Merge pull request #732 from zwicker-group/key_retrieval

Refactor key retrieval in various modules for consistency and clarity

Author: david-zwicker

pde/fields/base.py

@@ -244,7 +244,7 @@ def from_file(cls, filename: str) -> FieldBase:
 
             elif len(fp) == 1:
                 # a single field is stored in the data
-                dataset = fp[list(fp.keys())[0]]  # retrieve only dataset
+                dataset = fp[list(fp)[0]]  # retrieve only dataset
                 obj = cls._from_hdf_dataset(dataset)  # type: ignore
 
             else:

pde/grids/boundaries/axes.py

@@ -36,7 +36,7 @@
     BoundaryPairData, Sequence[BoundaryPairData], Callable, "BoundariesBase"
 ]
 
-BC_LOCAL_KEYS = ["type", "value"] + list(BCBase._conditions.keys())
+BC_LOCAL_KEYS = ["type", "value"] + list(BCBase._conditions)
 
 
 def _is_local_bc_data(data: dict[str, Any]) -> bool:
@@ -215,7 +215,7 @@ def _parse_from_dict(
 
         # warn if some keys were left over
         if data:
-            _logger.warning("Didn't use BC data from %s", list(data.keys()))
+            _logger.warning("Didn't use BC data from %s", list(data))
         # find boundary conditions that have not been specified
         bcs_unspecified = []
         for ax, bc_ax in enumerate(bc_data):

pde/grids/boundaries/axis.py

@@ -283,7 +283,7 @@ def from_data(cls, grid: GridBase, axis: int, data, rank: int = 0) -> BoundaryPa
                 d_high = data_copy.pop("high")
                 high = BCBase.from_data(grid, axis, upper=True, data=d_high, rank=rank)
                 if data_copy:
-                    raise BCDataError(f"Data items {data_copy.keys()} were not used.")
+                    raise BCDataError(f"Data items {list(data_copy)} were not used.")
             else:
                 # one condition for both sides
                 low = BCBase.from_data(grid, axis, upper=False, data=data, rank=rank)

pde/grids/boundaries/local.py

@@ -474,9 +474,7 @@ def from_dict(
             )
 
         else:
-            raise BCDataError(
-                f"Boundary conditions `{str(list(data.keys()))}` are not supported."
-            )
+            raise BCDataError(f"Boundary conditions `{list(data)}` are not supported.")
 
     @classmethod
     def from_data(

pde/grids/cartesian.py

@@ -161,7 +161,7 @@ def from_state(cls, state: dict[str, Any]) -> CartesianGrid:  # type: ignore
             periodic=state_copy.pop("periodic"),
         )
         if state_copy:
-            raise ValueError(f"State items {state_copy.keys()} were not used")
+            raise ValueError(f"State items {list(state_copy)} were not used")
         return obj
 
     @classmethod
@@ -504,7 +504,7 @@ def from_state(cls, state: dict[str, Any]) -> UnitGrid:  # type: ignore
         state_copy = state.copy()
         obj = cls(shape=state_copy.pop("shape"), periodic=state_copy.pop("periodic"))
         if state_copy:
-            raise ValueError(f"State items {state_copy.keys()} were not used")
+            raise ValueError(f"State items {list(state_copy)} were not used")
         return obj
 
     def to_cartesian(self) -> CartesianGrid:

pde/grids/cylindrical.py

@@ -156,7 +156,7 @@ def from_state(cls, state: dict[str, Any]) -> CylindricalSymGrid:  # type: ignor
             periodic_z=state_copy.pop("periodic_z"),
         )
         if state_copy:
-            raise ValueError(f"State items {state_copy.keys()} were not used")
+            raise ValueError(f"State items {list(state_copy)} were not used")
         return obj
 
     @classmethod

pde/grids/spherical.py

@@ -116,7 +116,7 @@ def from_state(cls, state: dict[str, Any]) -> SphericalSymGridBase:  # type: ign
         state_copy = state.copy()
         obj = cls(radius=state_copy.pop("radius"), shape=state_copy.pop("shape"))
         if state_copy:
-            raise ValueError(f"State items {state_copy.keys()} were not used")
+            raise ValueError(f"State items {list(state_copy)} were not used")
         return obj
 
     @classmethod

pde/solvers/base.py

@@ -111,7 +111,7 @@ def from_name(cls, name: str, pde: PDEBase, **kwargs) -> SolverBase:
             # solver was not registered
             solvers = (
                 f"'{solver}'"
-                for solver in sorted(cls._subclasses.keys())
+                for solver in sorted(cls._subclasses)
                 if not solver.endswith("Solver")
             )
             raise ValueError(
@@ -124,7 +124,7 @@ def from_name(cls, name: str, pde: PDEBase, **kwargs) -> SolverBase:
     @classproperty
     def registered_solvers(cls) -> list[str]:
         """list of str: the names of the registered solvers"""
-        return sorted(cls._subclasses.keys())
+        return sorted(cls._subclasses)
 
     @property
     def _compiled(self) -> bool:

pde/storage/base.py

@@ -187,7 +187,7 @@ def grid(self) -> GridBase | None:
                 else:
                     self._logger.warning(
                         "`grid` attribute was not stored. Available attributes: %s",
-                        ", ".join(sorted(attrs.keys())),
+                        ", ".join(sorted(attrs)),
                     )
 
             else:

pde/tools/spectral.py

@@ -96,7 +96,7 @@ def noise_corr() -> NumericArray:
     return noise_corr
 
 
-CorrelationType = Literal["none", "gaussian", "power law", "cosine"]
+CorrelationType = Literal["none", "delta", "gaussian", "power law", "cosine"]
 
 
 def make_correlated_noise(
@@ -164,7 +164,7 @@ def make_correlated_noise(
     """
     rng = np.random.default_rng(rng)
 
-    if correlation == "none":
+    if correlation == "none" or correlation == "delta":
         # no correlation
         corr_spectrum = None
 
@@ -204,7 +204,7 @@ def corr_spectrum(k2s):
         raise ValueError(f"Unknown correlation `{correlation}`")
 
     if kwargs:
-        _logger.warning("Unused arguments: %s", kwargs.keys())
+        _logger.warning("Unused arguments: %s", list(kwargs))
 
     if corr_spectrum is None:
         # fast case of uncorrelated white noise