NCHS data available at HHS, nation level

nchs_mortality/delphi_nchs_mortality/constants.py

@@ -25,7 +25,11 @@
         "prop"
 ]
 INCIDENCE_BASE = 100000
-GEO_RES = "state"
+GEO_RES = [
+    "nation",
+    "hhs",
+    "state"
+]
 
 # this is necessary as a delimiter in the f-string expressions we use to
 # construct detailed error reports

nchs_mortality/delphi_nchs_mortality/run.py

@@ -7,6 +7,7 @@
 import time
 from datetime import datetime, date, timedelta
 from typing import Dict, Any
+from itertools import product
 
 import numpy as np
 from delphi_utils import S3ArchiveDiffer, get_structured_logger
@@ -60,7 +61,7 @@ def run_module(params: Dict[str, Any]):
 
     stats = []
     df_pull = pull_nchs_mortality_data(token, test_file)
-    for metric in METRICS:
+    for metric,geo in product(METRICS,GEO_RES):
         if metric == 'percent_of_expected_deaths':
             print(metric)
             df = df_pull.copy()
@@ -71,7 +72,7 @@ def run_module(params: Dict[str, Any]):
             sensor_name = "_".join([SENSOR_NAME_MAP[metric]])
             dates = export_csv(
                 df,
-                geo_name=GEO_RES,
+                geo_name=geo,
                 export_dir=daily_export_dir,
                 start_date=datetime.strptime(export_start_date, "%Y-%m-%d"),
                 sensor=sensor_name,
@@ -92,7 +93,7 @@ def run_module(params: Dict[str, Any]):
                 sensor_name = "_".join([SENSOR_NAME_MAP[metric], sensor])
                 dates = export_csv(
                     df,
-                    geo_name=GEO_RES,
+                    geo_name=geo,
                     export_dir=daily_export_dir,
                     start_date=datetime.strptime(export_start_date, "%Y-%m-%d"),
                     sensor=sensor_name,

nchs_mortality/tests/test_run.py

@@ -36,17 +36,29 @@ def test_output_files_exist(self, run_as_module, date):
                        'deaths_pneumonia_or_flu_or_covid_incidence']
             sensors = ["num", "prop"]
 
-            expected_files = []
+            expected_files_nation = []
+            expected_files_state=[]
+            expected_files_hhs=[]
             for d in dates:
                 for metric in metrics:
                     if metric == "deaths_percent_of_expected":
-                        expected_files += ["weekly_" + d + "_state_" \
+                        expected_files_nation += ["weekly_" + d + "_nation_" \
+                                           + metric + ".csv"]
+                        expected_files_state += ["weekly_" + d + "_state_" \
+                                           + metric + ".csv"]
+                        expected_files_hhs += ["weekly_" + d + "_hhs_" \
                                            + metric + ".csv"]
                     else:
                         for sensor in sensors:
-                            expected_files += ["weekly_" + d + "_state_" \
+                            expected_files_nation += ["weekly_" + d + "_nation_" \
+                                               + metric + "_" + sensor + ".csv"]
+                            expected_files_state += ["weekly_" + d + "_state_" \
+                                               + metric + "_" + sensor + ".csv"]
+                            expected_files_hhs += ["weekly_" + d + "_hhs_" \
                                                + metric + "_" + sensor + ".csv"]
-            assert set(expected_files).issubset(set(csv_files))
+            assert set(expected_files_nation).issubset(set(csv_files))
+            assert set(expected_files_state).issubset(set(csv_files))
+            assert set(expected_files_hhs).issubset(set(csv_files))
 
     @pytest.mark.parametrize("date", ["2020-09-14", "2020-09-18"])
     def test_output_file_format(self, run_as_module, date):