Merge pull request #314 from chinandrew/hhs-support

Add HHS support and tests

Author: krivard

src/acquisition/covidcast/csv_importer.py

@@ -27,7 +27,7 @@ class CsvImporter:
   PATTERN_ISSUE_DIR = re.compile(r'^.*/([^/]*)/issue_(\d{8})$')
 
   # set of allowed resolutions (aka "geo_type")
-  GEOGRAPHIC_RESOLUTIONS = {'county', 'hrr', 'msa', 'dma', 'state', 'nation'}
+  GEOGRAPHIC_RESOLUTIONS = {'county', 'hrr', 'msa', 'dma', 'state', 'hhs', 'nation'}
 
   # set of required CSV columns
   REQUIRED_COLUMNS = {'geo_id', 'val', 'se', 'sample_size'}
@@ -209,7 +209,7 @@ def extract_and_check_row(row, geo_type):
       # geo_id was `None`
       return (None, 'geo_id')
 
-    if geo_type in ('hrr', 'msa', 'dma'):
+    if geo_type in ('hrr', 'msa', 'dma', 'hhs'):
       # these particular ids are prone to be written as ints -- and floats
       try:
         geo_id = str(CsvImporter.floaty_int(geo_id))
@@ -239,6 +239,10 @@ def extract_and_check_row(row, geo_type):
       if len(geo_id) != 2 or not 'aa' <= geo_id <= 'zz':
         return (None, 'geo_id')
 
+    elif geo_type == 'hhs':
+      if not 1 <= int(geo_id) <= 10:
+        return (None, 'geo_id')
+
     elif geo_type == 'nation':
       # geo_id is lowercase
       if len(geo_id) != 2 or not 'aa' <= geo_id <= 'zz':

tests/acquisition/covidcast/test_csv_importer.py

@@ -45,6 +45,8 @@ def test_find_csv_files(self):
       path_prefix + 'ght/20200408_state_rawsearch.csv',
       # valid national
       path_prefix + 'valid/20200408_nation_sig.csv',
+      # valid hhs
+      path_prefix + 'valid/20200408_hhs_sig.csv',
       # invalid
       path_prefix + 'invalid/hello_world.csv',
       # invalid day
@@ -68,10 +70,11 @@ def test_find_csv_files(self):
       (glob_paths[0], ('fb_survey', 'cli', 'week', 'county', 202015, expected_issue_week, delta_epiweeks(202015, expected_issue_week))),
       (glob_paths[1], ('ght', 'rawsearch', 'day', 'state', time_value_day, expected_issue_day, (date.today() - date(year=time_value_day // 10000, month=(time_value_day // 100) % 100, day=time_value_day % 100)).days)),
       (glob_paths[2], ('valid', 'sig', 'day', 'nation', time_value_day, expected_issue_day, (date.today() - date(year=time_value_day // 10000, month=(time_value_day // 100) % 100, day=time_value_day % 100)).days)),
-      (glob_paths[3], None),
+      (glob_paths[3], ('valid', 'sig', 'day', 'hhs', time_value_day, expected_issue_day, (date.today() - date(year=time_value_day // 10000, month=(time_value_day // 100) % 100, day=time_value_day % 100)).days)),
       (glob_paths[4], None),
       (glob_paths[5], None),
       (glob_paths[6], None),
+      (glob_paths[7], None),
     ])
     self.assertEqual(found, expected)
 
@@ -137,6 +140,7 @@ def make_row(
       (make_row(geo_type='state', geo_id='48'), 'geo_id'),
       (make_row(geo_type='state', geo_id='iowa'), 'geo_id'),
       (make_row(geo_type='nation', geo_id='0000'), 'geo_id'),
+      (make_row(geo_type='hhs', geo_id='0'), 'geo_id'),
       (make_row(geo_type='province', geo_id='ab'), 'geo_type'),
       (make_row(se='-1'), 'se'),
       (make_row(geo_type=None), 'geo_type'),