feature/decouple adapters from core

dagger/run_dbt_spark_tests.py

@@ -2,6 +2,7 @@
 
 import argparse
 import sys
+from typing import Dict
 
 import anyio as anyio
 import dagger as dagger
@@ -19,7 +20,7 @@
 TESTING_ENV_VARS.update({"ODBC_DRIVER": "Simba"})
 
 
-def env_variables(envs: dict[str, str]):
+def env_variables(envs: Dict[str, str]):
     def env_variables_inner(ctr: dagger.Container):
         for key, value in envs.items():
             ctr = ctr.with_env_variable(key, value)

dbt/adapters/spark/column.py

@@ -2,7 +2,7 @@
 from typing import Any, Dict, Optional, TypeVar, Union
 
 from dbt.adapters.base.column import Column
-from dbt.dataclass_schema import dbtClassMixin
+from dbt.common.dataclass_schema import dbtClassMixin
 
 Self = TypeVar("Self", bound="SparkColumn")
 

dbt/adapters/spark/connections.py

@@ -2,10 +2,12 @@
 
 import dbt.exceptions
 from dbt.adapters.base import Credentials
+from dbt.adapters.contracts.connection import AdapterResponse, ConnectionState, Connection
+from dbt.adapters.events.logging import AdapterLogger
 from dbt.adapters.sql import SQLConnectionManager
-from dbt.contracts.connection import ConnectionState, AdapterResponse
-from dbt.events import AdapterLogger
-from dbt.utils import DECIMALS
+from dbt.common.exceptions import DbtConfigError
+
+from dbt.common.utils.encoding import DECIMALS
 from dbt.adapters.spark import __version__
 
 try:
@@ -22,8 +24,7 @@
     pyodbc = None
 from datetime import datetime
 import sqlparams
-from dbt.contracts.connection import Connection
-from dbt.dataclass_schema import StrEnum
+from dbt.common.dataclass_schema import StrEnum
 from dataclasses import dataclass, field
 from typing import Any, Dict, Optional, Union, Tuple, List, Generator, Iterable, Sequence
 
@@ -390,7 +391,7 @@ def validate_creds(cls, creds: Any, required: Iterable[str]) -> None:
 
         for key in required:
             if not hasattr(creds, key):
-                raise dbt.exceptions.DbtProfileError(
+                raise DbtConfigError(
                     "The config '{}' is required when using the {} method"
                     " to connect to Spark".format(key, method)
                 )
@@ -481,7 +482,7 @@ def open(cls, connection: Connection) -> Connection:
                             endpoint=creds.endpoint
                         )
                     else:
-                        raise dbt.exceptions.DbtProfileError(
+                        raise DbtConfigError(
                             "Either `cluster` or `endpoint` must set when"
                             " using the odbc method to connect to Spark"
                         )
@@ -525,7 +526,7 @@ def open(cls, connection: Connection) -> Connection:
                         Connection(server_side_parameters=creds.server_side_parameters)
                     )
                 else:
-                    raise dbt.exceptions.DbtProfileError(
+                    raise DbtConfigError(
                         f"invalid credential method: {creds.method}"
                     )
                 break

dbt/adapters/spark/impl.py

@@ -4,7 +4,9 @@
 from typing import Any, Dict, Iterable, List, Optional, Union, Type, Tuple, Callable, Set
 
 from dbt.adapters.base.relation import InformationSchema
-from dbt.contracts.graph.manifest import Manifest
+from dbt.adapters.contracts.connection import AdapterResponse
+from dbt.adapters.events.logging import AdapterLogger
+from dbt.common.utils import AttrDict, executor
 
 from typing_extensions import TypeAlias
 
@@ -13,6 +15,7 @@
 import dbt
 import dbt.exceptions
 
+
 from dbt.adapters.base import AdapterConfig, PythonJobHelper
 from dbt.adapters.base.impl import catch_as_completed, ConstraintSupport
 from dbt.adapters.sql import SQLAdapter
@@ -24,12 +27,9 @@
     AllPurposeClusterPythonJobHelper,
 )
 from dbt.adapters.base import BaseRelation
-from dbt.clients.agate_helper import DEFAULT_TYPE_TESTER
-from dbt.contracts.connection import AdapterResponse
-from dbt.contracts.graph.nodes import ConstraintType
-from dbt.contracts.relation import RelationType
-from dbt.events import AdapterLogger
-from dbt.utils import executor, AttrDict
+from dbt.adapters.contracts.relation import RelationType, RelationConfig
+from dbt.common.clients.agate_helper import DEFAULT_TYPE_TESTER
+from dbt.common.contracts.constraints import ConstraintType
 
 logger = AdapterLogger("Spark")
 
@@ -352,9 +352,9 @@ def _get_columns_for_catalog(self, relation: BaseRelation) -> Iterable[Dict[str,
             yield as_dict
 
     def get_catalog(
-        self, manifest: Manifest, selected_nodes: Optional[Set] = None
+        self, relation_configs: Iterable[RelationConfig], selected_nodes: Optional[Set] = None
     ) -> Tuple[agate.Table, List[Exception]]:
-        schema_map = self._get_catalog_schemas(manifest)
+        schema_map = self._get_catalog_schemas(relation_configs)
         if len(schema_map) > 1:
             raise dbt.exceptions.CompilationError(
                 f"Expected only one database in get_catalog, found " f"{list(schema_map)}"
@@ -371,7 +371,7 @@ def get_catalog(
                             self._get_one_catalog,
                             info,
                             [schema],
-                            manifest,
+                            relation_configs,
                         )
                     )
             catalogs, exceptions = catch_as_completed(futures)
@@ -381,7 +381,7 @@ def _get_one_catalog(
         self,
         information_schema: InformationSchema,
         schemas: Set[str],
-        manifest: Manifest,
+        relation_configs: Iterable[RelationConfig],
     ) -> agate.Table:
         if len(schemas) != 1:
             raise dbt.exceptions.CompilationError(

dbt/adapters/spark/relation.py

@@ -2,9 +2,9 @@
 from dataclasses import dataclass, field
 
 from dbt.adapters.base.relation import BaseRelation, Policy
+from dbt.adapters.events.logging import AdapterLogger
 
-from dbt.exceptions import DbtRuntimeError
-from dbt.events import AdapterLogger
+from dbt.common.exceptions import DbtRuntimeError
 
 logger = AdapterLogger("Spark")
 

dbt/adapters/spark/session.py

@@ -7,9 +7,9 @@
 from typing import Any, Dict, List, Optional, Tuple, Union, Sequence
 
 from dbt.adapters.spark.connections import SparkConnectionWrapper
-from dbt.events import AdapterLogger
-from dbt.utils import DECIMALS
-from dbt.exceptions import DbtRuntimeError
+from dbt.adapters.events.logging import AdapterLogger
+from dbt.common.utils.encoding import DECIMALS
+from dbt.common.exceptions import DbtRuntimeError
 from pyspark.sql import DataFrame, Row, SparkSession
 from pyspark.sql.utils import AnalysisException
 

dev-requirements.txt

@@ -1,7 +1,7 @@
 # install latest changes in dbt-core
 # TODO: how to automate switching from develop to version branches?
-git+https://github.com/dbt-labs/dbt-core.git@c2bc2f009bbeeb46b3c69d082ab4d485597898af#egg=dbt-core&subdirectory=core
-git+https://github.com/dbt-labs/dbt-core.git@c2bc2f009bbeeb46b3c69d082ab4d485597898af#egg=dbt-tests-adapter&subdirectory=tests/adapter
+git+https://github.com/dbt-labs/dbt-core.git#egg=dbt-core&subdirectory=core
+git+https://github.com/dbt-labs/dbt-core.git#egg=dbt-tests-adapter&subdirectory=tests/adapter
 
 # if version 1.x or greater -> pin to major version
 # if version 0.x -> pin to minor

tests/unit/test_adapter.py

@@ -1,4 +1,5 @@
 import unittest
+from multiprocessing import get_context
 from unittest import mock
 
 import dbt.flags as flags
@@ -146,7 +147,7 @@ def _get_target_odbc_sql_endpoint(self, project):
 
     def test_http_connection(self):
         config = self._get_target_http(self.project_cfg)
-        adapter = SparkAdapter(config)
+        adapter = SparkAdapter(config, get_context("spawn"))
 
         def hive_http_connect(thrift_transport, configuration):
             self.assertEqual(thrift_transport.scheme, "https")
@@ -171,7 +172,7 @@ def hive_http_connect(thrift_transport, configuration):
 
     def test_thrift_connection(self):
         config = self._get_target_thrift(self.project_cfg)
-        adapter = SparkAdapter(config)
+        adapter = SparkAdapter(config, get_context("spawn"))
 
         def hive_thrift_connect(
             host, port, username, auth, kerberos_service_name, password, configuration
@@ -195,7 +196,7 @@ def hive_thrift_connect(
 
     def test_thrift_ssl_connection(self):
         config = self._get_target_use_ssl_thrift(self.project_cfg)
-        adapter = SparkAdapter(config)
+        adapter = SparkAdapter(config, get_context("spawn"))
 
         def hive_thrift_connect(thrift_transport, configuration):
             self.assertIsNotNone(thrift_transport)
@@ -215,7 +216,7 @@ def hive_thrift_connect(thrift_transport, configuration):
 
     def test_thrift_connection_kerberos(self):
         config = self._get_target_thrift_kerberos(self.project_cfg)
-        adapter = SparkAdapter(config)
+        adapter = SparkAdapter(config, get_context("spawn"))
 
         def hive_thrift_connect(
             host, port, username, auth, kerberos_service_name, password, configuration
@@ -239,7 +240,7 @@ def hive_thrift_connect(
 
     def test_odbc_cluster_connection(self):
         config = self._get_target_odbc_cluster(self.project_cfg)
-        adapter = SparkAdapter(config)
+        adapter = SparkAdapter(config, get_context("spawn"))
 
         def pyodbc_connect(connection_str, autocommit):
             self.assertTrue(autocommit)
@@ -266,7 +267,7 @@ def pyodbc_connect(connection_str, autocommit):
 
     def test_odbc_endpoint_connection(self):
         config = self._get_target_odbc_sql_endpoint(self.project_cfg)
-        adapter = SparkAdapter(config)
+        adapter = SparkAdapter(config, get_context("spawn"))
 
         def pyodbc_connect(connection_str, autocommit):
             self.assertTrue(autocommit)
@@ -329,7 +330,7 @@ def test_parse_relation(self):
         input_cols = [Row(keys=["col_name", "data_type"], values=r) for r in plain_rows]
 
         config = self._get_target_http(self.project_cfg)
-        rows = SparkAdapter(config).parse_describe_extended(relation, input_cols)
+        rows = SparkAdapter(config, get_context("spawn")).parse_describe_extended(relation, input_cols)
         self.assertEqual(len(rows), 4)
         self.assertEqual(
             rows[0].to_column_dict(omit_none=False),
@@ -418,7 +419,7 @@ def test_parse_relation_with_integer_owner(self):
         input_cols = [Row(keys=["col_name", "data_type"], values=r) for r in plain_rows]
 
         config = self._get_target_http(self.project_cfg)
-        rows = SparkAdapter(config).parse_describe_extended(relation, input_cols)
+        rows = SparkAdapter(config, get_context("spawn")).parse_describe_extended(relation, input_cols)
 
         self.assertEqual(rows[0].to_column_dict().get("table_owner"), "1234")
 
@@ -454,7 +455,7 @@ def test_parse_relation_with_statistics(self):
         input_cols = [Row(keys=["col_name", "data_type"], values=r) for r in plain_rows]
 
         config = self._get_target_http(self.project_cfg)
-        rows = SparkAdapter(config).parse_describe_extended(relation, input_cols)
+        rows = SparkAdapter(config, get_context("spawn")).parse_describe_extended(relation, input_cols)
         self.assertEqual(len(rows), 1)
         self.assertEqual(
             rows[0].to_column_dict(omit_none=False),
@@ -483,7 +484,7 @@ def test_parse_relation_with_statistics(self):
 
     def test_relation_with_database(self):
         config = self._get_target_http(self.project_cfg)
-        adapter = SparkAdapter(config)
+        adapter = SparkAdapter(config, get_context("spawn"))
         # fine
         adapter.Relation.create(schema="different", identifier="table")
         with self.assertRaises(DbtRuntimeError):
@@ -564,7 +565,7 @@ def test_parse_columns_from_information_with_table_type_and_delta_provider(self)
         )
 
         config = self._get_target_http(self.project_cfg)
-        columns = SparkAdapter(config).parse_columns_from_information(relation)
+        columns = SparkAdapter(config, get_context("spawn")).parse_columns_from_information(relation)
         self.assertEqual(len(columns), 4)
         self.assertEqual(
             columns[0].to_column_dict(omit_none=False),
@@ -649,7 +650,7 @@ def test_parse_columns_from_information_with_view_type(self):
         )
 
         config = self._get_target_http(self.project_cfg)
-        columns = SparkAdapter(config).parse_columns_from_information(relation)
+        columns = SparkAdapter(config, get_context("spawn")).parse_columns_from_information(relation)
         self.assertEqual(len(columns), 4)
         self.assertEqual(
             columns[1].to_column_dict(omit_none=False),
@@ -715,7 +716,7 @@ def test_parse_columns_from_information_with_table_type_and_parquet_provider(sel
         )
 
         config = self._get_target_http(self.project_cfg)
-        columns = SparkAdapter(config).parse_columns_from_information(relation)
+        columns = SparkAdapter(config, get_context("spawn")).parse_columns_from_information(relation)
         self.assertEqual(len(columns), 4)
 
         self.assertEqual(

tests/unit/utils.py

@@ -9,7 +9,7 @@
 
 import agate
 import pytest
-from dbt.dataclass_schema import ValidationError
+from dbt.common.dataclass_schema import ValidationError
 from dbt.config.project import PartialProject