(orc) branch main updated: ORC-1595: Add a short-cut to skip tiny inputs for `ZstdCodec.compress`

[dongjoon]

This is an automated email from the ASF dual-hosted git repository.

dongjoon pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/orc.git


The following commit(s) were added to refs/heads/main by this push:
     new 6b75eb18c ORC-1595: Add a short-cut to skip tiny inputs for 
`ZstdCodec.compress`
6b75eb18c is described below

commit 6b75eb18c6374eb7b445680981eea21648a4cb30
Author: Dongjoon Hyun <dongj...@apache.org>
AuthorDate: Thu Jan 18 13:57:32 2024 -0800

    ORC-1595: Add a short-cut to skip tiny inputs for `ZstdCodec.compress`
    
    ### What changes were proposed in this pull request?
    
    For `ZstdCodec.compress`, this PR aims to add a short-cut to skip tiny 
inputs which is smaller than the minimum ZStandard format.
    
    ### Why are the changes needed?
    
    To reduce the number of JNI call.
    
    ### How was this patch tested?
    
    Pass the CIs.
    
    ### Was this patch authored or co-authored using generative AI tooling?
    
    No.
    
    Closes #1762 from dongjoon-hyun/ORC-1595.
    
    Authored-by: Dongjoon Hyun <dongj...@apache.org>
    Signed-off-by: Dongjoon Hyun <dongj...@apache.org>
---
 java/core/src/java/org/apache/orc/impl/ZstdCodec.java | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/java/core/src/java/org/apache/orc/impl/ZstdCodec.java 
b/java/core/src/java/org/apache/orc/impl/ZstdCodec.java
index 8b5cd24e1..3400f2890 100644
--- a/java/core/src/java/org/apache/orc/impl/ZstdCodec.java
+++ b/java/core/src/java/org/apache/orc/impl/ZstdCodec.java
@@ -171,6 +171,12 @@ public class ZstdCodec implements CompressionCodec {
   public boolean compress(ByteBuffer in, ByteBuffer out,
       ByteBuffer overflow,
       Options options) throws IOException {
+    int inBytes = in.remaining();
+    // Skip with minimum ZStandard format size:
+    // https://datatracker.ietf.org/doc/html/rfc8878#name-zstandard-frames
+    // Magic Number (4 bytes) + Frame Header (2 bytes) + Data Block Header (3 
bytes)
+    if (inBytes < 10) return false;
+
     ZstdOptions zso = (ZstdOptions) options;
 
     zstdCompressCtx = new ZstdCompressCtx();
@@ -179,7 +185,6 @@ public class ZstdCodec implements CompressionCodec {
     zstdCompressCtx.setChecksum(false);
 
     try {
-      int inBytes = in.remaining();
       byte[] compressed = getBuffer((int) Zstd.compressBound(inBytes));
 
       int outBytes = zstdCompressCtx.compressByteArray(compressed, 0, 
compressed.length,