updated recommenders to use the GUID only amo curated whitelist

taar/recommenders/hybrid_recommender.py

@@ -11,19 +11,12 @@ import operator as op
 S3_BUCKET = 'telemetry-parquet'
 
 ENSEMBLE_WEIGHTS = 'taar/ensemble/ensemble_weight.json'
-CURATED_WHITELIST = 'telemetry-ml/addon_recommender/top_200_whitelist.json'
+CURATED_WHITELIST = 'telemetry-ml/addon_recommender/only_guids_top_200.json'
 
 
 class CuratedWhitelistCache:
     """
     This fetches the curated whitelist from S3.
-
-    A sample of the whitelist below :
-
-        [{'GUID': guid_string,
-          'Extension': extension_name,
-          'Copy (final)': english_description},
-        ]
     """
     def __init__(self, ctx):
         self._ctx = ctx
@@ -39,8 +32,7 @@ class CuratedWhitelistCache:
         """ Fetch a subset of randomzied GUIDs from the whitelist """
         dataset = self.get_whitelist()
         random.shuffle(dataset)
-        samples = dataset[:item_count]
-        return [s['GUID'] for s in samples]
+        return dataset[:item_count]
 
 
 class CuratedRecommender(AbstractRecommender):

taar/recommenders/recommendation_manager.py

@@ -138,7 +138,7 @@ class RecommendationManager:
 
         # The whitelist data is only used for test client IDs
         WHITELIST_S3_BUCKET = 'telemetry-parquet'
-        WHITELIST_S3_KEY = 'telemetry-ml/addon_recommender/top_200_whitelist.json'
+        WHITELIST_S3_KEY = 'telemetry-ml/addon_recommender/only_guids_top_200.json'
         self._whitelist_data = LazyJSONLoader(self._ctx, WHITELIST_S3_BUCKET, WHITELIST_S3_KEY)
 
     @schema_validate(RecommendationManagerQuerySchema)
@@ -165,7 +165,7 @@ class RecommendationManager:
             random.shuffle(data)
             samples = data[:limit]
             self.logger.info("Test ID detected [{}]".format(client_id))
-            return [(s['GUID'], 1.1) for s in samples]
+            return [(s, 1.1) for s in samples]
 
         if client_id in EMPTY_TEST_CLIENT_IDS:
             self.logger.info("Empty Test ID detected [{}]".format(client_id))

tests/test_hybrid_recommender.py

@@ -38,9 +38,7 @@ def install_no_curated_data(ctx):
 def install_mock_curated_data(ctx):
     mock_data = []
     for i in range(20):
-        mock_data.append({'GUID': str(i) * 16,
-                          'Extension': 'WebExt %d' % i,
-                          'Copy (final)': 'Copy for %d' % i})
+        mock_data.append(str(i) * 16)
 
     ctx = ctx.child()
     conn = boto3.resource('s3', region_name='us-west-2')