data that
+/// follows is not compressed, which can be useful for cases where
+/// compression does not yield appreciable savings.
+fn read_buffer(
+ buf: &ipc::Buffer,
+ a_data: &[u8],
+ compression_codec: &CompressionCodecType,
+) -> Buffer {
let start_offset = buf.offset() as usize;
let end_offset = start_offset + buf.length() as usize;
let buf_data = &a_data[start_offset..end_offset];
- Buffer::from(&buf_data)
+ // corner case: empty buffer
+ if buf_data.is_empty() {
+ return Buffer::from(buf_data);
+ }
+ match compression_codec {
+ CompressionCodecType::NoCompression => Buffer::from(buf_data),
+ CompressionCodecType::Lz4Frame | CompressionCodecType::Zstd => {
+ // 8byte + data
+ // read the first 8 bytes
+ // if the data is compressed, decompress the data, otherwise
return as is
+ let decompressed_length = read_uncompressed_size(buf_data);
+ if decompressed_length == LENGTH_EMPTY_COMPRESSED_DATA {
+ // emtpy
+ let empty = Vec::<u8>::new();
+ Buffer::from(empty)
+ } else if decompressed_length == LENGTH_NO_COMPRESSED_DATA {
+ // not compress
+ let data = &buf_data[(LENGTH_OF_PREFIX_DATA as usize)..];
+ Buffer::from(data)
+ } else {
+ // decompress data using the codec
+ let mut _uncompressed_buffer = Vec::new();
+ let _input_data = &buf_data[(LENGTH_OF_PREFIX_DATA as
usize)..];
+ // TODO consider the error result
+ #[cfg(any(feature = "zstd,lz4", test))]
Review Comment:
agree with you, @viirya . I have no idea and how to control this.
Could you please give me some suggestion or advice?
--
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]
