use unexpected key's ids as suffix; drop UT sex group handling

Author: nmdefries

src/acquisition/flusurv/api.py

@@ -202,9 +202,7 @@ def get_data(self, location):
 
         # extract
         print("[reformatting flusurv result...]")
-        data_out = self._add_sex_breakdowns_ut(
-            self._group_by_epiweek(data_in), location
-        )
+        data_out = self._group_by_epiweek(data_in), location
 
         # return
         print(f"[successfully fetched data for {location}]")
@@ -325,6 +323,8 @@ def _groupid_to_name(self, ageid, sexid, raceid, fluid):
             raise ValueError("Expect at least three of four group ids to be 0")
         if (ageid, sexid, raceid, fluid).count(0) == 4:
             group = "overall"
+        # In all cases, if id is not available as a key in the dict, use the
+        # raw id as the name suffix
         elif ageid != 0:
             if ageid == 6:
                 # Ageid of 6 used to be used for the "overall" category.
@@ -333,23 +333,13 @@ def _groupid_to_name(self, ageid, sexid, raceid, fluid):
                 #  has gone wrong.
                 raise ValueError("Ageid cannot be 6; please check for changes in the API")
             else:
-                age_group = self.metadata.id_to_group["Age"][ageid]
+                age_group = self.metadata.id_to_group["Age"].get(ageid, ageid)
             group = "age_" + age_group
         elif sexid != 0:
-            group = "sex_" + self.metadata.id_to_group["Sex"][sexid]
+            group = "sex_" + self.metadata.id_to_group["Sex"].get(sexid, ageid)
         elif raceid != 0:
-            group = "race_" + self.metadata.id_to_group["Race"][raceid]
+            group = "race_" + self.metadata.id_to_group["Race"].get(raceid, ageid)
         elif fluid != 0:
-            group = "flu_" + self.metadata.id_to_group["Flutype"][fluid]
+            group = "flu_" + self.metadata.id_to_group["Flutype"].get(fluid, ageid)
 
         return "rate_" + group
-
-    def _add_sex_breakdowns_ut(self, data, location):
-        # UT doesn't have sex breakdowns available at least for 2022-23. Fill
-        # in to avoid downstream errors.
-        if location == "UT":
-            for epiweek in data.keys():
-                for group in SEX_GROUPS:
-                    if group not in data[epiweek].keys():
-                        data[epiweek][group] = None
-        return(data)

src/acquisition/flusurv/constants.py

@@ -89,11 +89,11 @@
 )
 SEX_GROUPS = (
     "rate_sex_male",
-    "rate_sex_female"
+    "rate_sex_female",
 )
 FLU_GROUPS = (
     "rate_flu_a",
-    "rate_flu_b"
+    "rate_flu_b",
 )
 EXPECTED_GROUPS = HISTORICAL_GROUPS + NEW_AGE_GROUPS + RACE_GROUPS + SEX_GROUPS + FLU_GROUPS
 

tests/acquisition/flusurv/test_flusurv.py

@@ -357,6 +357,8 @@ def test_groupids_to_name(self):
             (0, 0, 0, 1),
             (13, 0, 0, 0),
             (97, 0, 0, 0),
+            (999, 0, 0, 0),
+            (0, 0, 111, 0),
         )
         expected_list = [
             "rate_age_0",
@@ -368,30 +370,17 @@ def test_groupids_to_name(self):
             "rate_flu_b",
             "rate_flu_a",
             "rate_age_0tlt1",
-            "rate_age_97",
+            "rate_age_lt18",
+            "rate_age_999",
+            "rate_race_111",
         ]
 
         for (ageid, sexid, raceid, fluid), expected in zip(ids, expected_list):
             self.assertEqual(api_fetcher._groupid_to_name(ageid, sexid, raceid, fluid), expected)
 
         with self.assertRaisesRegex(ValueError, "Ageid cannot be 6"):
-            api_fetcher._groupid_to_name(6, 0, 0)
+            api_fetcher._groupid_to_name(6, 0, 0, 0)
         with self.assertRaisesRegex(ValueError, "Expect at least three of four group ids to be 0"):
             api_fetcher._groupid_to_name(1, 1, 0, 0)
             api_fetcher._groupid_to_name(0, 1, 1, 0)
             api_fetcher._groupid_to_name(1, 1, 1, 1)
-
-    def test_groupids_to_name(self):
-        input_data = api_fetcher._group_by_epiweek(metadata_result)
-        self.assertEqual(api_fetcher._add_sex_breakdowns_ut(input_data, "network_all"), by_epiweek_example_data)
-
-        ut_expected = by_epiweek_example_data.copy()
-        ut_expected[201014]["rate_sex_male"] = None
-        ut_expected[200940]["rate_sex_female"] = None
-        ut_expected[200940]["rate_sex_male"] = None
-        ut_expected[201011]["rate_sex_female"] = None
-        ut_expected[201011]["rate_sex_male"] = None
-        ut_expected[201008]["rate_sex_female"] = None
-        ut_expected[201008]["rate_sex_male"] = None
-
-        self.assertEqual(api_fetcher._add_sex_breakdowns_ut(input_data, "UT"), ut_expected)