Drop Python 2 support (#759)

* remove python 2 support from setup.py

* remove python 2 drop warning from init

* remove is_python_2 from feature_set_calculator

* remove use of backports.tempfile module

* remove python 2 logic from load_primitive_from_file

* remove is python 2 checks from agg primitives

* remove python2 logic from test_demo_data

* remove is_python_2 function

* remove pathlib backport

* remove funcsigs

* remove future library

* remove py27 circleci tests

* remove python 2 package dependency filters

* bump pytest to latest version

* fix tests to work with updated pytest version

* remove python 2 type check entity repr

* more repr changes

* remove is_string utils function

* update changelog

* remove unused try/except

* use exist_ok on makedirs to avoid try/catch

* remove utf-8 coding headers since it is default in python3

* remove final builtins import

.circleci/config.yml

@@ -50,19 +50,13 @@ jobs:
       - run:
           name: "Trigger premium primitives tests for python 3.5"
           command: "curl -u ${PP_K}: -d build_parameters[CIRCLE_JOB]=python-35 https://circleci.com/api/v1.1/project/github/FeatureLabs/premium-primitives/tree/master"
-      - run:
-          name: "Trigger premium primitives tests for python 2.7"
-          command: "curl -u ${PP_K}: -d build_parameters[CIRCLE_JOB]=python-27 https://circleci.com/api/v1.1/project/github/FeatureLabs/premium-primitives/tree/master"
   run_premium_primitives_tests_on_release:
     docker:
       - image: circleci/buildpack-deps:curl
     steps:
       - run:
           name: "Trigger premium primitives tests for python 3.5 release"
           command: "curl -u ${PP_K}: -d build_parameters[CIRCLE_JOB]=python-35-ft-release https://circleci.com/api/v1.1/project/github/FeatureLabs/premium-primitives/tree/master"
-      - run:
-          name: "Trigger premium primitives tests for python 2.7 release"
-          command: "curl -u ${PP_K}: -d build_parameters[CIRCLE_JOB]=python-27-ft-release https://circleci.com/api/v1.1/project/github/FeatureLabs/premium-primitives/tree/master"
 
   unit_tests:
     working_directory: ~/featuretools
@@ -176,9 +170,6 @@ workflows:
   version: 2
   test_all_python_versions:
     jobs:
-      - install_ft:
-          name: "py27 install featuretools"
-          image_tag: "2.7"
       - install_ft:
           name: "py35 install featuretools"
           image_tag: "3.5"
@@ -187,11 +178,6 @@ workflows:
           image_tag: "3.6"
       - install_ft:
           name: "py37 install featuretools"
-      - unit_tests:
-          name: "py27 unit tests"
-          image_tag: "2.7"
-          requires:
-            - "py27 install featuretools"
       - unit_tests:
           name: "py35 unit tests"
           image_tag: "3.5"
@@ -207,11 +193,6 @@ workflows:
           name: "py37 unit tests"
           requires:
             - "py37 install featuretools"
-      - lint_test:
-          name: "py27 lint test"
-          image_tag: "2.7"
-          requires:
-            - "py27 install featuretools"
       - lint_test:
           name: "py35 lint test"
           image_tag: "3.5"
@@ -230,9 +211,6 @@ workflows:
           name: "build docs"
           requires:
             - "py37 install featuretools"
-      - install_ft_complete:
-          name: "py27 install featuretools complete test"
-          image_tag: "2.7"
       - install_ft_complete:
           name: "py35 install featuretools complete test"
           image_tag: "3.5"

dev-requirements.txt

@@ -2,16 +2,13 @@
 codecov==2.0.15
 flake8==3.7.8
 autopep8==1.4.4
-ipython==7.2.0; python_version>'3.4'
-ipython==5.8.0; python_version<'3'
+ipython==7.2.0
 isort==4.3.21
 jupyter==1.0.0
-matplotlib==3.0.2; python_version>'3.4'
-matplotlib==2.2.3; python_version<'3'
+matplotlib==3.0.2
 nbconvert==5.4.1
 nbsphinx==0.4.2
-Sphinx==2.0.1; python_version>'3.4'
-Sphinx==1.8.5; python_version<'3'
+Sphinx==2.0.1
 sphinx_rtd_theme==0.4.3
 nlp_primitives>=0.2.2
 autonormalize >= 1.0.0

docs/source/changelog.rst

@@ -7,6 +7,7 @@ Changelog
     * Fixes
     * Updates
     * Changes
+        * Drop Python 2 support (:pr:`759`)
     * Documentation Changes
     * Testing Changes
 

featuretools/__init__.py

@@ -1,5 +1,4 @@
 # flake8: noqa
-from __future__ import absolute_import
 import warnings
 from .config_init import config
 from . import variable_types
@@ -34,11 +33,3 @@
         sys.modules["featuretools." + entry_point.name] = entry_point.load()
     except Exception:
         pass
-
-
-from .utils.gen_utils import is_python_2
-if is_python_2():
-    warnings.warn(
-        "The next non-bugfix release of Featuretools will not support Python 2",
-        FutureWarning
-    )

featuretools/computational_backends/calculate_feature_matrix.py

@@ -1,12 +1,9 @@
-from __future__ import division
-
 import logging
 import math
 import os
 import shutil
 import time
 import warnings
-from builtins import zip
 from datetime import datetime
 
 import cloudpickle

featuretools/computational_backends/feature_set.py

@@ -1,6 +1,5 @@
 import itertools
 import logging
-from builtins import object
 from collections import defaultdict
 
 from featuretools.entityset.relationship import RelationshipPath

featuretools/computational_backends/feature_set_calculator.py

@@ -16,7 +16,7 @@
     IdentityFeature,
     TransformFeature
 )
-from featuretools.utils import Trie, is_python_2
+from featuretools.utils import Trie
 from featuretools.utils.gen_utils import get_relationship_variable_id
 
 warnings.simplefilter('ignore', np.RankWarning)
@@ -587,9 +587,7 @@ def last_n(df):
                     # for some reason, using the string count is significantly
                     # faster than any method a primitive can return
                     # https://stackoverflow.com/questions/55731149/use-a-function-instead-of-string-in-pandas-groupby-agg
-                    if is_python_2() and func == pd.Series.count.__func__:
-                        func = "count"
-                    elif func == pd.Series.count:
+                    if func == pd.Series.count:
                         func = "count"
 
                     funcname = func

featuretools/computational_backends/utils.py

@@ -1,5 +1,3 @@
-# -*- coding: utf-8 -*-
-
 import logging
 import os
 import warnings

featuretools/demo/flight.py

@@ -1,6 +1,5 @@
 import math
 import re
-from builtins import str
 
 import pandas as pd
 from tqdm import tqdm

featuretools/demo/mock_customer.py

@@ -1,7 +1,3 @@
-from __future__ import division
-
-from builtins import range
-
 import pandas as pd
 from numpy import random
 from numpy.random import choice

featuretools/entityset/deserialize.py

@@ -1,6 +1,7 @@
 import json
 import os
 import tarfile
+import tempfile
 from pathlib import Path
 
 import boto3
@@ -10,18 +11,12 @@
 from featuretools.entityset.serialize import FORMATS
 from featuretools.utils.gen_utils import (
     check_schema_version,
-    is_python_2,
     use_s3fs_es,
     use_smartopen_es
 )
 from featuretools.utils.wrangle import _is_s3, _is_url
 from featuretools.variable_types.variable import LatLong, find_variable_types
 
-if is_python_2():
-    from backports import tempfile
-else:
-    import tempfile
-
 
 def description_to_variable(description, entity=None):
     '''Deserialize variable from variable description.

featuretools/entityset/entity.py

@@ -1,15 +1,10 @@
-# -*- coding: utf-8 -*-
-from __future__ import absolute_import, division, print_function
-
 import logging
-from builtins import range
 
 import numpy as np
 import pandas as pd
 import pandas.api.types as pdtypes
 
 from featuretools import variable_types as vtypes
-from featuretools.utils import is_string
 from featuretools.utils.entity_utils import (
     col_is_datetime,
     convert_all_variable_data,
@@ -95,10 +90,6 @@ def __repr__(self):
         repr_out += u"\n  Shape:\n    (Rows: {}, Columns: {})".format(
             shape[0], shape[1])
 
-        # encode for python 2
-        if type(repr_out) != str:
-            repr_out = repr_out.encode("utf-8")
-
         return repr_out
 
     @property
@@ -559,10 +550,10 @@ def _create_index(index, make_index, df):
 
 def _validate_entity_params(id, df, time_index):
     '''Validation checks for Entity inputs'''
-    assert is_string(id), "Entity id must be a string"
+    assert isinstance(id, str), "Entity id must be a string"
     assert len(df.columns) == len(set(df.columns)), "Duplicate column names"
     for c in df.columns:
-        if not is_string(c):
+        if not isinstance(c, str):
             raise ValueError("All column names must be strings (Column {} "
                              "is not a string)".format(c))
     if time_index is not None and time_index not in df.columns:

featuretools/entityset/entityset.py

@@ -1,6 +1,5 @@
 import copy
 import logging
-from builtins import object
 from collections import defaultdict
 
 import cloudpickle
@@ -12,7 +11,6 @@
 from featuretools.entityset import deserialize, serialize
 from featuretools.entityset.entity import Entity
 from featuretools.entityset.relationship import Relationship, RelationshipPath
-from featuretools.utils import is_string
 
 pd.options.mode.chained_assignment = None  # default='warn'
 logger = logging.getLogger('featuretools.entityset')
@@ -213,10 +211,6 @@ def __repr__(self):
                 (r._child_entity_id, r._child_variable_id,
                  r._parent_entity_id, r._parent_variable_id)
 
-        # encode for python 2
-        if type(repr_out) != str:
-            repr_out = repr_out.encode("utf-8")
-
         return repr_out
 
     def add_relationships(self, relationships):
@@ -573,7 +567,7 @@ def normalize_entity(self, base_entity_id, new_entity_id, index,
             if v == index:
                 raise ValueError("Not copying {} as both index and variable".format(v))
                 break
-        if is_string(make_time_index):
+        if isinstance(make_time_index, str):
             if make_time_index not in base_entity.df.columns:
                 raise ValueError("'make_time_index' must be a variable in the base entity")
             elif make_time_index not in additional_variables + copy_variables:

featuretools/entityset/relationship.py

@@ -39,10 +39,6 @@ def __repr__(self):
             (self._child_entity_id, self._child_variable_id,
              self._parent_entity_id, self._parent_variable_id)
 
-        # encode for python 2
-        if type(ret) != str:
-            ret = ret.encode("utf-8")
-
         return ret
 
     def __eq__(self, other):

featuretools/entityset/serialize.py

@@ -1,23 +1,14 @@
 import datetime
-import errno
 import json
 import os
 import tarfile
+import tempfile
 
 import boto3
 
-from featuretools.utils.gen_utils import (
-    is_python_2,
-    use_s3fs_es,
-    use_smartopen_es
-)
+from featuretools.utils.gen_utils import use_s3fs_es, use_smartopen_es
 from featuretools.utils.wrangle import _is_s3, _is_url
 
-if is_python_2():
-    from backports import tempfile
-else:
-    import tempfile
-
 FORMATS = ['csv', 'pickle', 'parquet']
 SCHEMA_VERSION = "1.0.0"
 
@@ -145,12 +136,7 @@ def write_data_description(entityset, path, profile_name=None, **kwargs):
         raise ValueError("Writing to URLs is not supported")
     else:
         path = os.path.abspath(path)
-        try:
-            os.makedirs(os.path.join(path, 'data'))
-        except OSError as e:
-            if e.errno != errno.EEXIST:
-                raise
-
+        os.makedirs(os.path.join(path, 'data'), exist_ok=True)
         dump_data_description(entityset, path, **kwargs)
 
 

featuretools/entityset/timedelta.py

@@ -1,10 +1,6 @@
-from __future__ import division
-
 import pandas as pd
 from dateutil.relativedelta import relativedelta
 
-from featuretools.utils import is_string
-
 
 class Timedelta(object):
     """Represents differences in time.
@@ -115,7 +111,7 @@ def get_unit_type(self):
             return relativedelta(**readable_times)
 
     def check_value(self, value, unit):
-        if is_string(value):
+        if isinstance(value, str):
             from featuretools.utils.wrangle import _check_timedelta
             td = _check_timedelta(value)
             self.times = td.times

featuretools/feature_base/feature_base.py

@@ -1,5 +1,3 @@
-from builtins import zip
-
 from featuretools import Relationship, Timedelta, primitives
 from featuretools.entityset.relationship import RelationshipPath
 from featuretools.primitives.base import (
@@ -171,13 +169,7 @@ def number_output_features(self):
         return self.primitive.number_output_features
 
     def __repr__(self):
-        ret = "<Feature: %s>" % (self.get_name())
-
-        # encode for python 2
-        if type(ret) != str:
-            ret = ret.encode("utf-8")
-
-        return ret
+        return "<Feature: %s>" % (self.get_name())
 
     def hash(self):
         return hash(self.get_name() + self.entity.id)

featuretools/primitives/base/primitive_base.py

@@ -1,5 +1,3 @@
-from __future__ import absolute_import
-
 import os
 
 import numpy as np

featuretools/primitives/base/utils.py

@@ -1,9 +1,4 @@
-try:
-    # python 3
-    from inspect import signature
-except ImportError:
-    # python 2
-    from funcsigs import signature
+from inspect import signature
 
 
 def inspect_function_args(new_class, function, uses_calc_time):