Squash the hierarchies of databases & dialects from sqeleton & data_diff

Sergey Vasilyev · Sergey Vasilyev · commit bd78b5dbbe93 · 2023-09-22T17:07:16.000+02:00
diff --git a/data_diff/databases/__init__.py b/data_diff/databases/__init__.py
@@ -1,17 +1,17 @@
 from data_diff.sqeleton.databases import MD5_HEXDIGITS, CHECKSUM_HEXDIGITS, QueryError, ConnectError
 
-from data_diff.databases.postgresql import PostgreSQL
-from data_diff.databases.mysql import MySQL
-from data_diff.databases.oracle import Oracle
-from data_diff.databases.snowflake import Snowflake
-from data_diff.databases.bigquery import BigQuery
-from data_diff.databases.redshift import Redshift
-from data_diff.databases.presto import Presto
-from data_diff.databases.databricks import Databricks
-from data_diff.databases.trino import Trino
-from data_diff.databases.clickhouse import Clickhouse
-from data_diff.databases.vertica import Vertica
-from data_diff.databases.duckdb import DuckDB
-from data_diff.databases.mssql import MsSql
+from data_diff.sqeleton.databases.postgresql import PostgreSQL as PostgreSQL
+from data_diff.sqeleton.databases.mysql import MySQL as MySQL
+from data_diff.sqeleton.databases.oracle import Oracle as Oracle
+from data_diff.sqeleton.databases.snowflake import Snowflake as Snowflake
+from data_diff.sqeleton.databases.bigquery import BigQuery as BigQuery
+from data_diff.sqeleton.databases.redshift import Redshift as Redshift
+from data_diff.sqeleton.databases.presto import Presto as Presto
+from data_diff.sqeleton.databases.databricks import Databricks as Databricks
+from data_diff.sqeleton.databases.trino import Trino as Trino
+from data_diff.sqeleton.databases.clickhouse import Clickhouse as Clickhouse
+from data_diff.sqeleton.databases.vertica import Vertica as Vertica
+from data_diff.sqeleton.databases.duckdb import DuckDB as DuckDB
+from data_diff.sqeleton.databases.mssql import MsSQL as MsSql
 
 from data_diff.databases._connect import connect
diff --git a/data_diff/databases/base.py b/data_diff/databases/base.py
diff --git a/data_diff/databases/bigquery.py b/data_diff/databases/bigquery.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import bigquery
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(bigquery.Dialect, bigquery.Mixin_MD5, bigquery.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class BigQuery(bigquery.BigQuery):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.bigquery import Dialect as Dialect
+from data_diff.sqeleton.databases.bigquery import BigQuery as BigQuery
diff --git a/data_diff/databases/clickhouse.py b/data_diff/databases/clickhouse.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import clickhouse
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(clickhouse.Dialect, clickhouse.Mixin_MD5, clickhouse.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class Clickhouse(clickhouse.Clickhouse):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.clickhouse import Dialect as Dialect
+from data_diff.sqeleton.databases.clickhouse import Clickhouse as Clickhouse
diff --git a/data_diff/databases/databricks.py b/data_diff/databases/databricks.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import databricks
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(databricks.Dialect, databricks.Mixin_MD5, databricks.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class Databricks(databricks.Databricks):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.databricks import Dialect as Dialect
+from data_diff.sqeleton.databases.databricks import Databricks as Databricks
diff --git a/data_diff/databases/duckdb.py b/data_diff/databases/duckdb.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import duckdb
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(duckdb.Dialect, duckdb.Mixin_MD5, duckdb.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class DuckDB(duckdb.DuckDB):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.duckdb import Dialect as Dialect
+from data_diff.sqeleton.databases.duckdb import DuckDB as DuckDB
diff --git a/data_diff/databases/mssql.py b/data_diff/databases/mssql.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import mssql
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(mssql.Dialect, mssql.Mixin_MD5, mssql.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class MsSql(mssql.MsSQL):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.mssql import Dialect as Dialect
+from data_diff.sqeleton.databases.mssql import MsSQL as MsSql
diff --git a/data_diff/databases/mysql.py b/data_diff/databases/mysql.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import mysql
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(mysql.Dialect, mysql.Mixin_MD5, mysql.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class MySQL(mysql.MySQL):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.mysql import Dialect as Dialect
+from data_diff.sqeleton.databases.mysql import MySQL as MySQL
diff --git a/data_diff/databases/oracle.py b/data_diff/databases/oracle.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import oracle
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(oracle.Dialect, oracle.Mixin_MD5, oracle.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class Oracle(oracle.Oracle):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.oracle import Dialect as Dialect
+from data_diff.sqeleton.databases.oracle import Oracle as Oracle
diff --git a/data_diff/databases/postgresql.py b/data_diff/databases/postgresql.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import postgresql as pg
-from data_diff.databases.base import DatadiffDialect
-
-
-class PostgresqlDialect(pg.PostgresqlDialect, pg.Mixin_MD5, pg.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class PostgreSQL(pg.PostgreSQL):
-    dialect = PostgresqlDialect()
+from data_diff.sqeleton.databases.postgresql import PostgresqlDialect as PostgresqlDialect
+from data_diff.sqeleton.databases.postgresql import PostgreSQL as PostgreSQL
diff --git a/data_diff/databases/presto.py b/data_diff/databases/presto.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import presto
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(presto.Dialect, presto.Mixin_MD5, presto.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class Presto(presto.Presto):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.presto import Dialect as Dialect
+from data_diff.sqeleton.databases.presto import Presto as Presto
diff --git a/data_diff/databases/redshift.py b/data_diff/databases/redshift.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import redshift
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(redshift.Dialect, redshift.Mixin_MD5, redshift.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class Redshift(redshift.Redshift):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.redshift import Dialect as Dialect
+from data_diff.sqeleton.databases.redshift import Redshift as Redshift
diff --git a/data_diff/databases/snowflake.py b/data_diff/databases/snowflake.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import snowflake
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(snowflake.Dialect, snowflake.Mixin_MD5, snowflake.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class Snowflake(snowflake.Snowflake):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.snowflake import Dialect as Dialect
+from data_diff.sqeleton.databases.snowflake import Snowflake as Snowflake
diff --git a/data_diff/databases/trino.py b/data_diff/databases/trino.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import trino
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(trino.Dialect, trino.Mixin_MD5, trino.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class Trino(trino.Trino):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.trino import Dialect as Dialect
+from data_diff.sqeleton.databases.trino import Trino as Trino
diff --git a/data_diff/databases/vertica.py b/data_diff/databases/vertica.py
@@ -1,10 +1,2 @@
-from data_diff.sqeleton.databases import vertica
-from data_diff.databases.base import DatadiffDialect
-
-
-class Dialect(vertica.Dialect, vertica.Mixin_MD5, vertica.Mixin_NormalizeValue, DatadiffDialect):
-    pass
-
-
-class Vertica(vertica.Vertica):
-    dialect = Dialect()
+from data_diff.sqeleton.databases.vertica import Dialect as Dialect
+from data_diff.sqeleton.databases.vertica import Vertica as Vertica
diff --git a/data_diff/sqeleton/databases/bigquery.py b/data_diff/sqeleton/databases/bigquery.py
@@ -139,7 +139,7 @@ def time_travel(
         )
 
 
-class Dialect(BaseDialect, Mixin_Schema):
+class Dialect(BaseDialect, Mixin_Schema, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "BigQuery"
     ROUNDS_ON_PREC_LOSS = False  # Technically BigQuery doesn't allow implicit rounding or truncation
     TYPE_CLASSES = {
diff --git a/data_diff/sqeleton/databases/clickhouse.py b/data_diff/sqeleton/databases/clickhouse.py
@@ -99,7 +99,7 @@ def normalize_timestamp(self, value: str, coltype: TemporalType) -> str:
         return f"rpad({value}, {TIMESTAMP_PRECISION_POS + 6}, '0')"
 
 
-class Dialect(BaseDialect):
+class Dialect(BaseDialect, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "Clickhouse"
     ROUNDS_ON_PREC_LOSS = False
     TYPE_CLASSES = {
diff --git a/data_diff/sqeleton/databases/databricks.py b/data_diff/sqeleton/databases/databricks.py
@@ -60,7 +60,7 @@ def normalize_boolean(self, value: str, _coltype: Boolean) -> str:
         return self.to_string(f"cast ({value} as int)")
 
 
-class Dialect(BaseDialect):
+class Dialect(BaseDialect, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "Databricks"
     ROUNDS_ON_PREC_LOSS = True
     TYPE_CLASSES = {
diff --git a/data_diff/sqeleton/databases/duckdb.py b/data_diff/sqeleton/databases/duckdb.py
@@ -75,7 +75,7 @@ def test_regex(self, string: Compilable, pattern: Compilable) -> Compilable:
         return Func("regexp_matches", [string, pattern])
 
 
-class Dialect(BaseDialect, Mixin_Schema):
+class Dialect(BaseDialect, Mixin_Schema, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "DuckDB"
     ROUNDS_ON_PREC_LOSS = False
     SUPPORTS_PRIMARY_KEY = True
diff --git a/data_diff/sqeleton/databases/mssql.py b/data_diff/sqeleton/databases/mssql.py
@@ -58,7 +58,7 @@ def md5_as_int(self, s: str) -> str:
         return f"convert(bigint, convert(varbinary, '0x' + RIGHT(CONVERT(NVARCHAR(32), HashBytes('MD5', {s}), 2), {CHECKSUM_HEXDIGITS}), 1))"
 
 
-class Dialect(BaseDialect, Mixin_Schema, Mixin_OptimizerHints):
+class Dialect(BaseDialect, Mixin_Schema, Mixin_OptimizerHints, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "MsSQL"
     ROUNDS_ON_PREC_LOSS = True
     SUPPORTS_PRIMARY_KEY = True
diff --git a/data_diff/sqeleton/databases/mysql.py b/data_diff/sqeleton/databases/mysql.py
@@ -67,7 +67,7 @@ def test_regex(self, string: Compilable, pattern: Compilable) -> Compilable:
         return BinBoolOp("REGEXP", [string, pattern])
 
 
-class Dialect(BaseDialect, Mixin_Schema, Mixin_OptimizerHints):
+class Dialect(BaseDialect, Mixin_Schema, Mixin_OptimizerHints, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "MySQL"
     ROUNDS_ON_PREC_LOSS = True
     SUPPORTS_PRIMARY_KEY = True
diff --git a/data_diff/sqeleton/databases/oracle.py b/data_diff/sqeleton/databases/oracle.py
@@ -80,7 +80,7 @@ def list_tables(self, table_schema: str, like: Compilable = None) -> Compilable:
         )
 
 
-class Dialect(BaseDialect, Mixin_Schema, Mixin_OptimizerHints):
+class Dialect(BaseDialect, Mixin_Schema, Mixin_OptimizerHints, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "Oracle"
     SUPPORTS_PRIMARY_KEY = True
     SUPPORTS_INDEXES = True
diff --git a/data_diff/sqeleton/databases/postgresql.py b/data_diff/sqeleton/databases/postgresql.py
@@ -61,7 +61,7 @@ def normalize_json(self, value: str, _coltype: JSON) -> str:
         return f"{value}::text"
 
 
-class PostgresqlDialect(BaseDialect, Mixin_Schema):
+class PostgresqlDialect(BaseDialect, Mixin_Schema, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "PostgreSQL"
     ROUNDS_ON_PREC_LOSS = True
     SUPPORTS_PRIMARY_KEY = True
diff --git a/data_diff/sqeleton/databases/presto.py b/data_diff/sqeleton/databases/presto.py
@@ -76,7 +76,7 @@ def normalize_boolean(self, value: str, _coltype: Boolean) -> str:
         return self.to_string(f"cast ({value} as int)")
 
 
-class Dialect(BaseDialect, Mixin_Schema):
+class Dialect(BaseDialect, Mixin_Schema, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "Presto"
     ROUNDS_ON_PREC_LOSS = True
     TYPE_CLASSES = {
diff --git a/data_diff/sqeleton/databases/redshift.py b/data_diff/sqeleton/databases/redshift.py
@@ -7,7 +7,7 @@
     DbPath,
     TimestampTZ,
 )
-from data_diff.sqeleton.abcs.mixins import AbstractMixin_MD5
+from data_diff.sqeleton.abcs.mixins import AbstractMixin_MD5, AbstractMixin_NormalizeValue
 from data_diff.sqeleton.databases.postgresql import (
     PostgreSQL,
     MD5_HEXDIGITS,
@@ -51,7 +51,7 @@ def normalize_json(self, value: str, _coltype: JSON) -> str:
         return f"nvl2({value}, json_serialize({value}), NULL)"
 
 
-class Dialect(PostgresqlDialect):
+class Dialect(PostgresqlDialect, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "Redshift"
     TYPE_CLASSES = {
         **PostgresqlDialect.TYPE_CLASSES,
diff --git a/data_diff/sqeleton/databases/snowflake.py b/data_diff/sqeleton/databases/snowflake.py
@@ -104,7 +104,7 @@ def time_travel(
         return code(f"{{table}} {at_or_before}({key} => {{value}})", table=table, value=value)
 
 
-class Dialect(BaseDialect, Mixin_Schema):
+class Dialect(BaseDialect, Mixin_Schema, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "Snowflake"
     ROUNDS_ON_PREC_LOSS = False
     TYPE_CLASSES = {
diff --git a/data_diff/sqeleton/databases/trino.py b/data_diff/sqeleton/databases/trino.py
@@ -1,3 +1,4 @@
+from data_diff.sqeleton.abcs.mixins import AbstractMixin_MD5, AbstractMixin_NormalizeValue
 from data_diff.sqeleton.abcs.database_types import TemporalType, ColType_UUID
 from data_diff.sqeleton.databases import presto
 from data_diff.sqeleton.databases.base import import_helper
@@ -29,7 +30,7 @@ def normalize_uuid(self, value: str, coltype: ColType_UUID) -> str:
         return f"TRIM({value})"
 
 
-class Dialect(presto.Dialect):
+class Dialect(presto.Dialect, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "Trino"
 
 
diff --git a/data_diff/sqeleton/databases/vertica.py b/data_diff/sqeleton/databases/vertica.py
@@ -78,7 +78,7 @@ def list_tables(self, table_schema: str, like: Compilable = None) -> Compilable:
         )
 
 
-class Dialect(BaseDialect, Mixin_Schema):
+class Dialect(BaseDialect, Mixin_Schema, Mixin_MD5, Mixin_NormalizeValue, AbstractMixin_MD5, AbstractMixin_NormalizeValue):
     name = "Vertica"
     ROUNDS_ON_PREC_LOSS = True
 
