Add Python 3.9 support to Hive

Hive support for Python 3.9 has been removed in apache#15515 but cloudera
released new ssl 0.3.1 library version to support it and we should
be able to get Hive provider working for Python 3.9 too.

Fixes: apache#21891

airflow/providers/apache/hive/provider.yaml

@@ -36,9 +36,6 @@ versions:
 additional-dependencies:
   - apache-airflow>=2.1.0
 
-excluded-python-versions:
-  - "3.9"
-
 integrations:
   - integration-name: Apache Hive
     external-doc-url: https://hive.apache.org/

setup.py

@@ -380,7 +380,11 @@ def write_version(filename: str = os.path.join(*[my_dir, "airflow", "git_version
 ]
 hive = [
     'hmsclient>=0.1.0',
-    'pyhive[hive]>=0.6.0;python_version<"3.9"',
+    'pyhive[hive]>=0.6.0',
+    # in case of Python 3.9 sasl library needs to be installed with version higher or equal than
+    # 0.3.1 because only that version supports Python 3.9. For other Python version pyhive[hive] pulls
+    # the sasl library anyway (and there sasl library version is not relevant)
+    'sasl>=0.3.1; python_version>="3.9"',
     'thrift>=0.9.2',
     pandas_requirement,
 ]

tests/providers/apache/hive/hooks/test_hive.py

@@ -28,7 +28,6 @@
 import pytest
 from hmsclient import HMSClient
 
-from airflow import PY39
 from airflow.exceptions import AirflowException
 from airflow.models.connection import Connection
 from airflow.models.dag import DAG
@@ -55,12 +54,6 @@ def __init__(self):
         self.iterable = []
 
 
-@pytest.mark.skipif(
-    PY39,
-    reason="Hive does not run on Python 3.9 because it brings SASL via thrift-sasl."
-    " This could be removed when https://github.com/dropbox/PyHive/issues/380"
-    " is solved",
-)
 class TestHiveEnvironment(unittest.TestCase):
     def setUp(self):
         self.next_day = (DEFAULT_DATE + datetime.timedelta(days=1)).isoformat()[:10]
@@ -77,12 +70,6 @@ def setUp(self):
             self.hook = HiveMetastoreHook()
 
 
-@pytest.mark.skipif(
-    PY39,
-    reason="Hive does not run on Python 3.9 because it brings SASL via thrift-sasl."
-    " This could be removed when https://github.com/dropbox/PyHive/issues/380"
-    " is solved",
-)
 class TestHiveCliHook(unittest.TestCase):
     @mock.patch('tempfile.tempdir', '/tmp/')
     @mock.patch('tempfile._RandomNameSequence.__next__')
@@ -347,12 +334,6 @@ def test_load_df_with_data_types(self, mock_run_cli):
         assert_equal_ignore_multiple_spaces(self, mock_run_cli.call_args_list[0][0][0], query)
 
 
-@pytest.mark.skipif(
-    PY39,
-    reason="Hive does not run on Python 3.9 because it brings SASL via thrift-sasl."
-    " This could be removed when https://github.com/dropbox/PyHive/issues/380"
-    " is solved",
-)
 class TestHiveMetastoreHook(TestHiveEnvironment):
     VALID_FILTER_MAP = {'key2': 'value2'}
 
@@ -596,12 +577,6 @@ def test_drop_partition(self, get_metastore_client_mock, table_exist_mock):
         assert metastore_mock.drop_partition(self.table, db=self.database, part_vals=[DEFAULT_DATE_DS]), ret
 
 
-@pytest.mark.skipif(
-    PY39,
-    reason="Hive does not run on Python 3.9 because it brings SASL via thrift-sasl."
-    " This could be removed when https://github.com/dropbox/PyHive/issues/380"
-    " is solved",
-)
 class TestHiveServer2Hook(unittest.TestCase):
     def _upload_dataframe(self):
         df = pd.DataFrame({'a': [1, 2], 'b': [1, 2]})
@@ -857,12 +832,6 @@ def test_get_results_with_hive_conf(self):
         assert 'test_dag_run_id' in output
 
 
-@pytest.mark.skipif(
-    PY39,
-    reason="Hive does not run on Python 3.9 because it brings SASL via thrift-sasl."
-    " This could be removed when https://github.com/dropbox/PyHive/issues/380"
-    " is solved",
-)
 class TestHiveCli(unittest.TestCase):
     def setUp(self):
         self.nondefault_schema = "nondefault"

tests/providers/apache/hive/transfers/test_hive_to_mysql.py

@@ -20,9 +20,6 @@
 import unittest
 from unittest.mock import MagicMock, patch
 
-import pytest
-
-from airflow import PY39
 from airflow.providers.apache.hive.transfers.hive_to_mysql import HiveToMySqlOperator
 from airflow.utils import timezone
 from airflow.utils.operator_helpers import context_to_airflow_vars
@@ -31,12 +28,6 @@
 DEFAULT_DATE = timezone.datetime(2015, 1, 1)
 
 
-@pytest.mark.skipif(
-    PY39,
-    reason="Hive does not run on Python 3.9 because it brings SASL via thrift-sasl."
-    " This could be removed when https://github.com/dropbox/PyHive/issues/380"
-    " is solved",
-)
 class TestHiveToMySqlTransfer(TestHiveEnvironment):
     def setUp(self):
         self.kwargs = dict(

tests/providers/apache/hive/transfers/test_hive_to_samba.py

@@ -19,9 +19,6 @@
 import unittest
 from unittest.mock import MagicMock, Mock, PropertyMock, patch
 
-import pytest
-
-from airflow import PY39
 from airflow.providers.apache.hive.transfers.hive_to_samba import HiveToSambaOperator
 from airflow.providers.samba.hooks.samba import SambaHook
 from airflow.utils.operator_helpers import context_to_airflow_vars
@@ -44,12 +41,6 @@ def get_connection(self, *args):
         return self.conn
 
 
-@pytest.mark.skipif(
-    PY39,
-    reason="Hive does not run on Python 3.9 because it brings SASL via thrift-sasl."
-    " This could be removed when https://github.com/dropbox/PyHive/issues/380"
-    " is solved",
-)
 class TestHive2SambaOperator(TestHiveEnvironment):
     def setUp(self):
         self.kwargs = dict(

tests/providers/apache/hive/transfers/test_mssql_to_hive.py

@@ -23,7 +23,7 @@
 
 import pytest
 
-from airflow import PY38, PY39
+from airflow import PY38
 
 if PY38:
     MsSqlToHiveTransferOperator: None = None
@@ -36,12 +36,6 @@
     pymssql = None
 
 
-@pytest.mark.skipif(
-    PY39,
-    reason="Hive does not run on Python 3.9 because it brings SASL via thrift-sasl."
-    " This could be removed when https://github.com/dropbox/PyHive/issues/380"
-    " is solved",
-)
 @pytest.mark.skipif(PY38, reason="Mssql package not available when Python >= 3.8.")
 @pytest.mark.skipif(pymssql is None, reason='pymssql package not present')
 class TestMsSqlToHiveTransfer(unittest.TestCase):

tests/providers/apache/hive/transfers/test_mysql_to_hive.py

@@ -23,7 +23,6 @@
 
 import pytest
 
-from airflow import PY39
 from airflow.providers.apache.hive.hooks.hive import HiveCliHook
 from airflow.providers.apache.hive.transfers.mysql_to_hive import MySqlToHiveOperator
 from airflow.providers.mysql.hooks.mysql import MySqlHook
@@ -34,12 +33,6 @@
 DEFAULT_DATE_DS = DEFAULT_DATE_ISO[:10]
 
 
-@pytest.mark.skipif(
-    PY39,
-    reason="Hive does not run on Python 3.9 because it brings SASL via thrift-sasl."
-    " This could be removed when https://github.com/dropbox/PyHive/issues/380"
-    " is solved",
-)
 @pytest.mark.backend("mysql")
 class TestTransfer:
     env_vars = {