chore: remove python2 (#310)

* ci: remove py2 ci (#280)

(cherry picked from commit 2ac583cf)

* Remove all usages of six

* makefile: remove defunct python2 test command

* setup.py: bump version to 3.0.0

This change removes python2 support. Python 3.6 or up are now required.

There are no other breaking changes in this release.
Co-authored-by: Tao Feng <fengtao04@gmail.com>

.travis.yml

 language: python
 python: 
-- '2.7'
 - '3.6'
 install:
 - pip install -r requirements.txt

Makefile

@@ -4,7 +4,8 @@ clean:
 	rm -rf dist/
 
 .PHONY: test_unit
-test_unit: test_unit2_or3_if_its_default
+test_unit:
+	python3 -bb -m pytest tests
 
 lint:
 	flake8 .
@@ -12,9 +13,3 @@ lint:
 .PHONY: test
 test: test_unit lint
 
-.PHONY: test_unit
-test_unit2_or3_if_its_default:
-	python -bb -m pytest tests/unit
-
-test_unit3:
-	python3 -bb -m pytest tests/unit

README.md

@@ -13,7 +13,7 @@ Amundsen Databuilder is a data ingestion library, which is inspired by [Apache G
 For information about Amundsen and our other services, visit the [main repository](https://github.com/lyft/amundsen#amundsen) `README.md` . Please also see our instructions for a [quick start](https://github.com/lyft/amundsen/blob/master/docs/installation.md#bootstrap-a-default-version-of-amundsen-using-docker) setup  of Amundsen with dummy data, and an [overview of the architecture](https://github.com/lyft/amundsen/blob/master/docs/architecture.md#architecture).
 
 ## Requirements
-- Python = 2.7.x or Python >= 3.6.x
+- Python >= 3.6.x
 
 ## Doc
 - https://lyft.github.io/amundsen/

databuilder/__init__.py

@@ -2,13 +2,11 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import abc
-import six
 
 from pyhocon import ConfigTree, ConfigFactory  # noqa: F401
 
 
-@six.add_metaclass(abc.ABCMeta)
-class Scoped(object):
+class Scoped(object, metaclass=abc.ABCMeta):
     _EMPTY_CONFIG = ConfigFactory.from_dict({})
     """
     An interface for class that works with scoped (nested) config.

databuilder/callback/call_back.py

@@ -3,15 +3,13 @@
 
 import abc
 import logging
-import six
 
 from typing import List, Optional  # noqa: F401
 
 LOGGER = logging.getLogger(__name__)
 
 
-@six.add_metaclass(abc.ABCMeta)
-class Callback(object):
+class Callback(object, metaclass=abc.ABCMeta):
     """
     A callback interface that expected to fire "on_success" if the operation is successful, else "on_failure" if
     operation failed.

databuilder/extractor/db2_metadata_extractor.py

@@ -2,7 +2,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import logging
-import six
 from collections import namedtuple
 
 from pyhocon import ConfigFactory, ConfigTree  # noqa: F401
@@ -60,11 +59,7 @@ class Db2MetadataExtractor(Extractor):
 
         cluster_source = "'{}'".format(self._cluster)
 
-        database = conf.get_string(Db2MetadataExtractor.DATABASE_KEY, default='db2')
-        if six.PY2 and isinstance(database, six.text_type):
-            database = database.encode('utf-8', 'ignore')
-
-        self._database = database
+        self._database = conf.get_string(Db2MetadataExtractor.DATABASE_KEY, default='db2')
 
         self.sql_stmt = Db2MetadataExtractor.SQL_STATEMENT.format(
             where_clause_suffix=conf.get_string(Db2MetadataExtractor.WHERE_CLAUSE_SUFFIX_KEY),

databuilder/extractor/mssql_metadata_extractor.py

@@ -2,7 +2,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import logging
-import six
 from collections import namedtuple
 
 from pyhocon import ConfigFactory, ConfigTree  # noqa: F401
@@ -86,13 +85,9 @@ class MSSQLMetadataExtractor(Extractor):
         else:
             cluster_source = "'{}'".format(self._cluster)
 
-        database = conf.get_string(
+        self._database = conf.get_string(
             MSSQLMetadataExtractor.DATABASE_KEY,
             default='mssql')
-        if six.PY2 and isinstance(database, six.text_type):
-            database = database.encode('utf-8', 'ignore')
-
-        self._database = database
 
         config_where_clause = conf.get_string(
             MSSQLMetadataExtractor.WHERE_CLAUSE_SUFFIX_KEY)

databuilder/extractor/mysql_metadata_extractor.py

@@ -2,7 +2,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import logging
-import six
 from collections import namedtuple
 
 from pyhocon import ConfigFactory, ConfigTree  # noqa: F401
@@ -69,11 +68,7 @@ class MysqlMetadataExtractor(Extractor):
         else:
             cluster_source = "'{}'".format(self._cluster)
 
-        database = conf.get_string(MysqlMetadataExtractor.DATABASE_KEY, default='mysql')
-        if six.PY2 and isinstance(database, six.text_type):
-            database = database.encode('utf-8', 'ignore')
-
-        self._database = database
+        self._database = conf.get_string(MysqlMetadataExtractor.DATABASE_KEY, default='mysql')
 
         self.sql_stmt = MysqlMetadataExtractor.SQL_STATEMENT.format(
             where_clause_suffix=conf.get_string(MysqlMetadataExtractor.WHERE_CLAUSE_SUFFIX_KEY),

databuilder/extractor/postgres_metadata_extractor.py

@@ -2,7 +2,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import logging
-import six
 from collections import namedtuple
 
 from pyhocon import ConfigFactory, ConfigTree  # noqa: F401
@@ -64,11 +63,7 @@ class PostgresMetadataExtractor(Extractor):
         else:
             cluster_source = "'{}'".format(self._cluster)
 
-        database = conf.get_string(PostgresMetadataExtractor.DATABASE_KEY, default='postgres')
-        if six.PY2 and isinstance(database, six.text_type):
-            database = database.encode('utf-8', 'ignore')
-
-        self._database = database
+        self._database = conf.get_string(PostgresMetadataExtractor.DATABASE_KEY, default='postgres')
 
         self.sql_stmt = PostgresMetadataExtractor.SQL_STATEMENT.format(
             where_clause_suffix=conf.get_string(PostgresMetadataExtractor.WHERE_CLAUSE_SUFFIX_KEY),

databuilder/extractor/snowflake_metadata_extractor.py

@@ -3,7 +3,6 @@
 
 
 import logging
-import six
 from collections import namedtuple
 
 from pyhocon import ConfigFactory, ConfigTree  # noqa: F401
@@ -84,10 +83,6 @@ class SnowflakeMetadataExtractor(Extractor):
         self._database = conf.get_string(SnowflakeMetadataExtractor.DATABASE_KEY)
         self._snowflake_database = conf.get_string(SnowflakeMetadataExtractor.SNOWFLAKE_DATABASE_KEY)
 
-        if six.PY2:
-            self._database = self._database.encode('utf-8', 'ignore')
-            self._snowflake_database = self._snowflake_database.encode('utf-8', 'ignore')
-
         self.sql_stmt = SnowflakeMetadataExtractor.SQL_STATEMENT.format(
             where_clause_suffix=conf.get_string(SnowflakeMetadataExtractor.WHERE_CLAUSE_SUFFIX_KEY),
             cluster_source=cluster_source,

databuilder/loader/file_system_neo4j_csv_loader.py

 # Copyright Contributors to the Amundsen project.
 # SPDX-License-Identifier: Apache-2.0
 
-import six
+import csv
 import logging
 import os
 import shutil
@@ -17,11 +17,6 @@ from databuilder.models.neo4j_csv_serde import NODE_LABEL, \
 from databuilder.models.neo4j_csv_serde import Neo4jCsvSerializable  # noqa: F401
 from databuilder.utils.closer import Closer
 
-if six.PY2:
-    import unicodecsv as csv
-else:
-    import csv
-
 
 LOGGER = logging.getLogger(__name__)
 
@@ -170,15 +165,9 @@ class FsNeo4jCSVLoader(Loader):
 
         LOGGER.info('Creating file for {}'.format(key))
 
-        if six.PY2:
-
-            file_out = open('{}/{}.csv'.format(dir_path, file_suffix), 'w')
-            writer = csv.DictWriter(file_out, fieldnames=csv_record_dict.keys(),
-                                    quoting=csv.QUOTE_NONNUMERIC, encoding='utf-8')
-        else:
-            file_out = open('{}/{}.csv'.format(dir_path, file_suffix), 'w', encoding='utf8')
-            writer = csv.DictWriter(file_out, fieldnames=csv_record_dict.keys(),
-                                    quoting=csv.QUOTE_NONNUMERIC)
+        file_out = open('{}/{}.csv'.format(dir_path, file_suffix), 'w', encoding='utf8')
+        writer = csv.DictWriter(file_out, fieldnames=csv_record_dict.keys(),
+                                quoting=csv.QUOTE_NONNUMERIC)
 
         def file_out_close():
             # type: () -> None

databuilder/models/neo4j_csv_serde.py

@@ -3,7 +3,6 @@
 
 import abc
 
-import six
 from typing import Dict, Set, Any, Union  # noqa: F401
 
 NODE_KEY = 'KEY'
@@ -24,8 +23,7 @@ LABELS = {NODE_LABEL, RELATION_START_LABEL, RELATION_END_LABEL}
 TYPES = {RELATION_TYPE, RELATION_REVERSE_TYPE}
 
 
-@six.add_metaclass(abc.ABCMeta)
-class Neo4jCsvSerializable(object):
+class Neo4jCsvSerializable(object, metaclass=abc.ABCMeta):
     """
     A Serializable abstract class asks subclass to implement next node or
     next relation in dict form so that it can be serialized to CSV file.
@@ -120,7 +118,7 @@ class Neo4jCsvSerializable(object):
         required_count = 0
         for header_col, val_col in \
                 ((header_col, val_col) for header_col, val_col
-                 in six.iteritems(val_dict) if header_col in required_set):
+                 in val_dict.items() if header_col in required_set):
             required_count += 1
 
             if header_col in LABELS:

databuilder/models/table_metadata.py

@@ -3,7 +3,6 @@
 
 import copy
 from collections import namedtuple
-from six import string_types
 
 from typing import Iterable, Any, Union, Iterator, Dict, Set  # noqa: F401
 
@@ -334,7 +333,7 @@ class TableMetadata(Neo4jCsvSerializable):
 
     @staticmethod
     def format_tags(tags):
-        if isinstance(tags, string_types):
+        if isinstance(tags, str):
             tags = list(filter(None, tags.split(',')))
         if isinstance(tags, list):
             tags = [tag.lower().strip() for tag in tags]

databuilder/publisher/neo4j_csv_publisher.py

@@ -11,7 +11,6 @@ from os import listdir
 from os.path import isfile, join
 from string import Template
 
-import six
 from neo4j import GraphDatabase, Transaction  # noqa: F401
 import neo4j
 from neo4j.exceptions import CypherError
@@ -389,7 +388,7 @@ ON MATCH SET {update_prop_body}""".format(create_prop_body=create_prop_body,
         """
         template_params = {}
         props = []
-        for k, v in six.iteritems(record_dict):
+        for k, v in record_dict.items():
             if k in excludes:
                 template_params[k] = v
                 continue
@@ -437,10 +436,7 @@ ON MATCH SET {update_prop_body}""".format(create_prop_body=create_prop_body,
             if LOGGER.isEnabledFor(logging.DEBUG):
                 LOGGER.debug('Executing statement: {} with params {}'.format(stmt, params))
 
-            if six.PY2:
-                result = tx.run(unicode(stmt, errors='ignore'), parameters=params)  # noqa
-            else:
-                result = tx.run(str(stmt).encode('utf-8', 'ignore'), parameters=params)
+            result = tx.run(str(stmt).encode('utf-8', 'ignore'), parameters=params)
             if expect_result and not result.single():
                 raise RuntimeError('Failed to executed statement: {}'.format(stmt))
 

databuilder/publisher/neo4j_preprocessor.py

@@ -4,14 +4,12 @@
 import abc
 
 import logging
-import six
 import textwrap
 
 LOGGER = logging.getLogger(__name__)
 
 
-@six.add_metaclass(abc.ABCMeta)
-class RelationPreprocessor(object):
+class RelationPreprocessor(object, metaclass=abc.ABCMeta):
     """
     A Preprocessor for relations. Prior to publish Neo4j relations, RelationPreprocessor will be used for
     pre-processing.

databuilder/rest_api/base_rest_api_query.py

@@ -4,14 +4,12 @@
 import abc
 import logging
 
-import six
 from typing import Iterable, Any, Dict, Iterator  # noqa: F401
 
 LOGGER = logging.getLogger(__name__)
 
 
-@six.add_metaclass(abc.ABCMeta)
-class BaseRestApiQuery(object):
+class BaseRestApiQuery(object, metaclass=abc.ABCMeta):
 
     @abc.abstractmethod
     def execute(self):

databuilder/rest_api/rest_api_failure_handlers.py

@@ -3,13 +3,11 @@
 
 import abc
 
-import six
 from requests.exceptions import HTTPError
 from typing import Iterable, Union, List, Dict, Any, Optional  # noqa: F401
 
 
-@six.add_metaclass(abc.ABCMeta)
-class BaseFailureHandler(object):
+class BaseFailureHandler(object, metaclass=abc.ABCMeta):
 
     @abc.abstractmethod
     def can_skip_failure(self,

databuilder/transformer/regex_str_replace_transformer.py

@@ -2,7 +2,6 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import logging
-import six
 from pyhocon import ConfigTree  # noqa: F401
 from typing import Any  # noqa: F401
 
@@ -38,13 +37,9 @@ class RegexStrReplaceTransformer(Transformer):
         else:
             val = getattr(record, self._attribute_name)
 
-        if val is None or not isinstance(val, six.string_types):
+        if val is None or not isinstance(val, str):
             return record
 
-        # Encode unicode string
-        if six.PY2:
-            val = val.encode('utf-8', 'ignore')
-
         for regex_replace_tuple in self._regex_replace_tuples:
             val = val.replace(regex_replace_tuple[0], regex_replace_tuple[1])
 

setup.py

@@ -5,7 +5,7 @@ import os
 from setuptools import setup, find_packages
 
 
-__version__ = '2.6.5'
+__version__ = '3.0.0'
 
 requirements_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), 'requirements.txt')
 with open(requirements_path) as requirements_file:
@@ -56,7 +56,7 @@ setup(
     packages=find_packages(exclude=['tests*']),
     dependency_links=[],
     install_requires=requirements,
-    python_requires='>=2.7,!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,!=3.5.*',
+    python_requires='>=3.6,!=3.0.*,!=3.1.*,!=3.2.*,!=3.3.*,!=3.4.*,!=3.5.*',
     extras_require={
         ':python_version=="2.7"': ['typing>=3.6'],  # allow typehinting PY2
         'all': all_deps,
@@ -71,7 +71,6 @@ setup(
         'druid': druid,
     },
     classifiers=[
-        'Programming Language :: Python :: 2.7',
         'Programming Language :: Python :: 3.6',
         'Programming Language :: Python :: 3.7',
     ],

tests/unit/publisher/test_elasticsearch_publisher.py

@@ -3,7 +3,6 @@
 
 import json
 from mock import MagicMock, mock_open, patch
-import six
 import unittest
 
 from pyhocon import ConfigFactory
@@ -38,10 +37,7 @@ class TestElasticsearchPublisher(unittest.TestCase):
         """
         Test Publish functionality with no data
         """
-        target = 'builtins.open'
-        if six.PY2:
-            target = '__builtin__.open'
-        with patch(target, mock_open(read_data='')) as mock_file:
+        with patch('builtins.open', mock_open(read_data='')) as mock_file:
             publisher = ElasticsearchPublisher()
             publisher.init(conf=Scoped.get_scoped_conf(conf=self.conf,
                                                        scope=publisher.get_scope()))
@@ -62,10 +58,7 @@ class TestElasticsearchPublisher(unittest.TestCase):
                                 'KEY_DOESNOT_MATTER2': 'NO_VALUE2'})
         self.mock_es_client.indices.get_alias.return_value = {}
 
-        target = 'builtins.open'
-        if six.PY2:
-            target = '__builtin__.open'
-        with patch(target, mock_open(read_data=mock_data)) as mock_file:
+        with patch('builtins.open', mock_open(read_data=mock_data)) as mock_file:
             publisher = ElasticsearchPublisher()
             publisher.init(conf=Scoped.get_scoped_conf(conf=self.conf,
                                                        scope=publisher.get_scope()))
@@ -99,10 +92,7 @@ class TestElasticsearchPublisher(unittest.TestCase):
                                 'KEY_DOESNOT_MATTER2': 'NO_VALUE2'})
         self.mock_es_client.indices.get_alias.return_value = {'test_old_index': 'DOES_NOT_MATTER'}
 
-        target = 'builtins.open'
-        if six.PY2:
-            target = '__builtin__.open'
-        with patch(target, mock_open(read_data=mock_data)) as mock_file:
+        with patch('builtins.open', mock_open(read_data=mock_data)) as mock_file:
             publisher = ElasticsearchPublisher()
             publisher.init(conf=Scoped.get_scoped_conf(conf=self.conf,
                                                        scope=publisher.get_scope()))