diff --git a/crates/polars-mem-engine/src/executors/projection_utils.rs b/crates/polars-mem-engine/src/executors/projection_utils.rs
index 8287c923969a..6715796c5d30 100644
--- a/crates/polars-mem-engine/src/executors/projection_utils.rs
+++ b/crates/polars-mem-engine/src/executors/projection_utils.rs
@@ -291,6 +291,7 @@ pub(super) fn check_expand_literals(
             }
         }
     }
+
     // If all series are the same length it is ok. If not we can broadcast Series of length one.
     if !all_equal_len && should_broadcast {
         selected_columns = selected_columns
@@ -300,24 +301,26 @@ pub(super) fn check_expand_literals(
                 Ok(match series.len() {
                     0 if df_height == 1 => series,
                     1 => {
-                        if has_empty {
-                            polars_ensure!(df_height == 1,
-                                ComputeError: "Series length {} doesn't match the DataFrame height of {}",
-                                series.len(), df_height
-                            );
-                            series.slice(0, 0)
-                        } else if df_height == 1 {
+                         if !has_empty && df_height == 1 {
                             series
                         } else {
+                            if has_empty {
+                                polars_ensure!(df_height == 1,
+                                ShapeMismatch: "Series length {} doesn't match the DataFrame height of {}",
+                                series.len(), df_height
+                            );
+
+                            }
+
                             if verify_scalar {
                                 polars_ensure!(phys.is_scalar(),
-                                InvalidOperation: "Series: {}, length {} doesn't match the DataFrame height of {}\n\n\
+                                ShapeMismatch: "Series: {}, length {} doesn't match the DataFrame height of {}\n\n\
                                 If you want this Series to be broadcasted, ensure it is a scalar (for instance by adding '.first()').",
-                                series.name(), series.len(), df_height
+                                series.name(), series.len(), df_height *(!has_empty as usize)
                             );
 
                             }
-                            series.new_from_index(0, df_height)
+                            series.new_from_index(0, df_height * (!has_empty as usize) )
                         }
                     },
                     len if len == df_height => {
@@ -325,7 +328,7 @@ pub(super) fn check_expand_literals(
                     },
                     _ => {
                         polars_bail!(
-                        ComputeError: "Series length {} doesn't match the DataFrame height of {}",
+                        ShapeMismatch: "Series length {} doesn't match the DataFrame height of {}",
                         series.len(), df_height
                     )
                     }
diff --git a/py-polars/tests/unit/constructors/test_dataframe.py b/py-polars/tests/unit/constructors/test_dataframe.py
index 5e56630e7552..251ec5e7bce2 100644
--- a/py-polars/tests/unit/constructors/test_dataframe.py
+++ b/py-polars/tests/unit/constructors/test_dataframe.py
@@ -151,18 +151,6 @@ def test_df_init_nested_mixed_types() -> None:
     assert df.to_dicts() == [{"key": [{"value": 1.0}, {"value": 1.0}]}]
 
 
-def test_unit_and_empty_construction_15896() -> None:
-    # This is still incorrect.
-    # We should raise, but currently for len 1 dfs,
-    # we cannot tell if they come from a literal or expression.
-    assert "shape: (0, 2)" in str(
-        pl.DataFrame({"A": [0]}).select(
-            C="A",
-            A=pl.int_range("A"),  # creates empty series
-        )
-    )
-
-
 class CustomSchema(Mapping[str, Any]):
     """Dummy schema object for testing compatibility with Mapping."""
 
diff --git a/py-polars/tests/unit/dataframe/test_shape.py b/py-polars/tests/unit/dataframe/test_shape.py
new file mode 100644
index 000000000000..2409ee0c2f3f
--- /dev/null
+++ b/py-polars/tests/unit/dataframe/test_shape.py
@@ -0,0 +1,11 @@
+import pytest
+
+import polars as pl
+
+
+# TODO: remove this skip when streaming raises
+@pytest.mark.may_fail_auto_streaming
+def test_raise_invalid_shape_19108() -> None:
+    df = pl.DataFrame({"foo": [1, 2], "bar": [3, 4]})
+    with pytest.raises(pl.exceptions.ShapeError):
+        df.select(pl.col.foo.head(0), pl.col.bar.head(1))
diff --git a/py-polars/tests/unit/lazyframe/test_with_context.py b/py-polars/tests/unit/lazyframe/test_with_context.py
index 1c084c7f7c34..5a224a16c0a9 100644
--- a/py-polars/tests/unit/lazyframe/test_with_context.py
+++ b/py-polars/tests/unit/lazyframe/test_with_context.py
@@ -3,7 +3,6 @@
 import pytest
 
 import polars as pl
-from polars.exceptions import ComputeError
 from polars.testing import assert_frame_equal
 
 
@@ -19,7 +18,7 @@ def test_with_context() -> None:
 
     with pytest.deprecated_call():
         context = df_a.with_context(df_b.lazy())
-    with pytest.raises(ComputeError):
+    with pytest.raises(pl.exceptions.ShapeError):
         context.select("a", "c").collect()
 
 
diff --git a/py-polars/tests/unit/test_scalar.py b/py-polars/tests/unit/test_scalar.py
index d1f354d8e48e..6fa59a6c323d 100644
--- a/py-polars/tests/unit/test_scalar.py
+++ b/py-polars/tests/unit/test_scalar.py
@@ -11,7 +11,7 @@ def test_invalid_broadcast() -> None:
             "group": [0, 1],
         }
     )
-    with pytest.raises(pl.exceptions.InvalidOperationError):
+    with pytest.raises(pl.exceptions.ShapeError):
         df.select(pl.col("group").filter(pl.col("group") == 0), "a")