Fix dict extend validation

Author: FrancescAlted

src/blosc2/ctable.py

@@ -5699,6 +5699,8 @@ def extend(self, data: list | CTable | Any, *, validate: bool | None = None) ->
         if self.base is not None:
             raise TypeError("Cannot extend view.")
         if len(data) <= 0:
+            if isinstance(data, dict):
+                raise ValueError("No columns provided for extend().")
             return
 
         # Resolve effective validate flag: per-call override takes precedence
@@ -5720,9 +5722,25 @@ def extend(self, data: list | CTable | Any, *, validate: bool | None = None) ->
                     provided_names.add(name)
         else:
             if isinstance(data, dict):
-                provided_names = set(data) & set(current_col_names)
-                new_nrows = len(next(iter(data.values())))
-                raw_columns = {name: data[name] for name in provided_names}
+                known_names = [name for name in current_col_names if name in data]
+                if not known_names:
+                    raise ValueError("No known stored columns provided for extend().")
+                column_lengths = {}
+                for name in known_names:
+                    try:
+                        column_lengths[name] = len(data[name])
+                    except TypeError as exc:
+                        raise TypeError(f"Column {name!r} does not have a length.") from exc
+                new_nrows = column_lengths[known_names[0]]
+                mismatched = {name: n for name, n in column_lengths.items() if n != new_nrows}
+                if mismatched:
+                    details = ", ".join(f"{name}={n}" for name, n in mismatched.items())
+                    raise ValueError(
+                        f"All provided columns must have the same length; "
+                        f"expected {new_nrows}, got {details}."
+                    )
+                provided_names = set(known_names)
+                raw_columns = {name: data[name] for name in known_names}
             elif isinstance(data, np.ndarray) and data.dtype.names is not None:
                 new_nrows = len(data)
                 raw_columns = {name: data[name] for name in data.dtype.names if name in current_col_names}

tests/ctable/test_extend_delete.py

@@ -50,6 +50,40 @@ def assert_data_at_positions(table: CTable, positions: list, expected_ids: list)
 # -------------------------------------------------------------------
 
 
+def test_extend_dict_rejects_mismatched_column_lengths():
+    t = CTable(Row, expected_size=8)
+    with pytest.raises(ValueError, match="same length"):
+        t.extend(
+            {
+                "id": [1, 2],
+                "c_val": [0j],
+                "score": [10.0, 20.0],
+                "active": [True, False],
+            }
+        )
+
+
+def test_extend_dict_rejects_no_known_columns():
+    t = CTable(Row, expected_size=8)
+    with pytest.raises(ValueError, match="No known stored columns"):
+        t.extend({"unknown": [1, 2]})
+
+
+def test_extend_dict_uses_known_column_lengths():
+    t = CTable(Row, expected_size=8)
+    t.extend(
+        {
+            "unknown": [0],
+            "id": [1, 2],
+            "c_val": [0j, 1j],
+            "score": [10.0, 20.0],
+            "active": [True, False],
+        }
+    )
+    assert len(t) == 2
+    assert list(t["id"][:]) == [1, 2]
+
+
 def test_gap_fill_mask_and_positions():
     """extend and append fill from last valid position; mask is updated correctly."""
     # extend after deletions: mask and physical positions