GH-343: Fix ListVector offset buffer not properly serialized for nested empty arrays

vector/src/main/java/org/apache/arrow/vector/complex/LargeListVector.java

@@ -309,11 +309,14 @@ private void setReaderAndWriterIndex() {
     offsetBuffer.readerIndex(0);
     if (valueCount == 0) {
       validityBuffer.writerIndex(0);
-      offsetBuffer.writerIndex(0);
     } else {
       validityBuffer.writerIndex(BitVectorHelper.getValidityBufferSizeFromCount(valueCount));
-      offsetBuffer.writerIndex((valueCount + 1) * OFFSET_WIDTH);
     }
+    // IPC serializer will determine readable bytes based on `readerIndex` and `writerIndex`.
+    // Both are set to 0 means 0 bytes are written to the IPC stream which will crash IPC readers
+    // in other libraries. According to Arrow spec, we should still output the offset buffer which
+    // is [0].
+    offsetBuffer.writerIndex((long) (valueCount + 1) * OFFSET_WIDTH);
   }
 
   /**

vector/src/main/java/org/apache/arrow/vector/complex/ListVector.java

@@ -267,11 +267,14 @@ private void setReaderAndWriterIndex() {
     offsetBuffer.readerIndex(0);
     if (valueCount == 0) {
       validityBuffer.writerIndex(0);
-      offsetBuffer.writerIndex(0);
     } else {
       validityBuffer.writerIndex(BitVectorHelper.getValidityBufferSizeFromCount(valueCount));
-      offsetBuffer.writerIndex((valueCount + 1) * OFFSET_WIDTH);
     }
+    // IPC serializer will determine readable bytes based on `readerIndex` and `writerIndex`.
+    // Both are set to 0 means 0 bytes are written to the IPC stream which will crash IPC readers
+    // in other libraries. According to Arrow spec, we should still output the offset buffer which
+    // is [0].
+    offsetBuffer.writerIndex((long) (valueCount + 1) * OFFSET_WIDTH);
   }
 
   /**

vector/src/test/java/org/apache/arrow/vector/TestLargeListVector.java

@@ -1100,6 +1100,38 @@ public void testCopyValueSafeForExtensionType() throws Exception {
     }
   }
 
+  @Test
+  public void testNestedEmptyLargeListOffsetBuffer() {
+    // Test that nested LargeListVector properly allocates offset buffer
+    // even when nested writers are never invoked. According to Arrow spec,
+    // offset buffer must have N+1 entries. Even when N=0, it should contain [0].
+    try (LargeListVector outerList = LargeListVector.empty("outer", allocator)) {
+      // Setup LargeList<LargeList<Int>>
+      outerList.addOrGetVector(FieldType.nullable(MinorType.LARGELIST.getType()));
+      LargeListVector innerList = (LargeListVector) outerList.getDataVector();
+      innerList.addOrGetVector(FieldType.nullable(MinorType.INT.getType()));
+
+      // Allocate both outer and inner - simulates case where inner is never written to
+      outerList.allocateNew();
+      innerList.allocateNew();
+      outerList.setValueCount(0);
+      innerList.setValueCount(0);
+
+      // Get field buffers - this is what IPC serialization uses
+      List<ArrowBuf> innerBuffers = innerList.getFieldBuffers();
+
+      // Verify inner list offset buffer has at least OFFSET_WIDTH (8) bytes
+      assertTrue(
+          innerBuffers.get(1).readableBytes() >= LargeListVector.OFFSET_WIDTH,
+          "Inner LargeList offset buffer should have at least "
+              + LargeListVector.OFFSET_WIDTH
+              + " bytes for offset[0]");
+
+      // Verify offset[0] = 0
+      assertEquals(0L, innerList.getOffsetBuffer().getLong(0));
+    }
+  }
+
   private void writeIntValues(UnionLargeListWriter writer, int[] values) {
     writer.startList();
     for (int v : values) {

vector/src/test/java/org/apache/arrow/vector/TestListVector.java

@@ -1379,6 +1379,49 @@ public void testCopyValueSafeForExtensionType() throws Exception {
     }
   }
 
+  @Test
+  public void testNestedEmptyListOffsetBuffer() {
+    // Test that 3-level nested ListVector properly allocates offset buffers
+    // even when nested writers are never invoked. According to Arrow spec,
+    // offset buffer must have N+1 entries. Even when N=0, it should contain [0].
+    try (ListVector level0 = ListVector.empty("level0", allocator)) {
+      // Setup List<List<List<Int>>> - 3 levels
+      level0.addOrGetVector(FieldType.nullable(MinorType.LIST.getType()));
+      ListVector level1 = (ListVector) level0.getDataVector();
+      level1.addOrGetVector(FieldType.nullable(MinorType.LIST.getType()));
+      ListVector level2 = (ListVector) level1.getDataVector();
+      level2.addOrGetVector(FieldType.nullable(MinorType.INT.getType()));
+
+      // Allocate all levels - simulates case where nested levels are never written to
+      level0.allocateNew();
+      level1.allocateNew();
+      level2.allocateNew();
+      level0.setValueCount(0);
+      level1.setValueCount(0);
+      level2.setValueCount(0);
+
+      // Verify all levels have properly allocated offset buffers
+      List<ArrowBuf> level1Buffers = level1.getFieldBuffers();
+      List<ArrowBuf> level2Buffers = level2.getFieldBuffers();
+
+      assertTrue(
+          level1Buffers.get(1).readableBytes() >= BaseRepeatedValueVector.OFFSET_WIDTH,
+          "Level1 offset buffer should have at least "
+              + BaseRepeatedValueVector.OFFSET_WIDTH
+              + " bytes for offset[0]");
+
+      assertTrue(
+          level2Buffers.get(1).readableBytes() >= BaseRepeatedValueVector.OFFSET_WIDTH,
+          "Level2 offset buffer should have at least "
+              + BaseRepeatedValueVector.OFFSET_WIDTH
+              + " bytes for offset[0]");
+
+      // Verify offset[0] = 0 for all levels
+      assertEquals(0, level1.getOffsetBuffer().getInt(0));
+      assertEquals(0, level2.getOffsetBuffer().getInt(0));
+    }
+  }
+
   private void writeIntValues(UnionListWriter writer, int[] values) {
     writer.startList();
     for (int v : values) {