Wave: Revert cdip reader and test changes

These changes were a reaction to failing tests, but it is likely that
the failing tests were not caused by this code specifically, but pandas
3.0 updates.

This reverts these changes back to the develop branch as they are likely
unnecessary for the tests to pass when pandas is pinned below 3.0

Author: simmsa

mhkit/tests/wave/io/test_cdip.py

@@ -7,6 +7,7 @@
 import pytz
 import os
 
+
 testdir = dirname(abspath(__file__))
 datadir = normpath(join(testdir, "..", "..", "..", "..", "examples", "data", "wave"))
 
@@ -128,19 +129,19 @@ def test_request_parse_workflow_multiyear(self):
         expected_index0 = datetime(year1, 1, 1)
         expected_index_final = datetime(year2, 12, 31)
 
-        wave1D = data["wave"]
-        self.assertEqual(wave1D.index[0].floor("D").to_pydatetime(), expected_index0)
+        wave1D = data["data"]["wave"]
+        self.assertEqual(wave1D.index[0].floor("d").to_pydatetime(), expected_index0)
 
         self.assertEqual(
-            wave1D.index[-1].floor("D").to_pydatetime(), expected_index_final
+            wave1D.index[-1].floor("d").to_pydatetime(), expected_index_final
         )
 
-        for key, wave2D in data["wave2D"].items():
+        for key, wave2D in data["data"]["wave2D"].items():
             self.assertEqual(
-                wave2D.index[0].floor("D").to_pydatetime(), expected_index0
+                wave2D.index[0].floor("d").to_pydatetime(), expected_index0
             )
             self.assertEqual(
-                wave2D.index[-1].floor("D").to_pydatetime(), expected_index_final
+                wave2D.index[-1].floor("d").to_pydatetime(), expected_index_final
             )
 
     def test_plot_boxplot(self):
@@ -158,7 +159,7 @@ def test_plot_boxplot(self):
         )
 
         plt.figure()
-        wave.graphics.plot_boxplot(data["wave"]["waveHs"])
+        wave.graphics.plot_boxplot(data["data"]["wave"]["waveHs"])
         plt.savefig(filename, format="png")
         plt.close()
 
@@ -181,9 +182,9 @@ def test_plot_compendium(self):
 
         plt.figure()
         wave.graphics.plot_compendium(
-            data["wave"]["waveHs"],
-            data["wave"]["waveTp"],
-            data["wave"]["waveDp"],
+            data["data"]["wave"]["waveHs"],
+            data["data"]["wave"]["waveTp"],
+            data["data"]["wave"]["waveDp"],
         )
         plt.savefig(filename, format="png")
         plt.close()

mhkit/wave/io/cdip.py

@@ -310,72 +310,88 @@ def request_parse_workflow(
         nc.variables["metaStationName"][:].compressed().tobytes().decode("utf-8")
     )
 
-    if years is not None:
-        start_date = []
-        end_date = []
+    multiyear = False
+    if years:
         if isinstance(years, int):
-            years = [years]
-        for year in years:
-            start_date.append(datetime.datetime(year, 1, 1, tzinfo=pytz.UTC))
-            end_date.append(datetime.datetime(year + 1, 1, 1, tzinfo=pytz.UTC))
-    else:
-        start_date = [start_date]
-        end_date = [end_date]
-
-    data = {"metadata": {}}
-    multiyear_data = {}
-    for start, end in zip(start_date, end_date):
-        # Check the cache for each individual year
-        hash_params = f"{station_number}-{parameters}-{start}-{end}"
-        year_data, _, _ = handle_caching(
+            start_date = datetime.datetime(years, 1, 1, tzinfo=pytz.UTC)
+            end_date = datetime.datetime(years + 1, 1, 1, tzinfo=pytz.UTC)
+        elif isinstance(years, list):
+            if len(years) == 1:
+                start_date = datetime.datetime(years[0], 1, 1, tzinfo=pytz.UTC)
+                end_date = datetime.datetime(years[0] + 1, 1, 1, tzinfo=pytz.UTC)
+            else:
+                multiyear = True
+    if not multiyear:
+        # Check the cache first
+        hash_params = f"{station_number}-{parameters}-{start_date}-{end_date}"
+        data, _, _ = handle_caching(
             hash_params,
             cache_dir,
             cache_content={"data": None, "metadata": None, "write_json": None},
         )
-        if year_data is None:
-            new_data = get_netcdf_variables(
+
+        if data is None:
+            data = get_netcdf_variables(
                 nc,
-                start_date=start,
-                end_date=end,
+                start_date=start_date,
+                end_date=end_date,
                 parameters=parameters,
                 all_2D_variables=all_2D_variables,
                 silent=silent,
             )
-            year_data = new_data["data"]
-            # If waveFrequency is in the metadata, move it to the top level of metadata
-            if "wave" in new_data["metadata"]:
-                if "waveFrequency" in new_data["metadata"]["wave"]:
-                    new_data["metadata"]["waveFrequency"] = new_data["metadata"][
-                        "wave"
-                    ]["waveFrequency"]
-                new_data["metadata"].pop("wave")
-            # Cache the individual year's data
             handle_caching(
                 hash_params,
                 cache_dir,
-                cache_content={
-                    "data": year_data,
-                    "metadata": new_data["metadata"],
-                    "write_json": None,
-                },
+                cache_content={"data": data, "metadata": None, "write_json": None},
             )
-        year = start.year
-        multiyear_data[year] = year_data
-
-    for data_key in year_data.keys():
-        if data_key.endswith("2D"):
-            data[data_key] = {}
-            for data_key2D in year_data[data_key].keys():
-                data_list = []
-                for year in multiyear_data.keys():
-                    data2D = multiyear_data[year][data_key][data_key2D]
-                    data_list.append(data2D)
-                data[data_key][data_key2D] = pd.concat(data_list)
-        else:
-            data_list = [
-                multiyear_data[year][data_key] for year in multiyear_data.keys()
-            ]
-            data[data_key] = pd.concat(data_list)
+
+    else:
+        data = {"data": {}, "metadata": {}}
+        multiyear_data = {}
+        for year in years:
+            start_date = datetime.datetime(year, 1, 1, tzinfo=pytz.UTC)
+            end_date = datetime.datetime(year + 1, 1, 1, tzinfo=pytz.UTC)
+
+            # Check the cache for each individual year
+            hash_params = f"{station_number}-{parameters}-{start_date}-{end_date}"
+            year_data, _, _ = handle_caching(
+                hash_params,
+                cache_dir,
+                cache_content={"data": None, "metadata": None, "write_json": None},
+            )
+            if year_data is None:
+                year_data = get_netcdf_variables(
+                    nc,
+                    start_date=start_date,
+                    end_date=end_date,
+                    parameters=parameters,
+                    all_2D_variables=all_2D_variables,
+                    silent=silent,
+                )
+                # Cache the individual year's data
+                handle_caching(
+                    hash_params,
+                    cache_dir,
+                    cache_content={
+                        "data": year_data,
+                        "metadata": None,
+                        "write_json": None,
+                    },
+                )
+            multiyear_data[year] = year_data["data"]
+
+        for data_key in year_data["data"].keys():
+            if data_key.endswith("2D"):
+                data["data"][data_key] = {}
+                for data_key2D in year_data["data"][data_key].keys():
+                    data_list = []
+                    for year in years:
+                        data2D = multiyear_data[year][data_key][data_key2D]
+                        data_list.append(data2D)
+                    data["data"][data_key][data_key2D] = pd.concat(data_list)
+            else:
+                data_list = [multiyear_data[year][data_key] for year in years]
+                data["data"][data_key] = pd.concat(data_list)
 
     if buoy_name:
         try: