Specify a few invalid configuration options to avoid potential confusion for the users

CHANGELOG.md

@@ -6,12 +6,18 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [Unreleased]
+
+### Added
+
+- Added checks of invalid configuration combinations [!22](https://github.com/dmidk/sunflow/pull/22), @KristianHMoller
 - Added `{second}` and `{munite}` as a filename formatting variable [!15](https://github.com/dmidk/sunflow/pull/15), @JoachimKoenigslieb
-- Use `.expand_dims` instead of `.assign_coords` to make sure we have both time dimension and time coordinates when loading from files [!15](https://github.com/dmidk/sunflow/pull/15), @JoachimKoenigslieb
 - Added a `clearsky` config object which can load clear-sky data from files via `clearsky.path` or generate it with `pvlib` via the `simplified_solis` method [!15](https://github.com/dmidk/sunflow/pull/15), @JoachimKoenigslieb
-- `check_solar_elevation` now does not assume location is in Copenhagen by default [!15](https://github.com/dmidk/sunflow/pull/15), @JoachimKoenigslieb
 - Added `MIN_SOLAR_ELEVATION_DEGREES` (defaulting to 6 degrees) to configure the minimum maximum-corner solar elevation required to run [!15](https://github.com/dmidk/sunflow/pull/15), @JoachimKoenigslieb
 
+### Changed
+
+- Use `.expand_dims` instead of `.assign_coords` to make sure we have both time dimension and time coordinates when loading from files [!15](https://github.com/dmidk/sunflow/pull/15), @JoachimKoenigslieb
+- `check_solar_elevation` now does not assume location is in Copenhagen by default [!15](https://github.com/dmidk/sunflow/pull/15), @JoachimKoenigslieb
 
 ## [v1.2.0]
 

sunflow/downloaders.py

@@ -117,6 +117,11 @@ def download_current_data(
         merged_ds = subset_to_bbox(ds, bbox)
         current_time = merged_ds.time.values[0]
 
+    else:
+        raise ValueError(
+            f"Unsupported dataset_name: {dataset_name} " f"for run_mode 'download'"
+        )
+
     # Save data
     current_time_dt = current_time.astype("datetime64[s]").astype(datetime)
     filename = generate_input_filename(

sunflow/main.py

@@ -5,6 +5,7 @@
 from dataclasses import dataclass
 from datetime import datetime, timedelta, timezone
 from enum import Enum
+from typing import Any
 
 import isodate
 import numpy as np
@@ -221,6 +222,7 @@ def run_nowcast(
     domain_satellite_name: str,
     nowcast_config: NowcastConfig,
     s3_config: S3Config,
+    clearsky_config: dict[str, Any],
     full_ensemble: bool = False,
     custom_time: bool = True,
 ) -> RunResult:
@@ -236,6 +238,7 @@ def run_nowcast(
         domain_satellite_name: Domain identifier used for input filenames.
         nowcast_config: NowcastConfig object.
         s3_config: S3Config object.
+        clearsky_config: Clearsky configuration dict.
         full_ensemble: If True, save all ensemble members. If False,
             save configured ensemble statistics over ensemble members.
         custom_time: If True, skip the retry wait loop on missing data.
@@ -247,10 +250,6 @@ def run_nowcast(
     time_step_str = time_step.strftime("%Y-%m-%dT%H:%M:%SZ")
     logger.info(f"--- Running nowcast for {time_step_str} ---")
     nc_variable_names = config["nc_variable_names"].copy()
-    clearsky_config = config.get(
-        "clearsky",
-        {"method": "file", "path": config["filename_format"]},
-    )
 
     # Fetch current data (with retry loop in operational mode)
     fetch_current_data_with_retry(
@@ -349,7 +348,7 @@ def run_nowcast(
             nowcast_config.max_clearsky_fallback_days,
             clearsky_config["path"],
             config,
-            domain_nowcast,
+            domain_satellite,
             dataset_name,
             domain_satellite_name,
             nowcast_config,
@@ -553,14 +552,20 @@ def cli() -> None:
         f"Starting solarnowcasting.main (with configuration version: {model_version})..."
     )
 
-    # Load configuration
     args = parse_arguments()
-    nowcast_config = NowcastConfig.from_env(ensemble_members=args.ensemble_members)
-    s3_config = S3Config.from_env()
-
     run_mode = args.run_mode
     dataset_name = args.dataset
     domain_satellite_name = args.domain_satellite
+
+    # Load configurations
+    nowcast_config = NowcastConfig.from_env(ensemble_members=args.ensemble_members)
+    s3_config = S3Config.from_env()
+    config = yaml.safe_load(open("config.yaml"))[dataset_name]
+    clearsky_config = config.get(
+        "clearsky",
+        {"method": "file", "path": config["filename_format"]},
+    )
+
     domain_satellite = resolve_domain_bbox(
         domain_satellite_name,
         args.custom_domain_satellite,
@@ -593,8 +598,6 @@ def cli() -> None:
             f"domain_nowcast={domain_nowcast}."
         )
 
-    config = yaml.safe_load(open("config.yaml"))[dataset_name]
-
     if run_mode != "s3":
         os.makedirs(nowcast_config.nowcast_directory, exist_ok=True)
 
@@ -627,7 +630,7 @@ def cli() -> None:
             "Consider setting alpha=0.0 and beta=0.0 for a single-member run."
         )
 
-    validate_run_mode(run_mode, dataset_name)
+    validate_run_mode(run_mode, dataset_name, clearsky_config)
     validate_config(config, dataset_name)
     validate_nowcast_config(nowcast_config)
     verify_environment_variables(run_mode, dataset_name)
@@ -681,6 +684,7 @@ def cli() -> None:
                 domain_satellite_name,
                 nowcast_config,
                 s3_config,
+                clearsky_config,
                 full_ensemble=args.full_ensemble,
                 custom_time=custom_time,
             )

sunflow/validation.py

@@ -91,7 +91,9 @@ def validate_nowcast_config(nowcast_config: NowcastConfig) -> None:
         sys.exit(1)
 
 
-def validate_run_mode(run_mode: str, dataset_name: str) -> None:
+def validate_run_mode(
+    run_mode: str, dataset_name: str, clearsky_config: dict[str, Any]
+) -> None:
     """Validate that the run mode is compatible with the dataset.
 
     Exits immediately if the combination of run mode and dataset name
@@ -100,6 +102,7 @@ def validate_run_mode(run_mode: str, dataset_name: str) -> None:
     Args:
         run_mode: The requested run mode ('download', 'files', or 's3').
         dataset_name: Name of the dataset.
+        clearsky_config: Clearsky configuration dictionary.
 
     Raises:
         SystemExit: If the run mode is incompatible with the dataset.
@@ -111,6 +114,13 @@ def validate_run_mode(run_mode: str, dataset_name: str) -> None:
             "Use run_mode 'files' or 's3' instead. Exiting.\n"
         )
         sys.exit(1)
+    if run_mode == "download" and clearsky_config["method"] == "pvlib":
+        logger.error(
+            "Currently the pvlib clearsky method is not supported "
+            "for run_mode 'download'. "
+            "Use run_mode 'files' or 's3' instead. Exiting.\n"
+        )
+        sys.exit(1)
 
 
 def verify_environment_variables(run_mode: str, dataset_name: str) -> None: