chore(pre-commit): add ty pre-commit repo

.pre-commit-config.yaml

@@ -6,17 +6,18 @@
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.12.11
+    rev: v0.15.1
     hooks:
       # Run the linter.
       - id: ruff-check
         args: [--fix]
       # Run the formatter.
       - id: ruff-format
 
-  - repo: local
+  - repo: https://github.com/allganize/ty-pre-commit
+    rev: v0.0.17
     hooks:
-      - id: type-check
+      - id: ty-check
         name: Type check
         entry: ty check .
         language: system

requirements.txt

@@ -16,3 +16,5 @@ requests>=2.32.5
 scipy>=1.17.0
 tqdm>=4.67.3
 wget>=3.2
+ty
+ruff

swvo/io/RBMDataSet/scripts/create_RBSP_line_data.py

@@ -201,8 +201,8 @@ def create_RBSP_line_data(
                     rbm_data_set_result.line_data_flux[:, e] = np.squeeze(
                         rbm_data[min_offset_instrument].interp_flux(
                             target_en_single,
-                            target_al[e],
-                            TargetType.TargetPairs,  # ty:ignore[not-subscriptable]
+                            target_al[e],  # ty:ignore[not-subscriptable]
+                            TargetType.TargetPairs,
                         )
                     )
 

swvo/io/dst/wdc.py

@@ -57,7 +57,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
         self.data_dir.mkdir(parents=True, exist_ok=True)
 
         logger.info(f"WDC Dst data directory: {self.data_dir}")
@@ -199,7 +199,7 @@ def _process_single_file(self, file_path: Path, year, month) -> pd.DataFrame:
                 if len(val) > 4:
                     val = val[:4] if not val.startswith("9999") else None
                 try:
-                    dst = float(val)
+                    dst = float(val)  # ty:ignore[invalid-argument-type]
                 except:  # noqa: E722
                     continue
                 dt = datetime(year, month, day, hour)

swvo/io/f10_7/omni.py

@@ -67,7 +67,7 @@ def read(self, start_time: datetime, end_time: datetime, download: bool = False)
         f107_df["file_name"] = data_out["file_name"]
 
         # we return it just every 24 hours
-        f107_df = f107_df.drop(f107_df[data_out.index.hour % 24 != 0].index, axis=0)  # ty: ignore[possibly-missing-attribute]
+        f107_df = f107_df.drop(f107_df[data_out.index.hour % 24 != 0].index, axis=0)  # ty: ignore[unresolved-attribute]
         f107_df = f107_df.replace(999.9, np.nan)
         f107_df = f107_df.truncate(
             before=start_time - timedelta(hours=23.9999),

swvo/io/f10_7/swpc.py

@@ -59,7 +59,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
                 raise ValueError(msg)
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
         self.data_dir.mkdir(parents=True, exist_ok=True)
 
         logger.info(f"SWPC F10.7 data directory: {self.data_dir}")
@@ -130,7 +130,7 @@ def download_and_process(self) -> None:
                     logger.debug(f"Updating {file_path}...")
 
                     existing_data = pd.read_csv(file_path, parse_dates=["date"])
-                    existing_data["date"] = pd.to_datetime(existing_data["date"]).dt.tz_localize(None)  # ty: ignore[unresolved-attribute]
+                    existing_data["date"] = pd.to_datetime(existing_data["date"]).dt.tz_localize(None)
                     combined_data = pd.concat([existing_data, year_data])
                     combined_data = combined_data.drop_duplicates(subset=["date"], keep="last")
                     combined_data = combined_data.sort_values("date")
@@ -199,7 +199,7 @@ def _process_single_file(self, file_path: Path) -> pd.DataFrame:
             skiprows=13,
             usecols=[0, 1, 2, 3],
             names=["year", "month", "day", "f107"],
-        )
+        )  # ty:ignore[no-matching-overload]
 
         data["date"] = pd.to_datetime(data[["year", "month", "day"]].assign(hour=0))
         data = data[["date", "f107"]]

swvo/io/hp/ensemble.py

@@ -59,7 +59,7 @@ def __init__(self, index: str, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
 
         logger.info(f"{self.index.upper()} Ensemble data directory: {self.data_dir}")
 
@@ -178,7 +178,7 @@ def read_with_horizon(self, start_time: datetime, end_time: datetime, horizon: N
         if end_time is not None:
             end_time = enforce_utc_timezone(end_time)
 
-        if not (0 <= horizon <= 72):  # ty: ignore[unsupported-operator]
+        if not (0 <= horizon <= 72):
             raise ValueError("Horizon must be between 0 and 72 hours")
 
         if self.index == "hp30":
@@ -187,7 +187,7 @@ def read_with_horizon(self, start_time: datetime, end_time: datetime, horizon: N
                 raise ValueError("Horizon for hp30 must be in 0.5 hour increments")
         elif self.index == "hp60":
             freq = "1h"
-            if horizon % 1 != 0:  # ty: ignore[unsupported-operator]
+            if horizon % 1 != 0:
                 raise ValueError("Horizon for hp60 must be in 1 hour increments")
 
         align_start_to_hp_hr = start_time.replace(hour=start_time.hour, minute=0, second=0, microsecond=0)

swvo/io/hp/gfz.py

@@ -61,7 +61,7 @@ def __init__(self, index: str, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
         self.data_dir.mkdir(parents=True, exist_ok=True)
         self.index_number: int = index[2:]
 
@@ -161,7 +161,7 @@ def _download(self, temporary_dir: Path, start_time: datetime, end_time: datetim
             with open(output_file, "w") as f:
                 json.dump(data, f)
 
-        except requests.exceptions.RequestException as e:  # ty: ignore[unresolved-attribute]
+        except requests.exceptions.RequestException as e:
             logger.error(f"API request failed: {e}")
             raise
 

swvo/io/kp/ensemble.py

@@ -53,7 +53,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
 
         logger.info(f"Kp Ensemble data directory: {self.data_dir}")
 

swvo/io/kp/niemegk.py

@@ -59,7 +59,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
         self.data_dir.mkdir(parents=True, exist_ok=True)
 
         logger.info(f"Kp Niemegk  data directory: {self.data_dir}")

swvo/io/kp/omni.py

@@ -62,7 +62,7 @@ def read(self, start_time: datetime, end_time: datetime, download: bool = False)
         kp_df["kp"] = data_out["kp"]
         kp_df["file_name"] = data_out["file_name"]
         # we return it just every 3 hours
-        kp_df = kp_df.drop(kp_df[data_out.index.hour % 3 != 0].index, axis=0)  # ty: ignore[possibly-missing-attribute]
+        kp_df = kp_df.drop(kp_df[data_out.index.hour % 3 != 0].index, axis=0)  # ty: ignore[unresolved-attribute]
         kp_df = kp_df.truncate(
             before=start_time - timedelta(hours=2.9999),
             after=end_time + timedelta(hours=2.9999),

swvo/io/kp/swpc.py

@@ -61,7 +61,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
         self.data_dir.mkdir(parents=True, exist_ok=True)
 
         logger.info(f"Kp SWPC  data directory: {self.data_dir}")
@@ -253,7 +253,7 @@ def _process_single_file(self, temporary_dir: Path) -> pd.DataFrame:
             lines = f.readlines()
             for line in lines:
                 if ":Issued:" in line:
-                    year = int(re.search(r"(\d{4})", line).group(1))  # ty: ignore[possibly-missing-attribute]
+                    year = int(re.search(r"(\d{4})", line).group(1))  # ty: ignore[unresolved-attribute]
                     break
 
             for i, line in enumerate(lines):
@@ -286,7 +286,7 @@ def _process_single_file(self, temporary_dir: Path) -> pd.DataFrame:
                 kp.append(kp_data[j][i])
 
         time_in = [timestamp[0]] * 24
-        df = pd.DataFrame({"t_forecast": timestamp}, index=time_in)
+        df = pd.DataFrame({"t_forecast": timestamp}, index=time_in)  # ty:ignore[invalid-argument-type]
         df["kp"] = kp
 
         df.loc[round(df["kp"] % 1, 2) == 0.67, "kp"] = round(df.loc[round(df["kp"] % 1, 2) == 0.67, "kp"]) + 2 / 3

swvo/io/omni/omni_high_res.py

@@ -55,7 +55,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
         self.data_dir.mkdir(parents=True, exist_ok=True)
 
         logger.info(f"OMNI high resolution data directory: {self.data_dir}")
@@ -296,7 +296,7 @@ def _process_single_month(self, data: list[str]) -> pd.DataFrame:
             logger.error(msg)
             raise ValueError(msg)
 
-        df = pd.DataFrame([line.split() for line in data_lines], columns=columns)
+        df = pd.DataFrame([line.split() for line in data_lines], columns=columns)  # ty:ignore[invalid-argument-type]
         df = df.apply(pd.to_numeric)
 
         df["timestamp"] = df["YYYY"].map(str).apply(lambda x: x + "-01-01 ") + df["HR"].map(str).apply(

swvo/io/omni/omni_low_res.py

@@ -114,7 +114,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
         self.data_dir.mkdir(parents=True, exist_ok=True)
 
         logger.info(f"OMNI Low Res  data directory: {self.data_dir}")

swvo/io/solar_wind/ace.py

@@ -64,7 +64,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
         self.data_dir.mkdir(parents=True, exist_ok=True)
 
         logger.info(f"ACE data directory: {self.data_dir}")
@@ -107,7 +107,7 @@ def download_and_process(self, request_time: datetime) -> None:
         logger.debug("Processing file ...")
         processed_df = self._process_single_file(temporary_dir)
 
-        unique_dates = np.unique(processed_df.index.date)  # ty: ignore[possibly-missing-attribute]
+        unique_dates = np.unique(processed_df.index.date)  # ty: ignore[unresolved-attribute]
 
         for date in unique_dates:
             file_path = self.data_dir / date.strftime("%Y/%m") / f"ACE_SW_NOWCAST_{date.strftime('%Y%m%d')}.csv"

swvo/io/solar_wind/dscovr.py

@@ -62,7 +62,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
         self.data_dir.mkdir(parents=True, exist_ok=True)
 
         logger.info(f"DSCOVR data directory: {self.data_dir}")
@@ -103,7 +103,7 @@ def download_and_process(self, request_time: datetime) -> None:
         logger.debug("Processing file ...")
         processed_df = self._process_single_file(temporary_dir)
 
-        unique_dates = np.unique(processed_df.index.date)  # ty: ignore[possibly-missing-attribute]
+        unique_dates = np.unique(processed_df.index.date)  # ty: ignore[unresolved-attribute]
 
         for date in unique_dates:
             file_path = self.data_dir / date.strftime("%Y/%m") / f"DSCOVR_SW_NOWCAST_{date.strftime('%Y%m%d')}.csv"

swvo/io/solar_wind/swift.py

@@ -58,7 +58,7 @@ def __init__(self, data_dir: Optional[Path] = None) -> None:
 
             data_dir = os.environ.get(self.ENV_VAR_NAME)  # ty: ignore[invalid-assignment]
 
-        self.data_dir: Path = Path(data_dir)
+        self.data_dir: Path = Path(data_dir)  # ty:ignore[invalid-argument-type]
 
         logger.info(f"SWIFT ensemble data directory: {self.data_dir}")
 
@@ -212,7 +212,7 @@ def _read_single_file(self, file_name, use_old_column_names=False) -> pd.DataFra
                     "uz": uz,
                     "pdyn": pdyn,
                 },
-                index=time,
+                index=time,  # ty:ignore[invalid-argument-type]
             )
         else:
             df = pd.DataFrame(
@@ -226,7 +226,7 @@ def _read_single_file(self, file_name, use_old_column_names=False) -> pd.DataFra
                     "bz_gsm": bz,
                     "pdyn": pdyn,
                 },
-                index=time,
+                index=time,  # ty:ignore[invalid-argument-type]
             )
 
         df["file_name"] = file_name

swvo/io/utils.py

@@ -214,7 +214,7 @@ def sw_mag_propagation(sw_data: pd.DataFrame) -> pd.DataFrame:
         Data frame with propagated solar wind data, indexed by time.
     """
 
-    sw_data["t"] = [t.timestamp() for t in sw_data.index.to_pydatetime()]  # ty: ignore[possibly-missing-attribute]
+    sw_data["t"] = [t.timestamp() for t in sw_data.index.to_pydatetime()]  # ty: ignore[unresolved-attribute]
     sw_data = sw_data.dropna(how="any")
 
     distance = 1.5e6