Remove backwards compatibility for Python <= 3.10

tests/memory_benchmark.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-# encoding: utf-8
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -17,13 +16,14 @@
 #
 # To run:
 # python tika/tests/memory_benchmark.py
+import gzip
 import os
 import zlib
-import gzip
 
-import tika.parser
 from memory_profiler import profile
 
+import tika.parser
+
 
 @profile
 def test_parser_binary():

tests/test_benchmark.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-# encoding: utf-8
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -16,11 +15,10 @@
 # limitations under the License.
 #
 # pytest --benchmark-enable --benchmark-timer=time.process_time tika/tests/test_benchmark.py
-# pytest --benchmark-enable --benchmark-timer=time.process_time tika/tests/test_benchmark.py
+import gzip
 import os
 import unittest
 import zlib
-import gzip
 from http import HTTPStatus
 
 import tika.parser

tests/test_from_file_service.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-# encoding: utf-8
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -17,12 +16,9 @@
 #
 # python -m unittest tika.tests.test_from_file_service
 
-import sys
 import unittest
-if sys.version_info >= (3, 3):
-    from unittest import mock
-else:
-    import mock
+from unittest import mock
+
 import tika.parser
 
 

tests/test_tika.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-# encoding: utf-8
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.

tests/tests_params.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# encoding: utf-8
+
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -19,14 +19,16 @@
 #https://docs.python.org/2/library/unittest.html
 #http://eli.thegreenplace.net/2011/08/02/python-unit-testing-parametrized-test-cases
 #public domain license reference: http://eli.thegreenplace.net/pages/code
- 
+
 #Run
 #python tika/tests/tests_params.py
 
 import csv
 import unittest
+
 import tika.parser
 
+
 class CreateTest(unittest.TestCase):
     "test for file types"
     def __init__(self, methodName='runTest', param1=None, param2=None):
@@ -64,17 +66,17 @@ def test_suite():
         try:
             suite.addTest(CreateTest.parameterize(RemoteTest,param1=x))
         except IOError as e:
-            print(e.strerror)    
-    return suite        
+            print(e.strerror)
+    return suite
 
 def test_url():
     with open('tika/tests/arguments/test_remote_content.csv', 'r') as csvfile:
             urlread = csv.reader(csvfile)
             for url in urlread:
                 yield url[1]
-                
+
 
 
 if __name__ == '__main__':
     suite = test_suite()
-    unittest.TextTestRunner(verbosity=2).run(suite)
+    unittest.TextTestRunner(verbosity=2).run(suite)

tests/tests_unpack.py

@@ -1,7 +1,6 @@
-# coding=utf8
-
 import unittest
 from tempfile import NamedTemporaryFile
+
 from tika import unpack
 
 

tika/__init__.py

@@ -1,4 +1,3 @@
-# encoding: utf-8
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -17,6 +16,7 @@
 __version__ = "3.1.0"
 
 from pkgutil import extend_path
+
 __path__ = extend_path(__path__, __name__)
 
 def initVM():

tika/config.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# encoding: utf-8
+
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -14,10 +14,11 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-# 
+#
 
 from .tika import getConfig
 
+
 def getParsers():
     return getConfig('parsers')[1]
 

tika/detector.py

@@ -1,5 +1,4 @@
 #!/usr/bin/env python
-# encoding: utf-8
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -14,9 +13,10 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-# 
+#
+
+from .tika import ServerEndpoint, callServer, detectType1
 
-from .tika import detectType1, callServer, ServerEndpoint
 
 def from_file(filename, config_path=None, requestOptions={}):
     '''

tika/language.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# encoding: utf-8
+
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -14,9 +14,10 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-# 
+#
+
+from .tika import ServerEndpoint, callServer, detectLang1
 
-from .tika import detectLang1, callServer, ServerEndpoint
 
 def from_file(filename, requestOptions={}):
     '''

tika/parser.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# encoding: utf-8
+
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -16,9 +16,11 @@
 # limitations under the License.
 #
 
-from .tika import parse1, callServer, ServerEndpoint
 import json
 
+from .tika import ServerEndpoint, callServer, parse1
+
+
 def from_file(filename, serverEndpoint=ServerEndpoint, service='all', xmlContent=False, headers=None, config_path=None, requestOptions={}, raw_response=False):
     '''
     Parses a file for metadata and content

tika/pdf.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# encoding: utf-8
+
 # Licensed to the Apache Software Foundation (ASF) under one or more
 # contributor license agreements.  See the NOTICE file distributed with
 # this work for additional information regarding copyright ownership.
@@ -16,10 +16,13 @@
 # limitations under the License.
 #
 
-from tika import parser
 from io import StringIO
+
 from bs4 import BeautifulSoup
 
+from tika import parser
+
+
 def text_from_pdf_pages(filename):
     pages_txt = []