sekikn commented on issue #6399: [AIRFLOW-5730] Enable get_pandas_df on PinotDbApiHook URL: https://github.com/apache/airflow/pull/6399#issuecomment-546126938 Updated the PR. The CI failure seems to be a transient one. The newly added test works on my local environment, as follows: ``` $ git diff upstream/master diff --git a/airflow/contrib/hooks/pinot_hook.py b/airflow/contrib/hooks/pinot_hook.py index e617f8e9b..0864b3584 100644 --- a/airflow/contrib/hooks/pinot_hook.py +++ b/airflow/contrib/hooks/pinot_hook.py @@ -90,8 +90,5 @@ class PinotDbApiHook(DbApiHook): def set_autocommit(self, conn, autocommit): raise NotImplementedError() - def get_pandas_df(self, sql, parameters=None): - raise NotImplementedError() - def insert_rows(self, table, rows, target_fields=None, commit_every=1000): raise NotImplementedError() diff --git a/tests/contrib/hooks/test_pinot_hook.py b/tests/contrib/hooks/test_pinot_hook.py index 72cee01ea..489af15aa 100644 --- a/tests/contrib/hooks/test_pinot_hook.py +++ b/tests/contrib/hooks/test_pinot_hook.py @@ -34,6 +34,7 @@ class TestPinotDbApiHook(unittest.TestCase): self.conn.conn_type = 'http' self.conn.extra_dejson = {'endpoint': 'pql'} self.cur = mock.MagicMock() + self.conn.cursor.return_value = self.cur self.conn.__enter__.return_value = self.cur self.conn.__exit__.return_value = None @@ -74,3 +75,14 @@ class TestPinotDbApiHook(unittest.TestCase): result_sets = [('row1',), ('row2',)] self.cur.fetchone.return_value = result_sets[0] self.assertEqual(result_sets[0], self.db_hook().get_first(statement)) + + def test_get_pandas_df(self): + statement = 'SQL' + column = 'col' + result_sets = [('row1',), ('row2',)] + self.cur.description = [(column,)] + self.cur.fetchall.return_value = result_sets + df = self.db_hook().get_pandas_df(statement) + self.assertEqual(column, df.columns[0]) + for i in range(len(result_sets)): # pylint: disable=consider-using-enumerate + self.assertEqual(result_sets[i][0], df.values.tolist()[i][0]) $ ./run-tests tests.contrib.hooks.test_pinot_hook:TestPinotDbApiHook.test_get_pandas_df AIRFLOW__CORE__SQL_ALCHEMY_CONN not set - using default Airflow home: /home/sekikn Airflow root: /home/sekikn/repos/airflow Home of the user: /home/sekikn Skipping initializing of the DB as it was initialized already You can re-initialize the database by adding --with-db-init flag when running tests KRB5_KTNAME variable is empty - no kerberos intialisation Starting the tests with arguments: tests.contrib.hooks.test_pinot_hook:TestPinotDbApiHook.test_get_pandas_df . ---------------------------------------------------------------------- Ran 1 test in 0.289s OK [2019-10-25 07:20:36,090] {settings.py:200} DEBUG - Disposing DB connection pool (PID 3587) ```
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
