cleanup

Author: dlawin

data_diff/sqeleton/databases/base.py

@@ -527,8 +527,6 @@ def _query_cursor(self, c, sql_code: str) -> QueryResult:
             if sql_code.lower().startswith(("select", "explain", "show")):
                 columns = [col[0] for col in c.description]
 
-                # TODO FIXME pyodbc.Row seems to be causing a pydantic error
-                # [ConstantTable] Attribute 'rows' expected value of type Sequence[Sequence[Any]]
                 fetched = c.fetchall()
                 result = QueryResult(fetched, columns)
                 return result

data_diff/sqeleton/databases/mssql.py

@@ -35,8 +35,6 @@ def import_mssql():
 
 
 class Mixin_NormalizeValue(AbstractMixin_NormalizeValue):
-    # TODO FIXME the output in this method seems correct but timestamp precision isn't
-    # being properly concatenated in tests
     def normalize_timestamp(self, value: str, coltype: TemporalType) -> str:
         if coltype.precision > 0:
             formatted_value = (
@@ -57,11 +55,6 @@ def normalize_number(self, value: str, coltype: FractionalType) -> str:
 
 class Mixin_MD5(AbstractMixin_MD5):
     def md5_as_int(self, s: str) -> str:
-        # TODO FYI
-        # I didn't see another way to do this
-        # keep in mind this also required reducing
-        # CHECKSUM_HEXDIGITS = 15 -> 14 due to the convert needing an even number of hex digits
-        # (affects all dbs)
         return f"convert(bigint, convert(varbinary, '0x' + RIGHT(CONVERT(NVARCHAR(32), HashBytes('MD5', {s}), 2), {CHECKSUM_HEXDIGITS}), 1))"
 
 
@@ -168,8 +161,6 @@ class MsSQL(ThreadedDatabase):
     def __init__(self, host, port, user, password, *, database, thread_count, **kw):
         args = dict(server=host, port=port, database=database, user=user, password=password, **kw)
         self._args = {k: v for k, v in args.items() if v is not None}
-
-        # TODO User supplied value?
         self._args["driver"] = "{ODBC Driver 18 for SQL Server}"
 
         # TODO temp dev debug

tests/common.py

@@ -42,8 +42,7 @@
 N_SAMPLES = int(os.environ.get("N_SAMPLES", DEFAULT_N_SAMPLES))
 BENCHMARK = os.environ.get("BENCHMARK", False)
 N_THREADS = int(os.environ.get("N_THREADS", 1))
-# TODO temp dev
-TEST_ACROSS_ALL_DBS = os.environ.get("TEST_ACROSS_ALL_DBS", "full")  # Should we run the full db<->db test suite?
+TEST_ACROSS_ALL_DBS = os.environ.get("TEST_ACROSS_ALL_DBS", True)  # Should we run the full db<->db test suite?
 
 
 def get_git_revision_short_hash() -> str:

tests/sqeleton/test_database.py

@@ -7,35 +7,33 @@
 from data_diff.sqeleton import connect
 from data_diff.sqeleton import databases as dbs
 from data_diff.sqeleton.queries import table, current_timestamp, NormalizeAsString
-from tests.common import TEST_MYSQL_CONN_STRING, TEST_MSSQL_CONN_STRING
+from tests.common import TEST_MYSQL_CONN_STRING
 from tests.sqeleton.common import str_to_checksum, test_each_database_in_list, get_conn, random_table_suffix
 from data_diff.sqeleton.abcs.database_types import TimestampTZ
 
 TEST_DATABASES = {
-    # TODO dev
-    # dbs.MySQL,
-    # dbs.PostgreSQL,
-    # dbs.Oracle,
-    # dbs.Redshift,
-    # dbs.Snowflake,
-    # dbs.DuckDB,
-    # dbs.BigQuery,
-    # dbs.Presto,
-    # dbs.Trino,
-    # dbs.Vertica,
-    dbs.MsSQL
+    dbs.MySQL,
+    dbs.PostgreSQL,
+    dbs.Oracle,
+    dbs.Redshift,
+    dbs.Snowflake,
+    dbs.DuckDB,
+    dbs.BigQuery,
+    dbs.Presto,
+    dbs.Trino,
+    dbs.Vertica,
+    dbs.MsSQL,
 }
 
 test_each_database: Callable = test_each_database_in_list(TEST_DATABASES)
 
 
 class TestDatabase(unittest.TestCase):
     def setUp(self):
-        # TODO temp dev
-        self.mssql = connect(TEST_MSSQL_CONN_STRING)
+        self.mysql = connect(TEST_MYSQL_CONN_STRING)
 
     def test_connect_to_db(self):
-        self.assertEqual(1, self.mssql.query("SELECT 1", int))
+        self.assertEqual(1, self.mysql.query("SELECT 1", int))
 
 
 class TestMD5(unittest.TestCase):

tests/test_database_types.py

@@ -353,7 +353,6 @@ def init_conns():
     },
     db.MsSql: {
         "int": ["INT", "BIGINT"],
-        # TODO precision is being truncated to 4 somewhere
         "datetime": ["datetime2(6)"],
         "float": ["DECIMAL(6, 2)", "FLOAT", "REAL"],
         "uuid": ["VARCHAR(100)", "CHAR(100)", "UNIQUEIDENTIFIER"],