Parquet: Backport variant BINARY upper bound fix (#16880) to 1.9.x

parquet/src/main/java/org/apache/iceberg/parquet/ParquetVariantUtil.java

@@ -333,7 +333,7 @@ private static VariantValue truncateUpperBound(VariantValue value) {
           return truncatedString != null ? Variants.of(PhysicalType.STRING, truncatedString) : null;
         case BINARY:
           ByteBuffer truncatedBuffer =
-              BinaryUtil.truncateBinaryMin((ByteBuffer) value.asPrimitive().get(), 16);
+              BinaryUtil.truncateBinaryMax((ByteBuffer) value.asPrimitive().get(), 16);
           return truncatedBuffer != null ? Variants.of(PhysicalType.BINARY, truncatedBuffer) : null;
         default:
           return value;

parquet/src/test/java/org/apache/iceberg/parquet/TestVariantMetrics.java

@@ -199,6 +199,56 @@ public void testShreddedPrimitiveTypeMismatch(VariantValue value) throws IOExcep
         .isEqualTo(Map.of(1, Conversions.toByteBuffer(Type.TypeID.LONG, 1L)));
   }
 
+  @Test
+  public void testShreddedBinaryBoundsTruncation() throws IOException {
+    // binary longer than the 16-byte truncation length so the bounds are truncated
+    byte[] bytes = new byte[20];
+    for (int i = 0; i < bytes.length; i += 1) {
+      bytes[i] = (byte) (i + 1);
+    }
+    VariantValue value = Variants.of(ByteBuffer.wrap(bytes));
+
+    Metrics metrics =
+        writeParquet(
+            (id, name) -> ParquetVariantUtil.toParquetSchema(value),
+            Variant.of(EMPTY, value),
+            Variant.of(EMPTY, Variants.ofNull()),
+            null);
+
+    assertThat(metrics.lowerBounds().get(2))
+        .extracting(b -> Variant.from(b).value().asObject().get(ROOT_FIELD))
+        .isEqualTo(Variants.of(BinaryUtil.truncateBinaryMin(ByteBuffer.wrap(bytes), 16)));
+
+    assertThat(metrics.upperBounds().get(2))
+        .extracting(b -> Variant.from(b).value().asObject().get(ROOT_FIELD))
+        .isEqualTo(Variants.of(BinaryUtil.truncateBinaryMax(ByteBuffer.wrap(bytes), 16)));
+  }
+
+  @Test
+  public void testShreddedBinaryUpperBoundOverflow() throws IOException {
+    // an all-0xFF binary cannot be truncated up so the upper bound is omitted
+    byte[] bytes = new byte[20];
+    for (int i = 0; i < bytes.length; i += 1) {
+      bytes[i] = (byte) 0xFF;
+    }
+    VariantValue value = Variants.of(ByteBuffer.wrap(bytes));
+
+    Metrics metrics =
+        writeParquet(
+            (id, name) -> ParquetVariantUtil.toParquetSchema(value),
+            Variant.of(EMPTY, value),
+            Variant.of(EMPTY, Variants.ofNull()),
+            null);
+
+    assertThat(metrics.lowerBounds().get(2))
+        .extracting(b -> Variant.from(b).value().asObject().get(ROOT_FIELD))
+        .isEqualTo(Variants.of(BinaryUtil.truncateBinaryMin(ByteBuffer.wrap(bytes), 16)));
+
+    assertThat(metrics.upperBounds().get(2))
+        .extracting(b -> Variant.from(b).value().asObject().get(ROOT_FIELD))
+        .isNull();
+  }
+
   @Test
   public void testVariantFloatNaN() throws IOException {
     // NaN values are not counted because there is no ID for FieldMetrics