[AIRFLOW-4201] Replace unicode strings by normal strings (apache#5026)

airflow/_vendor/nvd3/NVD3Chart.py

@@ -145,9 +145,9 @@ def __init__(self, **kwargs):
         self.htmlheader = ''
         #: Place holder for the graph (the HTML div)
         #: Written by ``buildcontainer``
-        self.container = u''
+        self.container = ''
         #: Header for javascript code
-        self.containerheader = u''
+        self.containerheader = ''
         # CDN http://cdnjs.com/libraries/nvd3/ needs to make sure it's up to
         # date
         self.header_css = [

airflow/example_dags/tutorial.py

@@ -50,7 +50,7 @@
     # 'on_failure_callback': some_function,
     # 'on_success_callback': some_other_function,
     # 'on_retry_callback': another_function,
-    # 'trigger_rule': u'all_success'
+    # 'trigger_rule': 'all_success'
 }
 
 dag = DAG(

dev/airflow-license

@@ -30,7 +30,7 @@ _licenses = {'MIT': ['Permission is hereby granted free of charge', 'The above c
 
 def get_notices():
     license_file = open("../LICENSE", "r")
-    regex = ur"\((.+?)\) (.+?) \((http.+?)\)"
+    regex = r"\((.+?)\) (.+?) \((http.+?)\)"
 
     return list(filter(None, [re.findall(regex, line) for line in license_file]))
 

dev/airflow-pr

@@ -313,7 +313,7 @@ def merge_pr(pr_num, target_ref, title, body, pr_repo_desc, local):
         if primary_author == "":
             primary_author = distinct_authors[0]
 
-        merge_message_flags.append(u'--author="{}"'.format(primary_author))
+        merge_message_flags.append('--author="{}"'.format(primary_author))
 
     else:
         # This will mark the PR as merged

docs/conf.py

@@ -141,8 +141,8 @@
 master_doc = 'index'
 
 # General information about the project.
-project = u'Airflow'
-# copyright = u''
+project = 'Airflow'
+# copyright = ''
 
 # The version info for the project you're documenting, acts as replacement for
 # |version| and |release|, also used in various other places throughout the
@@ -344,8 +344,8 @@
 # (source start file, target name, title,
 #  author, documentclass [howto, manual, or own class]).
 latex_documents = [
-    ('index', 'Airflow.tex', u'Airflow Documentation',
-     u'Apache Airflow', 'manual'),
+    ('index', 'Airflow.tex', 'Airflow Documentation',
+     'Apache Airflow', 'manual'),
 ]
 
 # The name of an image file (relative to this directory) to place at the top of
@@ -374,8 +374,8 @@
 # One entry per manual page. List of tuples
 # (source start file, name, description, authors, manual section).
 man_pages = [
-    ('index', 'airflow', u'Airflow Documentation',
-     [u'Apache Airflow'], 1)
+    ('index', 'airflow', 'Airflow Documentation',
+     ['Apache Airflow'], 1)
 ]
 
 # If true, show URL addresses after external links.
@@ -388,8 +388,8 @@
 # (source start file, target name, title, author,
 #  dir menu entry, description, category)
 texinfo_documents = [(
-    'index', 'Airflow', u'Airflow Documentation',
-    u'Apache Airflow', 'Airflow',
+    'index', 'Airflow', 'Airflow Documentation',
+    'Apache Airflow', 'Airflow',
     'Airflow is a system to programmaticaly author, schedule and monitor data pipelines.',
     'Miscellaneous'
 ), ]

docs/howto/write-logs.rst

@@ -132,7 +132,7 @@ example:
 
   *** Reading remote log from gs://<bucket where logs should be persisted>/example_bash_operator/run_this_last/2017-10-03T00:00:00/16.log.
   [2017-10-03 21:57:50,056] {cli.py:377} INFO - Running on host chrisr-00532
-  [2017-10-03 21:57:50,093] {base_task_runner.py:115} INFO - Running: ['bash', '-c', u'airflow run example_bash_operator run_this_last 2017-10-03T00:00:00 --job_id 47 --raw -sd DAGS_FOLDER/example_dags/example_bash_operator.py']
+  [2017-10-03 21:57:50,093] {base_task_runner.py:115} INFO - Running: ['bash', '-c', 'airflow run example_bash_operator run_this_last 2017-10-03T00:00:00 --job_id 47 --raw -sd DAGS_FOLDER/example_dags/example_bash_operator.py']
   [2017-10-03 21:57:51,264] {base_task_runner.py:98} INFO - Subtask: [2017-10-03 21:57:51,263] {__init__.py:45} INFO - Using executor SequentialExecutor
   [2017-10-03 21:57:51,306] {base_task_runner.py:98} INFO - Subtask: [2017-10-03 21:57:51,306] {models.py:186} INFO - Filling up the DagBag from /airflow/dags/example_dags/example_bash_operator.py
 

tests/contrib/hooks/test_sqoop_hook.py

@@ -98,11 +98,11 @@ def setUp(self):
     @patch('subprocess.Popen')
     def test_popen(self, mock_popen):
         # Given
-        mock_popen.return_value.stdout = StringIO(u'stdout')
-        mock_popen.return_value.stderr = StringIO(u'stderr')
+        mock_popen.return_value.stdout = StringIO('stdout')
+        mock_popen.return_value.stderr = StringIO('stderr')
         mock_popen.return_value.returncode = 0
         mock_popen.return_value.communicate.return_value = \
-            [StringIO(u'stdout\nstdout'), StringIO(u'stderr\nstderr')]
+            [StringIO('stdout\nstdout'), StringIO('stderr\nstderr')]
 
         # When
         hook = SqoopHook(conn_id='sqoop_test')

tests/contrib/operators/test_dataproc_operator.py

@@ -370,7 +370,7 @@ def test_cluster_name_log_sub(self):
                 setattr(dataproc_task, 'cluster_name', rendered)
                 with self.assertRaises(TypeError):
                     dataproc_task.execute(None)
-                mock_info.assert_called_with('Creating cluster: %s', u'smoke-cluster-testnodash')
+                mock_info.assert_called_with('Creating cluster: %s', 'smoke-cluster-testnodash')
 
     def test_build_cluster_data_internal_ip_only_without_subnetwork(self):
 
@@ -457,7 +457,7 @@ def test_cluster_name_log_sub(self):
                 setattr(dataproc_task, 'cluster_name', rendered)
                 with self.assertRaises(TypeError):
                     dataproc_task.execute(None)
-                mock_info.assert_called_with('Scaling cluster: %s', u'smoke-cluster-testnodash')
+                mock_info.assert_called_with('Scaling cluster: %s', 'smoke-cluster-testnodash')
 
 
 class DataprocClusterDeleteOperatorTest(unittest.TestCase):
@@ -518,7 +518,7 @@ def test_cluster_name_log_sub(self):
                 with self.assertRaises(TypeError):
                     dataproc_task.execute(None)
                 mock_info.assert_called_with('Deleting cluster: %s',
-                                             u'smoke-cluster-testnodash')
+                                             'smoke-cluster-testnodash')
 
 
 class DataProcHadoopOperatorTest(unittest.TestCase):

tests/contrib/operators/test_mongo_to_s3_operator.py

@@ -76,7 +76,7 @@ def test_render_template(self):
         ti = TaskInstance(self.mock_operator, DEFAULT_DATE)
         ti.render_templates()
 
-        expected_rendered_template = {'$lt': u'2017-01-01T00:00:00+00:00Z'}
+        expected_rendered_template = {'$lt': '2017-01-01T00:00:00+00:00Z'}
 
         self.assertDictEqual(
             expected_rendered_template,

tests/contrib/operators/test_spark_submit_operator.py

@@ -154,13 +154,13 @@ def test_render_template(self):
         ti.render_templates()
 
         # Then
-        expected_application_args = [u'-f', 'foo',
-                                     u'--bar', 'bar',
-                                     u'--start', (DEFAULT_DATE - timedelta(days=1))
+        expected_application_args = ['-f', 'foo',
+                                     '--bar', 'bar',
+                                     '--start', (DEFAULT_DATE - timedelta(days=1))
                                      .strftime("%Y-%m-%d"),
-                                     u'--end', DEFAULT_DATE.strftime("%Y-%m-%d"),
-                                     u'--with-spaces',
-                                     u'args should keep embdedded spaces',
+                                     '--end', DEFAULT_DATE.strftime("%Y-%m-%d"),
+                                     '--with-spaces',
+                                     'args should keep embdedded spaces',
                                      ]
         expected_name = "spark_submit_job"
         self.assertListEqual(expected_application_args,

tests/contrib/utils/test_sendgrid.py

@@ -36,12 +36,12 @@ def setUp(self):
         self.cc = ['[email protected]', '[email protected]']
         self.bcc = ['[email protected]', '[email protected]']
         self.expected_mail_data = {
-            'content': [{'type': u'text/html', 'value': self.html_content}],
+            'content': [{'type': 'text/html', 'value': self.html_content}],
             'personalizations': [
                 {'cc': [{'email': '[email protected]'}, {'email': '[email protected]'}],
                  'to': [{'email': '[email protected]'}, {'email': '[email protected]'}],
                  'bcc': [{'email': '[email protected]'}, {'email': '[email protected]'}]}],
-            'from': {'email': u'[email protected]'},
+            'from': {'email': '[email protected]'},
             'subject': 'sendgrid-send-email unit test',
             'mail_settings': {},
         }

tests/core.py

@@ -448,7 +448,7 @@ def test_bash_operator(self):
     def test_bash_operator_multi_byte_output(self):
         t = BashOperator(
             task_id='test_multi_byte_bash_operator',
-            bash_command=u"echo \u2600",
+            bash_command="echo \u2600",
             dag=self.dag,
             output_encoding='utf-8')
         t.run(start_date=DEFAULT_DATE, end_date=DEFAULT_DATE, ignore_ti_state=True)
@@ -612,7 +612,7 @@ def test_template_with_json_variable_as_value(self):
 
         def verify_templated_field(context):
             self.assertEqual(context['ti'].task.some_templated_field,
-                             u'{"foo": "bar"}')
+                             '{"foo": "bar"}')
 
         t = OperatorSubclass(
             task_id='test_complex_template',
@@ -967,10 +967,9 @@ def test_run_command(self):
 
         cmd = 'import sys; {0}; sys.stdout.flush()'.format(write)
 
-        self.assertEqual(run_command("python -c '{0}'".format(cmd)),
-                         u'\u1000foo')
+        self.assertEqual(run_command("python -c '{0}'".format(cmd)), '\u1000foo')
 
-        self.assertEqual(run_command('echo "foo bar"'), u'foo bar\n')
+        self.assertEqual(run_command('echo "foo bar"'), 'foo bar\n')
         self.assertRaises(AirflowConfigException, run_command, 'bash -c "exit 1"')
 
     def test_trigger_dagrun_with_execution_date(self):
@@ -1972,123 +1971,123 @@ def ls(self, path, include_toplevel=False):
             return []
         elif path[0] == '/datadirectory/datafile':
             return [{
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 420,
                 'file_type': 'f',
                 'access_time': 1481122343796,
                 'block_replication': 3,
                 'modification_time': 1481122343862,
                 'length': 0,
                 'blocksize': 134217728,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/datafile'
             }]
         elif path[0] == '/datadirectory/empty_directory' and include_toplevel:
             return [{
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 493,
                 'file_type': 'd',
                 'access_time': 0,
                 'block_replication': 0,
                 'modification_time': 1481132141540,
                 'length': 0,
                 'blocksize': 0,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/empty_directory'
             }]
         elif path[0] == '/datadirectory/not_empty_directory' and include_toplevel:
             return [{
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 493,
                 'file_type': 'd',
                 'access_time': 0,
                 'block_replication': 0,
                 'modification_time': 1481132141540,
                 'length': 0,
                 'blocksize': 0,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/empty_directory'
             }, {
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 420,
                 'file_type': 'f',
                 'access_time': 1481122343796,
                 'block_replication': 3,
                 'modification_time': 1481122343862,
                 'length': 0,
                 'blocksize': 134217728,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/not_empty_directory/test_file'
             }]
         elif path[0] == '/datadirectory/not_empty_directory':
             return [{
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 420,
                 'file_type': 'f',
                 'access_time': 1481122343796,
                 'block_replication': 3,
                 'modification_time': 1481122343862,
                 'length': 0,
                 'blocksize': 134217728,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/not_empty_directory/test_file'
             }]
         elif path[0] == '/datadirectory/not_existing_file_or_directory':
             raise FakeSnakeBiteClientException
         elif path[0] == '/datadirectory/regex_dir':
             return [{
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 420,
                 'file_type': 'f',
                 'access_time': 1481122343796,
                 'block_replication': 3,
                 'modification_time': 1481122343862, 'length': 12582912,
                 'blocksize': 134217728,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/regex_dir/test1file'
             }, {
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 420,
                 'file_type': 'f',
                 'access_time': 1481122343796,
                 'block_replication': 3,
                 'modification_time': 1481122343862,
                 'length': 12582912,
                 'blocksize': 134217728,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/regex_dir/test2file'
             }, {
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 420,
                 'file_type': 'f',
                 'access_time': 1481122343796,
                 'block_replication': 3,
                 'modification_time': 1481122343862,
                 'length': 12582912,
                 'blocksize': 134217728,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/regex_dir/test3file'
             }, {
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 420,
                 'file_type': 'f',
                 'access_time': 1481122343796,
                 'block_replication': 3,
                 'modification_time': 1481122343862,
                 'length': 12582912,
                 'blocksize': 134217728,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/regex_dir/copying_file_1.txt._COPYING_'
             }, {
-                'group': u'supergroup',
+                'group': 'supergroup',
                 'permission': 420,
                 'file_type': 'f',
                 'access_time': 1481122343796,
                 'block_replication': 3,
                 'modification_time': 1481122343862,
                 'length': 12582912,
                 'blocksize': 134217728,
-                'owner': u'hdfs',
+                'owner': 'hdfs',
                 'path': '/datadirectory/regex_dir/copying_file_3.txt.sftp'
             }]
         else:
@@ -2286,8 +2285,8 @@ def test_send_smtp(self, mock_send_mime):
         self.assertEqual('subject', msg['Subject'])
         self.assertEqual(configuration.conf.get('smtp', 'SMTP_MAIL_FROM'), msg['From'])
         self.assertEqual(2, len(msg.get_payload()))
-        filename = u'attachment; filename="' + os.path.basename(attachment.name) + '"'
-        self.assertEqual(filename, msg.get_payload()[-1].get(u'Content-Disposition'))
+        filename = 'attachment; filename="' + os.path.basename(attachment.name) + '"'
+        self.assertEqual(filename, msg.get_payload()[-1].get('Content-Disposition'))
         mimeapp = MIMEApplication('attachment')
         self.assertEqual(mimeapp.get_payload(), msg.get_payload()[-1].get_payload())
 
@@ -2314,8 +2313,8 @@ def test_send_bcc_smtp(self, mock_send_mime):
         self.assertEqual('subject', msg['Subject'])
         self.assertEqual(configuration.conf.get('smtp', 'SMTP_MAIL_FROM'), msg['From'])
         self.assertEqual(2, len(msg.get_payload()))
-        self.assertEqual(u'attachment; filename="' + os.path.basename(attachment.name) + '"',
-                         msg.get_payload()[-1].get(u'Content-Disposition'))
+        self.assertEqual('attachment; filename="' + os.path.basename(attachment.name) + '"',
+                         msg.get_payload()[-1].get('Content-Disposition'))
         mimeapp = MIMEApplication('attachment')
         self.assertEqual(mimeapp.get_payload(), msg.get_payload()[-1].get_payload())
 

tests/hooks/test_hive_hook.py

@@ -167,7 +167,7 @@ def test_load_df(self, mock_to_csv, mock_load_file):
         assert mock_load_file.call_count == 1
         kwargs = mock_load_file.call_args[1]
         self.assertEqual(kwargs["delimiter"], delimiter)
-        self.assertEqual(kwargs["field_dict"], {"c": u"STRING"})
+        self.assertEqual(kwargs["field_dict"], {"c": "STRING"})
         self.assertTrue(isinstance(kwargs["field_dict"], OrderedDict))
         self.assertEqual(kwargs["table"], table)