SQLAlchemy v1.4.x support (sqlalchemy-redshift#221)

.travis.yml

@@ -3,10 +3,11 @@ sudo: false
 language: python
 python:
   - "2.7"
-  - "3.4"
   - "3.5"
   - "3.6"
   - "3.7"
+  - "3.8"
+  - "3.9"
 
 env:
   global:

CHANGES.rst

@@ -1,7 +1,7 @@
 0.8.3 (unreleased)
 ------------------
 
-- Nothing changed yet.
+- SQLAlchemy 1.4.x support
 
 
 0.8.2 (2021-01-08)

sqlalchemy_redshift/dialect.py

@@ -659,13 +659,16 @@ def _get_table_or_view_names(self, relkind, connection, schema=None, **kw):
     def _get_column_info(self, *args, **kwargs):
         kw = kwargs.copy()
         encode = kw.pop('encode', None)
-        if sa_version < Version('1.2.0'):
-            # SQLAlchemy 1.2.0 introduced the 'comment' param
-            del kw['comment']
         if sa_version >= Version('1.3.16'):
             # SQLAlchemy 1.3.16 introduced generated columns,
             # not supported in redshift
             kw['generated'] = ''
+
+        if sa_version < Version('1.4.0') and 'identity' in kw:
+            del kw['identity']
+        elif sa_version >= Version('1.4.0') and 'identity' not in kw:
+            kw['identity'] = None
+
         column_info = super(RedshiftDialect, self)._get_column_info(
             *args,
             **kw
@@ -744,7 +747,7 @@ def _get_all_relation_info(self, connection, **kw):
     @reflection.cache
     def _get_all_column_info(self, connection, **kw):
         all_columns = defaultdict(list)
-        with connection.contextual_connect() as cc:
+        with connection.connect() as cc:
             result = cc.execute("""
             SELECT
               n.nspname as "schema",
@@ -918,17 +921,26 @@ def visit_delete_stmt(element, compiler, **kwargs):
     #   the tables in the using clause are sorted in the order in
     #   which they first appear in the where clause.
     delete_stmt_table = compiler.process(element.table, asfrom=True, **kwargs)
-    whereclause_tuple = element.get_children()
-    if whereclause_tuple:
-        usingclause_tables = []
-        whereclause = ' WHERE {clause}'.format(
-            clause=compiler.process(*whereclause_tuple, **kwargs)
-        )
 
+    if sa_version >= Version('1.4.0'):
+        if element.whereclause is not None:
+            clause = compiler.process(element.whereclause, **kwargs)
+            if clause:
+                whereclause = ' WHERE {clause}'.format(clause=clause)
+    else:
+        whereclause_tuple = element.get_children()
+        if whereclause_tuple:
+            whereclause = ' WHERE {clause}'.format(
+                clause=compiler.process(*whereclause_tuple, **kwargs)
+            )
+
+    if whereclause:
+        usingclause_tables = []
         whereclause_columns = gen_columns_from_children(element)
         for col in whereclause_columns:
             table = compiler.process(col.table, asfrom=True, **kwargs)
-            if table != delete_stmt_table and table not in usingclause_tables:
+            if table != delete_stmt_table and \
+                    table not in usingclause_tables:
                 usingclause_tables.append(table)
         if usingclause_tables:
             usingclause = ' USING {clause}'.format(

tests/conftest.py

@@ -8,7 +8,7 @@
 
 try:
     from urllib import parse as urlparse
-except:
+except ImportError:
     import urlparse
 
 import requests
@@ -31,6 +31,7 @@ def database_name_generator():
             count=i,
         )
 
+
 database_name = functools.partial(next, database_name_generator())
 
 
@@ -50,7 +51,10 @@ def _database(self):
             conn.execute('CREATE DATABASE {db_name}'.format(db_name=db_name))
 
         dburl = copy.deepcopy(self.engine.url)
-        dburl.database = db_name
+        try:
+            dburl.database = db_name
+        except AttributeError:
+            dburl = dburl.set(database=db_name)
 
         try:
             yield db.EngineDefinition(

tests/rs_sqla_test_utils/models.py

@@ -1,12 +1,16 @@
 import sqlalchemy as sa
 
 from sqlalchemy import event
+from sqlalchemy import DDL
 from sqlalchemy.ext import declarative
-from sqlalchemy.schema import CreateSchema
 
 
 Base = declarative.declarative_base()
-event.listen(Base.metadata, 'before_create', CreateSchema('other_schema'))
+event.listen(
+    Base.metadata,
+    'before_create',
+    DDL('CREATE SCHEMA IF NOT EXISTS other_schema')
+)
 
 
 class Basic(Base):

tests/test_column_loading.py

@@ -1,24 +1,45 @@
 from unittest import TestCase
 
+from packaging.version import Version
+import sqlalchemy as sa
 from sqlalchemy.types import NullType, VARCHAR
 
 from sqlalchemy_redshift.dialect import RedshiftDialect
 
+sa_version = Version(sa.__version__)
+
 
 class TestColumnReflection(TestCase):
     def test_varchar_as_nulltype(self):
         """
         Varchar columns with no length should be considered NullType columns
         """
         dialect = RedshiftDialect()
-        column_info = dialect._get_column_info(
-            'Null Column',
-            'character varying', None, False, {}, {}, 'default', 'test column'
+
+        null_info = dialect._get_column_info(
+            name='Null Column',
+            format_type='character varying',
+            default=None,
+            notnull=False,
+            domains={},
+            enums=[],
+            schema='default',
+            encode='',
+            comment='test column',
+            identity=None
         )
-        assert isinstance(column_info['type'], NullType)
-        column_info_1 = dialect._get_column_info(
-            'character column',
-            'character varying(30)', None, False, {}, {}, 'default',
-            comment='test column'
+        assert isinstance(null_info['type'], NullType)
+
+        varchar_info = dialect._get_column_info(
+            name='character column',
+            format_type='character varying(30)',
+            default=None,
+            notnull=False,
+            domains={},
+            enums=[],
+            schema='default',
+            encode='',
+            comment='test column',
+            identity=None
         )
-        assert isinstance(column_info_1['type'], VARCHAR)
+        assert isinstance(varchar_info['type'], VARCHAR)

tests/test_delete_stmt.py

@@ -32,9 +32,12 @@
 """
 
 import sqlalchemy as sa
+from packaging.version import Version
 
 from rs_sqla_test_utils.utils import clean, compile_query
 
+sa_version = Version(sa.__version__)
+
 
 meta = sa.MetaData()
 
@@ -116,10 +119,14 @@ def test_delete_stmt_simplewhereclause2():
     del_stmt = sa.delete(customers).where(
         customers.c.email.endswith('test.com')
     )
-    expected = """
-        DELETE FROM customers
-        WHERE customers.email
-        LIKE '%%' || 'test.com'"""
+    if sa_version >= Version('1.4.0'):
+        expected = """
+            DELETE FROM customers
+            WHERE (customers.email LIKE '%%' || 'test.com')"""
+    else:
+        expected = """
+            DELETE FROM customers
+            WHERE customers.email LIKE '%%' || 'test.com'"""
     assert clean(compile_query(del_stmt)) == clean(expected)
 
 

tests/test_reflection_views.py

@@ -14,14 +14,18 @@ def test_view_reflection(redshift_engine):
     view_query = "SELECT my_table.col1, my_table.col2 FROM my_table"
     view_ddl = "CREATE VIEW my_view AS %s" % view_query
     conn = redshift_engine.connect()
-    conn.execute(table_ddl)
-    conn.execute(view_ddl)
-    insp = inspect(redshift_engine)
-    view_definition = insp.get_view_definition('my_view')
-    assert(clean(compile_query(view_definition)) == clean(view_query))
-    view = Table('my_view', MetaData(),
-                 autoload=True, autoload_with=redshift_engine)
-    assert(len(view.columns) == 2)
+    try:
+        conn.execute(table_ddl)
+        conn.execute(view_ddl)
+        insp = inspect(redshift_engine)
+        view_definition = insp.get_view_definition('my_view')
+        assert(clean(compile_query(view_definition)) == clean(view_query))
+        view = Table('my_view', MetaData(),
+                     autoload=True, autoload_with=redshift_engine)
+        assert(len(view.columns) == 2)
+    finally:
+        conn.execute('DROP TABLE IF EXISTS my_table CASCADE')
+        conn.execute('DROP VIEW IF EXISTS my_view CASCADE')
 
 
 def test_late_binding_view_reflection(redshift_engine):
@@ -30,13 +34,17 @@ def test_late_binding_view_reflection(redshift_engine):
     view_ddl = ("CREATE VIEW my_late_view AS "
                 "%s WITH NO SCHEMA BINDING" % view_query)
     conn = redshift_engine.connect()
-    conn.execute(table_ddl)
-    conn.execute(view_ddl)
-    insp = inspect(redshift_engine)
-    view_definition = insp.get_view_definition('my_late_view')
-
-    # For some reason, Redshift returns the entire DDL for late binding views.
-    assert(clean(compile_query(view_definition)) == clean(view_ddl))
-    view = Table('my_late_view', MetaData(),
-                 autoload=True, autoload_with=redshift_engine)
-    assert(len(view.columns) == 2)
+    try:
+        conn.execute(table_ddl)
+        conn.execute(view_ddl)
+        insp = inspect(redshift_engine)
+        view_definition = insp.get_view_definition('my_late_view')
+
+        # Redshift returns the entire DDL for late binding views.
+        assert(clean(compile_query(view_definition)) == clean(view_ddl))
+        view = Table('my_late_view', MetaData(),
+                     autoload=True, autoload_with=redshift_engine)
+        assert(len(view.columns) == 2)
+    finally:
+        conn.execute('DROP TABLE IF EXISTS my_table CASCADE')
+        conn.execute('DROP VIEW IF EXISTS my_late_view CASCADE')

tox.ini

@@ -1,16 +1,21 @@
 [tox]
-envlist = py{27,34,35,36,37}, lint, docs
+envlist =
+    py35-sa13,
+    py{27,36,37,38,39}-sa{13,14},
+    lint,
+    docs
 
 [testenv]
 passenv = PGPASSWORD
-commands = py.test {posargs}
+commands = pytest {posargs}
 deps =
-    requests==2.7.0
-    psycopg2==2.7.3.2
-    sqlalchemy==1.3.0
-    pytest==3.10.1
     alembic==1.4.2
     packaging==20.4
+    psycopg2==2.8.6
+    pytest==3.10.1
+    requests==2.7.0
+    sa13: sqlalchemy==1.3.24
+    sa14: sqlalchemy==1.4.15
 
 [testenv:lint]
 deps =