[SPARK-52181][4.0] Increase variant size limit to 128MiB

common/variant/src/main/java/org/apache/spark/types/variant/VariantBuilder.java

@@ -542,12 +542,13 @@ private void buildJson(JsonParser parser) throws IOException {
   }
 
   // Choose the smallest unsigned integer type that can store `value`. It must be within
-  // `[0, U24_MAX]`.
+  // `[0, SIZE_LIMIT]`.
   private int getIntegerSize(int value) {
-    assert value >= 0 && value <= U24_MAX;
+    assert value >= 0 && value <= SIZE_LIMIT;
     if (value <= U8_MAX) return 1;
     if (value <= U16_MAX) return 2;
-    return U24_SIZE;
+    if (value <= U24_MAX) return 3;
+    return 4;
   }
 
   private void parseFloatingPoint(JsonParser parser) throws IOException {

common/variant/src/main/java/org/apache/spark/types/variant/VariantUtil.java

@@ -137,8 +137,8 @@ public class VariantUtil {
   public static final int U24_SIZE = 3;
   public static final int U32_SIZE = 4;
 
-  // Both variant value and variant metadata need to be no longer than 16MiB.
-  public static final int SIZE_LIMIT = U24_MAX + 1;
+  // Both variant value and variant metadata need to be no longer than 128MiB.
+  public static final int SIZE_LIMIT = 128 * 1024 * 1024;
 
   public static final int MAX_DECIMAL4_PRECISION = 9;
   public static final int MAX_DECIMAL8_PRECISION = 18;

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/variant/VariantExpressionEvalUtilsSuite.scala

@@ -133,10 +133,9 @@ class VariantExpressionEvalUtilsSuite extends SparkFunSuite {
       checkException(json, "MALFORMED_RECORD_IN_PARSING.WITHOUT_SUGGESTION",
         Map("badRecord" -> json, "failFastMode" -> "FAILFAST"))
     }
-    for (json <- Seq("\"" + "a" * (16 * 1024 * 1024) + "\"",
-      (0 to 4 * 1024 * 1024).mkString("[", ",", "]"))) {
+    for (json <- Seq((0 to 32 * 1024 * 1024).mkString("[", ",", "]"))) {
       checkException(json, "VARIANT_SIZE_LIMIT",
-        Map("sizeLimit" -> "16.0 MiB", "functionName" -> "`parse_json`"))
+        Map("sizeLimit" -> "128.0 MiB", "functionName" -> "`parse_json`"))
     }
   }
 

sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/expressions/variant/VariantExpressionSuite.scala

@@ -93,21 +93,21 @@ class VariantExpressionSuite extends SparkFunSuite with ExpressionEvalHelper {
     check(Array(primitiveHeader(INT1), 0), Array[Byte](3, 0, 0))
     check(Array(primitiveHeader(INT1), 0), Array[Byte](2, 0, 0))
 
-    // Construct binary values that are over 1 << 24 bytes, but otherwise valid.
+    // Construct binary values that are over SIZE_LIMIT bytes, but otherwise valid.
     val bigVersion = Array[Byte]((VERSION | (3 << 6)).toByte)
-    val a = Array.fill(1 << 24)('a'.toByte)
+    val a = Array.fill(SIZE_LIMIT)('a'.toByte)
     val hugeMetadata = bigVersion ++ Array[Byte](2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1) ++
       a ++ Array[Byte]('b')
     check(Array(primitiveHeader(TRUE)), hugeMetadata, "VARIANT_CONSTRUCTOR_SIZE_LIMIT")
 
     // The keys are 'aaa....' and 'b'. Values are "yyy..." and 'true'.
-    val y = Array.fill(1 << 24)('y'.toByte)
+    val y = Array.fill(SIZE_LIMIT)('y'.toByte)
     val hugeObject = Array[Byte](objectHeader(true, 4, 4)) ++
       /* size */ padded(Array(2), 4) ++
       /* id list */ padded(Array(0, 1), 4) ++
-      // Second value starts at offset 5 + (1 << 24), which is `5001` little-endian. The last value
-      // is 1 byte, so the one-past-the-end value is `6001`
-      /* offset list */ Array[Byte](0, 0, 0, 0, 5, 0, 0, 1, 6, 0, 0, 1) ++
+      // Second value starts at offset 5 + (SIZE_LIMIT), which is `5008` little-endian. The last
+      // value is 1 byte, so the one-past-the-end value is `6008`
+      /* offset list */ Array[Byte](0, 0, 0, 0, 5, 0, 0, 8, 6, 0, 0, 8) ++
       /* field data */ Array[Byte](primitiveHeader(LONG_STR), 0, 0, 0, 1) ++ y ++ Array[Byte](
         primitiveHeader(TRUE)
       )

sql/core/src/test/scala/org/apache/spark/sql/VariantEndToEndSuite.scala

@@ -135,7 +135,8 @@ class VariantEndToEndSuite extends QueryTest with SharedSparkSession {
     check("{1:2}", null)
     check("{\"a\":1", null)
     check("{\"a\":[a,b,c]}", null)
-    check("\"" + "a" * (16 * 1024 * 1024) + "\"", null)
+    check("\"" + "a" * (16 * 1024 * 1024) + "\"")
+    check("\"" + "a" * (128 * 1024 * 1024) + "\"", null)
   }
 
   test("to_json with nested variant") {