Updating gotm tests expected values and regen script

github-actions/gotm-fabm-ersem/regen_state_var_expected.py → github-actions/gotm-fabm-ersem/regen_expected_results.py

@@ -7,7 +7,7 @@
 import argparse
 import json
 import netCDF4 as nc
-from numpy import ndarray
+from numpy import ndarray, interp
 
 class NumpyEncoder(json.JSONEncoder):
     def default(self, obj):
@@ -36,17 +36,40 @@ def default(self, obj):
     "G2_o_deep" , "G3_c" , "ben_nit_G4n" , "H1_c" , "H2_c" , "Y2_c" , "Y3_c" ,
     "Y4_c"]
 
-data = nc.Dataset(data_path, 'r')
-expected_results = {}
-for v in state_vars:
-    if data.variables[v].ndim == 4:
-        expected_results[v] = data.variables[v][:].squeeze()[-1, :]
-    elif data.variables[v].ndim == 3:
-        expected_results[v] = float(data.variables[v][:].squeeze()[-1])
-    else:
-        raise RuntimeError
+gotm_vars_test = ["dates", "N1_p", "N3_n", "N5_s"]
 
+data_dict = {"expected": gotm_vars_test, "expected_state": state_vars}
 
-with open('data.json', 'w') as f:
-    json.dump(expected_results, f, cls=NumpyEncoder)
+for key, items in data_dict.items():
+    data = nc.Dataset(data_path, 'r')
+    expected_results = {}
+    for v in items:
+        if key == "expected":
+            if v == "dates":
+                times = data.variables['time']
+                dates = nc.num2date(times[:],
+                                    units=times.units,
+                                    calendar=times.calendar)
+                dates = [str(d).split(" ")[0] for d in dates]
+                expected_results[v] = dates
+            else:
+                depth = 0.0
+                var = data.variables[v]
+                zi = data.variables['zi'][:].squeeze()
+                z = data.variables['z'][:].squeeze()
+                var_time_series = []
+                for i in range(var.shape[0]):
+                    depth_offset = depth + zi[i, -1]
+                    var_time_series.append(interp(depth_offset, z[i, :], var[i, :].squeeze()))
+                expected_results[v] = var_time_series
+
+        elif data.variables[v].ndim == 4:
+            expected_results[v] = data.variables[v][:].squeeze()[-1, :]
+        elif data.variables[v].ndim == 3:
+            expected_results[v] = float(data.variables[v][:].squeeze()[-1])
+        else:
+            raise RuntimeError
+
+    with open(f'{key}.json', 'w') as f:
+        json.dump(expected_results, f, cls=NumpyEncoder)