Initial commit

barra2_dl/globals.py

@@ -49,10 +49,29 @@ class LatLonBBox(TypedDict):
                         "{var}_AUS-11_ERA5_historical_hres_BOM_BARRA-R2_v1_1hr_{year}{month:02d}-{year}{month:02d}.nc")
 
 # index for barra2 used to join separate files
-barra2_index = ['time', 'station', 'latitude[unit="degrees_north"]', 'longitude[unit="degrees_east"]']
+barra2_aus11__index = ['time', 'station', 'latitude[unit="degrees_north"]', 'longitude[unit="degrees_east"]']
+
+# BARRA2 wind speed variable pairs
+barra2_aus11_wind_all = [('ua10m', 'va10m', '10m[unit="m s-1"]'),
+                         ('ua20m', 'va20m', '20m[unit="m s-1"]'),
+                         ('ua30m', 'va30m', '30m[unit="m s-1"]'),
+                         ('ua50m', 'va50m', '50m[unit="m s-1"]'),
+                         ('ua70m', 'va70m', '70m[unit="m s-1"]'),
+                         ('ua100m', 'va100m', '100m[unit="m s-1"]'),
+                         ('ua150m', 'va150m', '150m[unit="m s-1"]'),
+                         ('ua200m', 'va200m', '200m[unit="m s-1"]'),
+                         ('ua250m', 'va250m', '250m[unit="m s-1"]'),
+                         ('ua300m', 'va300m', '30m[unit="m s-1"]'),
+                         ('ua400m', 'va400m', '400m[unit="m s-1"]'),
+                         ('ua500m', 'va500m', '500m[unit="m s-1"]'),
+                         ('ua600m', 'va600m', '600m[unit="m s-1"]'),
+                         ('ua700m', 'va700m', '700m[unit="m s-1"]'),
+                         ('ua850m', 'va850m', '850m[unit="m s-1"]'),
+                         ('ua925m', 'va925m', '925m[unit="m s-1"]'),
+                         ('ua1000m', 'va1000m', '1000m[unit="m s-1"]')]
 
 # set list of BARRA2 variables to download default list is eastward wind (ua*), northward wind (va*), and air temperature at 50m (ta50m)
-barra2_var_wind_all = ["ua50m", "va50m", "ua100m", "va100m", "ua150m", "va150m", "ta50m"]
+barra2_var_wind_custom = ["ua50m", "va50m", "ua100m", "va100m", "ua150m", "va150m", "ta50m"]
 
 # optional limited variables to test
 barra2_var_wind_50m = ["ua50m", "va50m", "ta50m"]

barra2_dl/helpers.py

@@ -2,10 +2,11 @@
 This module contains helper functions.
 """
 
+from pathlib import Path
 import pandas as pd
+import numpy as np
 import fnmatch
-from pathlib import Path
-from typing import List
+from typing import List, Union
 
 
 def list_months(start_datetime: str, end_datetime: str, freq: str ='MS', **kwargs) -> list:
@@ -162,3 +163,129 @@ def get_timestamp_range_list(dataframe: pd.DataFrame, timestamp_column: str) ->
     last_timestamp = dataframe[timestamp_column].iloc[-1]
 
     return [first_timestamp, last_timestamp]
+
+# todo draft function to combine csv files on index need to check and add tests
+def combine_csv_files(file_paths: list[str | Path], output_file: str | Path, index_col: str) -> None:
+    """
+    Combine multiple CSV files with a common index column into a single CSV file.
+
+    Args:
+        file_paths (list[str | Path]): A list of file paths to the CSV files to be combined.
+        output_file (str | Path): The path where the combined CSV file should be saved.
+        index_col (str): The name of the column to be used as the index for combining the CSV files.
+
+    Returns:
+        None
+    """
+    combined_df = None
+
+    for file_path in file_paths:
+        # Read the current CSV file
+        df = pd.read_csv(file_path, index_col=index_col)
+
+        if combined_df is None:
+            combined_df = df
+        else:
+            combined_df = combined_df.join(df, how='outer')
+
+    # Save the combined DataFrame to the output file
+    combined_df.to_csv(output_file)
+
+    return
+
+# todo draft function to process csvs
+def process_csvs:
+      # process barra2 variables to wind speed and direction todo split into modules
+
+    # initiate DataFrame for adding new columns
+    df_processed = df_combined
+
+    # loop through df_combined to df_processed
+    for tup in barra2_wind_speeds:
+        mask_wind_speed_h = tup[0][2:]
+        mask_ua = df_combined.columns.str.contains(tup[0])  # selects column header
+        mask_va = df_combined.columns.str.contains(tup[1])  # selects column header
+
+        if np.any(mask_ua == True) and np.any(mask_va == True):
+            df_processed_ua = df_combined.loc[:, mask_ua]  # selects mask
+            df_processed_va = df_combined.loc[:, mask_va]  # selects mask
+
+            print('Converted: ' + tup.__str__())
+
+            df_processed_v = pd.DataFrame(np.sqrt(df_processed_ua.iloc[:, 0] ** 2 + df_processed_va.iloc[:, 0] ** 2))
+            df_processed_v.columns = ['v' + mask_wind_speed_h + '[unit="m s-1"]']
+
+            df_processed_phi_met = pd.DataFrame()
+
+            for index, row in df_combined.iterrows():
+                if (df_processed_ua.iloc[index, 0] == 0) and (df_processed_va.iloc[index, 0] == 0):
+                    df_processed_phi_met.loc[index, 'v' + mask_wind_speed_h + '_' + 'phi_met[unit="degrees"]'] = 0.0
+                else:
+                    df_processed_phi_met.loc[index, 'v' + mask_wind_speed_h + '_' + 'phi_met[unit="degrees"]'] = (
+                        np.mod(180 + np.rad2deg(
+                            np.arctan2(df_processed_ua.iloc[index, 0], df_processed_va.iloc[index, 0])), 360))
+
+            # Merge the current variable DataFrame with the combined DataFrame
+            df_processed = df_processed.join(df_processed_v)
+            df_processed = df_processed.join(df_processed_phi_met)
+
+    # export combined to csv
+    df_processed.to_csv(
+        os.path.join(output_dir,
+                     f"{output_filename_prefix}_processed_{start_date_time.strftime("%Y%m%d")}_{end_date_time.strftime("%Y%m%d")}.csv"))
+
+    return
+
+
+# todo add tests
+def calculate_wind_speed(ua: Union[float, int, List[float], List[int]], va: Union[float, int, List[float], List[int]]) -> Union[float, List[float]]:
+    """
+    Convert wind components ua and va to wind speed v.
+    Args:
+        ua (Union[float, int, List[float], List[int]]): The u-component of the wind.
+        va (Union[float, int, List[float], List[int]]): The v-component of the wind.
+    Returns:
+        float or List[float]: The calculated wind speed.
+    Raises:
+        ValueError: If the input types do not match or if they are neither List[float] nor float.
+    """
+    if isinstance(ua, (float, int)) and isinstance(va, (float, int)):
+        if ua == 0 and va == 0:
+            return 0.0
+        return np.sqrt(ua ** 2 + va ** 2)
+    elif isinstance(ua, list) and isinstance(va, list):
+        if not all(isinstance(num, (float, int)) for num in ua + va):
+            raise ValueError("All elements in both lists must be either float or int.")
+        if len(ua) != len(va):
+            raise ValueError("Both lists must be of the same length.")
+        return [0.0 if u == 0 and v == 0 else np.sqrt(u ** 2 + v ** 2) for u, v in zip(ua, va)]
+    else:
+        raise ValueError("Both arguments must be either both float/int or both lists of float/int.")
+
+
+# todo add tests
+def calculate_wind_direction(ua: Union[float, int, List[float], List[int]], va: Union[float, int, List[float], List[int]]) -> Union[float, List[float]]:
+    """
+    Convert wind components ua and va to wind direction phi.
+    Args:
+        ua (Union[float, int, List[float], List[int]]): The u-component of the wind.
+        va (Union[float, int, List[float], List[int]]): The v-component of the wind.
+    Returns:
+        float or List[float]: The calculated wind speed.
+    Raises:
+        ValueError: If the input types and incorrect or do not match or if lists of different lengths are provided.
+    """
+
+    if isinstance(ua, (float, int)) and isinstance(va, (float, int)):
+        if ua == 0 and va == 0:
+            return 0.0
+        return np.mod(180 + np.rad2deg(np.arctan2(ua, va)), 360)
+    elif isinstance(ua, list) and isinstance(va, list):
+        if not all(isinstance(num, (float, int)) for num in ua + va):
+            raise ValueError("All elements in both lists must be either float or int.")
+        if len(ua) != len(va):
+            raise ValueError("Both lists must be of the same length.")
+        return [0.0 if u == 0 and v == 0 else np.mod(180 + np.rad2deg(np.arctan2(u, v)), 360) for u, v in zip(ua, va)]
+    else:
+        raise ValueError("Both arguments must be either both float/int or both lists of float/int.")
+