feat(misc): modularize kwargs filter

cryoswath/l2.py

@@ -1,5 +1,4 @@
 import geopandas as gpd
-import inspect
 from multiprocessing import Pool
 import numpy as np
 import os
@@ -200,12 +199,8 @@ def process_track(idx, reprocess, l2_paths, save_or_return, data_path, current_s
                 cs_full_file_names = kwargs.pop("cs_full_file_names")
             else:
                 cs_full_file_names = load_cs_full_file_names(update="no")
-        # filter l1b_data kwargs
-        params = inspect.signature(l1b.l1b_data).parameters
-        l1b_kwargs = {k: v for k, v in kwargs.items() if k in params}
-        # filter to_l2 kwargs
-        params = inspect.signature(l1b.l1b_data.to_l2).parameters
-        to_l2_kwargs = {k: v for k, v in kwargs.items() if k in params and k != "swath_or_poca"}
+        l1b_kwargs = filter_kwargs(l1b.l1b_data, kwargs)
+        to_l2_kwargs = filter_kwargs(l1b.l1b_data.to_l2, kwargs, blacklist=["swath_or_poca"])
         swath_poca_tuple = l1b.l1b_data.from_id(cs_time_to_id(idx), **l1b_kwargs)\
                                        .to_l2(swath_or_poca="both", **to_l2_kwargs)
         if save_or_return != "return":

cryoswath/l3.py

@@ -43,8 +43,10 @@ def build_dataset(region_of_interest: str|shapely.Polygon,
     # I believe passing loading l2 data to the function prevents copying
     # on .drop. an alternative would be to define l2_data nonlocal
     # within the gridding function
-    l3_data =  med_mad_cnt_grid(l2.from_id(cs_tracks.index), start_datetime=start_datetime, end_datetime=end_datetime,
-                                aggregation_period=aggregation_period, timestep=timestep, spatial_res_meter=spatial_res_meter)
+    l3_data =  med_mad_cnt_grid(l2.from_id(cs_tracks.index, **filter_kwargs(l2.from_id, kwargs)),
+                                start_datetime=start_datetime, end_datetime=end_datetime,
+                                aggregation_period=aggregation_period, timestep=timestep,
+                                spatial_res_meter=spatial_res_meter)
     l3_data.to_netcdf(build_path(region_of_interest, timestep, spatial_res_meter, aggregation_period))
     return l3_data
 __all__.append("build_dataset")

cryoswath/misc.py

@@ -3,6 +3,7 @@
 import ftplib
 import geopandas as gpd
 import glob
+import inspect
 import numpy as np
 import os
 import pandas as pd
@@ -135,6 +136,23 @@ def convert_all_esri_to_feather(dir_path: str = None) -> None:
 #     return local_filename
 
 
+# ! make recursive
+def filter_kwargs(func: callable,
+                  kwargs: dict, *,
+                  blacklist: list[str] = None,
+                  whitelist: list[str] = None,
+                  ) -> dict:
+    def ensure_list(tmp_list):
+        if tmp_list is None: return []
+        elif isinstance(tmp_list, str): return [tmp_list]
+        else: return tmp_list
+    blacklist = ensure_list(blacklist)
+    whitelist = ensure_list(whitelist)
+    params = inspect.signature(func).parameters
+    return {k: v for k, v in kwargs.items() if (k in params and k not in blacklist) or k in whitelist}
+__all__.append("filter_kwargs")
+
+
 def find_region_id(location: any, scope: str = "o2") -> str:
     if isinstance(location, gpd.GeoDataFrame):
         location = location.geometry