def test_get_similar_artist_candidate_set_df(self):
similar_artist_df = listenbrainz_spark.session.createDataFrame([
Row(similar_artist_credit_id=2, similar_artist_name='martinkemp', user_name='rob'),
Row(similar_artist_credit_id=2, similar_artist_name='martinkemp', user_name='vansika_1'),
])
similar_artist_candidate_set_df, similar_artist_candidate_set_df_html = candidate_sets.get_similar_artist_candidate_set(
similar_artist_df, self.recordings_df, self.users_df, self.mapped_listens_subset
)
self.assertCountEqual(['recording_id', 'user_id', 'user_name'], similar_artist_candidate_set_df.columns)
self.assertEqual(similar_artist_candidate_set_df.count(), 2)
self.assertCountEqual(
['similar_artist_credit_id', 'artist_credit_id', 'recording_mbid', 'recording_id', 'user_name', 'user_id'],
similar_artist_candidate_set_df_html.columns
)
self.assertEqual(similar_artist_candidate_set_df_html.count(), 2)
def test_get_similar_artist_candidate_set_df(self):
mapped_listens_df = utils.read_files_from_HDFS(
self.mapped_listens_path)
recordings_df = create_dataframes.get_recordings_df(
mapped_listens_df, {}, self.recordings_path)
users = create_dataframes.get_users_dataframe(mapped_listens_df, {},
self.users_path)
mapped_listens_subset = utils.read_files_from_HDFS(
self.mapped_listens_subset_path)
df = utils.create_dataframe(Row(similar_artist_credit_id=2,
similar_artist_name='martinkemp',
user_name='rob'),
schema=None)
similar_artist_df = df.union(
utils.create_dataframe(Row(similar_artist_credit_id=2,
similar_artist_name='martinkemp',
user_name='vansika_1'),
schema=None))
similar_artist_candidate_set_df, similar_artist_candidate_set_df_html = candidate_sets.get_similar_artist_candidate_set(
similar_artist_df, recordings_df, users, mapped_listens_subset)
cols = ['recording_id', 'user_id', 'user_name']
self.assertListEqual(sorted(cols),
sorted(similar_artist_candidate_set_df.columns))
self.assertEqual(similar_artist_candidate_set_df.count(), 2)
cols = [
'similar_artist_credit_id', 'similar_artist_name',
'mb_artist_credit_id', 'mb_artist_credit_mbids',
'mb_recording_mbid', 'msb_artist_credit_name_matchable',
'msb_recording_name_matchable', 'recording_id', 'user_name',
'user_id'
]
self.assertListEqual(
sorted(cols), sorted(similar_artist_candidate_set_df_html.columns))
self.assertEqual(similar_artist_candidate_set_df_html.count(), 2)