fix oob in get operation of list namespace

polars/polars-arrow/src/index.rs

@@ -1,18 +1,18 @@
 pub(crate) trait IndexToUsize {
     /// Translate the negative index to an offset.
-    fn to_usize(self, length: usize) -> Option<usize>;
+    fn negative_to_usize(self, index: usize) -> Option<usize>;
 }
 
 impl IndexToUsize for i64 {
-    fn to_usize(self, length: usize) -> Option<usize> {
-        if self >= 0 && (self as usize) < length {
+    fn negative_to_usize(self, index: usize) -> Option<usize> {
+        if self >= 0 && (self as usize) < index {
             Some(self as usize)
         } else {
             let subtract = self.abs() as usize;
-            if subtract > length {
+            if subtract > index {
                 None
             } else {
-                Some(length - subtract)
+                Some(index - subtract)
             }
         }
     }

polars/polars-arrow/src/kernels/list.rs

@@ -38,10 +38,14 @@ fn sublist_get_indexes(arr: &ListArray<i64>, index: i64) -> PrimitiveArray<u32>
         let a: PrimitiveArray<u32> = iter
             .map(|&offset| {
                 let len = offset - previous;
+                // make sure that empty lists don't get accessed
+                if len == 0 {
+                    return None;
+                }
                 previous = offset;
 
                 let out = index
-                    .to_usize(len as usize)
+                    .negative_to_usize(len as usize)
                     .map(|idx| idx as u32 + cum_offset);
                 cum_offset += len as u32;
                 out

py-polars/tests/test_lists.py

@@ -119,3 +119,23 @@ def test_list_append() -> None:
 
     out_s = df["a"].arr.concat(([4, 1]))
     assert out_s[0].to_list() == [1, 2, 4, 1]
+
+
+def test_list_arr_empty() -> None:
+    df = pl.DataFrame({"cars": [[1, 2, 3], [2, 3], [4], []]})
+
+    out = df.select(
+        [
+            pl.col("cars").arr.first().alias("cars_first"),
+            pl.when(pl.col("cars").arr.first() == 2)
+            .then(1)
+            .when(pl.col("cars").arr.contains(2))
+            .then(2)
+            .otherwise(3)
+            .alias("cars_literal"),
+        ]
+    )
+    expected = pl.DataFrame(
+        {"cars_first": [1, 2, 4, None], "cars_literal": [2, 1, 3, 3]}
+    )
+    assert out.frame_equal(expected)