GH-3449: Fix BinaryStats int overflow (#3448)

Author: Jiayi-Wang-db

parquet-column/src/main/java/org/apache/parquet/column/statistics/BinaryStatistics.java

@@ -105,16 +105,16 @@ String stringify(Binary value) {
 
   @Override
   public boolean isSmallerThan(long size) {
-    return !hasNonNullValue() || ((min.length() + max.length()) < size);
+    return !hasNonNullValue() || (((long) min.length() + max.length()) < size);
   }
 
   public boolean isSmallerThanWithTruncation(long size, int truncationLength) {
     if (!hasNonNullValue()) {
       return true;
     }
 
-    int minTruncateLength = Math.min(min.length(), truncationLength);
-    int maxTruncateLength = Math.min(max.length(), truncationLength);
+    long minTruncateLength = Math.min(min.length(), truncationLength);
+    long maxTruncateLength = Math.min(max.length(), truncationLength);
 
     return minTruncateLength + maxTruncateLength < size;
   }

parquet-column/src/test/java/org/apache/parquet/column/statistics/TestStatistics.java

@@ -927,4 +927,15 @@ public void testNoopStatistics() {
     assertThrows(UnsupportedOperationException.class, stats::minAsString);
     assertThrows(UnsupportedOperationException.class, () -> stats.isSmallerThan(0));
   }
+
+  @Test
+  public void testBinaryIsSmallerThanNoOverflowForLargeValues() {
+    BinaryStatistics stats = new BinaryStatistics();
+    // Create a Binary whose length() reports 2^30 without allocating 1 GB
+    Binary fakeLarge = Binary.fromConstantByteArray(new byte[0], 0, 1 << 30);
+    stats.updateStats(fakeLarge);
+
+    // min.length() + max.length() = 2^31, must not overflow int to negative
+    assertFalse(stats.isSmallerThan(4096));
+  }
 }