From 803fcbb9d647878431f8d0bbcb23591e281e68fb Mon Sep 17 00:00:00 2001
From: Andrew Lamb <andrew@nerdnetworks.org>
Date: Wed, 1 Jul 2026 18:00:30 -0400
Subject: [PATCH] test(parquet): cover selective list padding

---
 parquet/src/arrow/arrow_reader/read_plan.rs   |  4 ++--
 parquet/src/arrow/arrow_writer/mod.rs         | 17 ++++++++++++++++
 .../arrow/record_reader/definition_levels.rs  | 20 +++++++++++++++++++
 3 files changed, 39 insertions(+), 2 deletions(-)

diff --git a/parquet/src/arrow/arrow_reader/read_plan.rs b/parquet/src/arrow/arrow_reader/read_plan.rs
index 7ca54f961e2a..3d80526c8282 100644
--- a/parquet/src/arrow/arrow_reader/read_plan.rs
+++ b/parquet/src/arrow/arrow_reader/read_plan.rs
@@ -541,13 +541,13 @@ mod tests {
 
         let data: Vec<i32> = (0..TOTAL_ROWS as i32).collect();
         let levels = vec![0; TOTAL_ROWS];
-        let leaf = make_int32_page_reader(&data, &levels, &levels, 0, 0);
+        let leaf = make_int32_page_reader(&data, &levels, &levels, 0, 0, None);
         let struct_type = ArrowType::Struct(Fields::from(vec![Field::new(
             "c0",
             ArrowType::Int32,
             false,
         )]));
-        let struct_reader = StructArrayReader::new(struct_type, vec![leaf], 0, 0, false);
+        let struct_reader = StructArrayReader::new(struct_type, vec![leaf], 0, 0, false, None);
 
         let mut predicate = ArrowPredicateFn::new(ProjectionMask::all(), |batch| {
             Ok((0..batch.num_rows())
diff --git a/parquet/src/arrow/arrow_writer/mod.rs b/parquet/src/arrow/arrow_writer/mod.rs
index 063d5abcf1c4..d558886ba5fa 100644
--- a/parquet/src/arrow/arrow_writer/mod.rs
+++ b/parquet/src/arrow/arrow_writer/mod.rs
@@ -3860,6 +3860,23 @@ mod tests {
         one_column_roundtrip(Arc::new(list), true);
     }
 
+    #[test]
+    fn list_utf8_view_selective_padding_roundtrip() {
+        let item = Arc::new(Field::new_list_field(DataType::Utf8View, true));
+        let mut builder = ListBuilder::new(StringViewBuilder::new()).with_field(item);
+        builder.values().append_value("a");
+        builder.values().append_null();
+        builder.append(true);
+        // The null parent list covers selective padding dropping values below
+        // the list definition level while preserving the preceding item null.
+        builder.append(false);
+        // The long string covers the non-inlined Utf8View buffer path.
+        builder.values().append_value("large payload over 12 bytes");
+        builder.append(true);
+
+        one_column_roundtrip(Arc::new(builder.finish()), true);
+    }
+
     #[test]
     fn struct_single_column() {
         let a_values = Int32Array::from(vec![1, 2, 3, 4, 5, 6, 7, 8, 9, 10]);
diff --git a/parquet/src/arrow/record_reader/definition_levels.rs b/parquet/src/arrow/record_reader/definition_levels.rs
index 0720c6cdbe0d..a104c93a2c13 100644
--- a/parquet/src/arrow/record_reader/definition_levels.rs
+++ b/parquet/src/arrow/record_reader/definition_levels.rs
@@ -504,6 +504,26 @@ mod tests {
     use crate::encodings::rle::RleEncoder;
     use rand::{Rng, rng};
 
+    #[test]
+    fn test_build_validity_bitmap_unfiltered_word_chunk() {
+        // 65 levels forces the unfiltered path to process one full u64 word
+        // with append_word, plus a remainder bit.
+        let def_levels = (0..65)
+            .map(|i| if i % 3 == 0 { 2 } else { 1 })
+            .collect::<Vec<_>>();
+        let mut bitmap = BooleanBufferBuilder::new(0);
+
+        assert_eq!(
+            build_filtered_validity_bitmap(&def_levels, None, None, 2, &mut bitmap),
+            def_levels.len()
+        );
+
+        let bitmap = bitmap.finish();
+        for (idx, def) in def_levels.iter().enumerate() {
+            assert_eq!(bitmap.value(idx), *def >= 2);
+        }
+    }
+
     #[test]
     fn test_packed_decoder() {
         let mut rng = rng();