Merge pull request NOAA-GFDL#35 from aradhakrishnanGFDL/cfname-realm

Get standard name considering realm
aradhakrishnanGFDL · Aug 15, 2024 · f0b2cff · f0b2cff
2 parents cb007c1 + 509aae9
commit f0b2cff
Show file tree

Hide file tree

Showing 2 changed files with 21 additions and 8 deletions.
diff --git a/catalogbuilder/intakebuilder/getinfo.py b/catalogbuilder/intakebuilder/getinfo.py
@@ -242,7 +242,7 @@ def getInfoFromGlobalAtts(fname,dictInfo,filexra=None):
     dictInfo["frequency"] = frequency
     return dictInfo
 
-def getStandardName(list_variable_id):
+def getStandardName(list_variable_id,list_realm):
   '''
   Returns dict standard name for the variable in question
   ''' 
@@ -256,7 +256,9 @@ def getStandardName(list_variable_id):
             sys.exit(1)
   #search for variable and its cf name
   for variable_id in list_variable_id:
-     cfname = (df[df['GFDL_varname'] == variable_id]["standard_name"])
+   for realm in list_realm: 
+     cfname = df[(df['GFDL_varname'] == variable_id) & (realm in df['modeling_realm'])]["standard_name"]
+     #cfname = (df[df['GFDL_varname'] == variable_id]["standard_name"])
      #print(cfname,variable_id)
      list_cfname = cfname.tolist()
      if(len(list_cfname) == 0):
@@ -266,5 +268,7 @@ def getStandardName(list_variable_id):
         #print(list_cfname)
      if len(list_cfname) > 0:
        unique_cf = list(set(list_cfname))[0]
-       dictCF[variable_id] = unique_cf
+       varrealm = "{0},{1}".format(variable_id,realm)
+       dictCF[varrealm] = unique_cf
+       #print(varrealm,unique_cf)
   return (dictCF)
diff --git a/catalogbuilder/scripts/gen_intake_gfdl.py b/catalogbuilder/scripts/gen_intake_gfdl.py
@@ -100,13 +100,22 @@ def create_catalog(input_path=None, output_path=None, config=None, filter_realm=
                #If we badly need standard name, we use gfdl cmip mapping tables especially when one does not prefer the slow option. Useful for MDTF runs
                       df = pd.read_csv(os.path.abspath(csv_path), sep=",", header=0,index_col=False)
                       list_variable_id = []
-                      list_variable_id = df["variable_id"].tolist()
-                      dictVarCF = getinfo.getStandardName(list_variable_id)
+                      list_variable_id = df["variable_id"].unique().tolist()
+                      list_realm = df["realm"].unique().tolist()
+                      dictVarCF = getinfo.getStandardName(list_variable_id,list_realm)
                       #print("standard name from look-up table-", dictVarCF)
                       for k, v in dictVarCF.items():
-                         #if(df['variable_id'].eq(k)).any():
-                         df['standard_name'].loc[(df['variable_id'] == k)] = v
-                             #df['standard_name'] = v 
+                        try:
+                           var = k.split(",")[0]
+                        except ValueError:
+                           continue
+                        try:
+                           realm = k.split(",")[1]
+                        except ValueError:
+                           continue 
+                        if(var is not None) & (realm is not None):
+                            df['standard_name'].loc[(df['variable_id'] == var) & (df['realm'] == realm) ] = v
+                        #df['standard_name'].loc[(df['variable_id'] == k)] = v
     if(slow == False) & ('standard_name' in headers):
        if ((df is not None) & (len(df) != 0) ):
            with open(csv_path, 'w') as csvfile: