From 45165dacca76b93c23d0351e61614ab792f976e6 Mon Sep 17 00:00:00 2001
From: Gengliang Wang <gengliang@apache.org>
Date: Sat, 30 May 2026 11:06:26 +0000
Subject: [PATCH] [SPARK-57172][SQL] Simplify Crc32 codegen by extracting a
 static Java helper

### What changes were proposed in this pull request?

Add `ExpressionImplUtils.crc32(byte[] bytes)` and route `Crc32`'s eval and
codegen paths through it. `Crc32.doGenCode` previously emitted a 3-line
allocate / `update` / `getValue` sequence inline; it now emits a single
`ExpressionImplUtils.crc32(...)` call, and the eval path calls the same helper.

This is a plain (non-ANSI, non-try/catch) type-independent block, in line with
the broadened goal of SPARK-56908 to move fixed generated-Java logic into static
Java helpers.

### Why are the changes needed?

Part of SPARK-56908 (umbrella). Collapsing the inline CRC32 sequence to one call
shrinks the generated Java for every stage that computes `crc32`.

### Does this PR introduce _any_ user-facing change?

No. The compiled behavior is identical; only the emitted Java source text changes.

### How was this patch tested?

```
build/sbt "catalyst/testOnly *HashExpressionsSuite"
```

40/40 pass, including `crc32` (exercised both with and without whole-stage codegen).

### Was this patch authored or co-authored using generative AI tooling?

Generated-by: Claude Code (Opus 4.8)

Co-authored-by: Isaac
---
 .../catalyst/expressions/ExpressionImplUtils.java  | 12 ++++++++++++
 .../spark/sql/catalyst/expressions/hash.scala      | 14 +++-----------
 2 files changed, 15 insertions(+), 11 deletions(-)

diff --git a/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/expressions/ExpressionImplUtils.java b/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/expressions/ExpressionImplUtils.java
index a5228edc33c83..7bad7c430b862 100644
--- a/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/expressions/ExpressionImplUtils.java
+++ b/sql/catalyst/src/main/java/org/apache/spark/sql/catalyst/expressions/ExpressionImplUtils.java
@@ -25,6 +25,7 @@
 import java.util.ArrayList;
 import java.util.List;
 import java.util.Locale;
+import java.util.zip.CRC32;
 import javax.crypto.Cipher;
 import javax.crypto.spec.GCMParameterSpec;
 import javax.crypto.spec.IvParameterSpec;
@@ -342,4 +343,15 @@ public static UTF8String quote(UTF8String str) {
     String sp = str.toString().replaceAll(qtChar, qtCharRep);
     return UTF8String.fromString(qtChar + sp + qtChar);
   }
+
+  /**
+   * Computes the CRC32 checksum of {@code bytes} for the {@code crc32} expression.
+   * Shared by the eval and codegen paths so the per-stage generated Java is a
+   * single call rather than an inline allocate / update / getValue sequence.
+   */
+  public static long crc32(byte[] bytes) {
+    CRC32 checksum = new CRC32();
+    checksum.update(bytes, 0, bytes.length);
+    return checksum.getValue();
+  }
 }
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/hash.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/hash.scala
index 0f27dee9dbc84..0b0c84176691e 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/hash.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/hash.scala
@@ -19,7 +19,6 @@ package org.apache.spark.sql.catalyst.expressions
 
 import java.math.{BigDecimal, RoundingMode}
 import java.util.concurrent.TimeUnit._
-import java.util.zip.CRC32
 
 import scala.annotation.tailrec
 
@@ -214,20 +213,13 @@ case class Crc32(child: Expression)
   override def contextIndependentFoldable: Boolean = child.contextIndependentFoldable
 
   protected override def nullSafeEval(input: Any): Any = {
-    val checksum = new CRC32
-    checksum.update(input.asInstanceOf[Array[Byte]], 0, input.asInstanceOf[Array[Byte]].length)
-    checksum.getValue
+    ExpressionImplUtils.crc32(input.asInstanceOf[Array[Byte]])
   }
 
   override def doGenCode(ctx: CodegenContext, ev: ExprCode): ExprCode = {
-    val CRC32 = "java.util.zip.CRC32"
-    val checksum = ctx.freshName("checksum")
+    val utils = classOf[ExpressionImplUtils].getName
     nullSafeCodeGen(ctx, ev, value => {
-      s"""
-        $CRC32 $checksum = new $CRC32();
-        $checksum.update($value, 0, $value.length);
-        ${ev.value} = $checksum.getValue();
-      """
+      s"${ev.value} = $utils.crc32($value);"
     })
   }