resolve MyPy type errors to satisfy CI checks

Author: mohamed-laarej

malariagen_data/anoph/frq_base.py

@@ -210,7 +210,7 @@ def plot_frequencies_heatmap(
 
         # Indexing.
         if index is None:
-            index = list(df.index.names)
+            index = [str(name) for name in df.index.names]
         df = df.reset_index().copy()
         if isinstance(index, list):
             index_col = (

malariagen_data/anoph/sample_metadata.py

@@ -39,7 +39,7 @@ def __init__(
         # data resources, and so column names and dtype need to be
         # passed in as parameters.
         self._aim_metadata_columns: Optional[List[str]] = None
-        self._aim_metadata_dtype: Dict[str, Any] = dict()
+        self._aim_metadata_dtype: Dict[str, Union[str, type, np.dtype]] = dict()
         if isinstance(aim_metadata_dtype, Mapping):
             self._aim_metadata_columns = list(aim_metadata_dtype.keys())
             self._aim_metadata_dtype.update(aim_metadata_dtype)
@@ -140,6 +140,16 @@ def _parse_general_metadata(
                 "longitude": "float64",
                 "sex_call": "object",
             }
+            # Mapping of string dtypes to actual dtypes
+            dtype_map = {
+                "object": str,
+                "int64": np.int64,
+                "float64": np.float64,
+            }
+
+            # Convert string dtypes to actual dtypes
+            dtype = {col: dtype_map.get(dtype[col], str) for col in dtype}
+
             df = pd.read_csv(io.BytesIO(data), dtype=dtype, na_values="")
 
             # Ensure all column names are lower case.

tests/anoph/test_snp_frq.py

@@ -153,7 +153,7 @@ def test_snp_effects(fixture, api: AnophelesSnpFrequencyAnalysis):
 
     # Check some values.
     assert np.all(df["contig"] == transcript["contig"])
-    position = df["position"].values
+    position = df["position"].to_numpy()
     assert np.all(position >= transcript["start"])
     assert np.all(position <= transcript["end"])
     assert np.all(position[1:] >= position[:-1])