sqla_models_tests.py

# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.
# isort:skip_file
import re
from datetime import datetime
from typing import Any, Dict, List, NamedTuple, Optional, Pattern, Tuple, Union
from unittest.mock import patch
import pytest

import numpy as np
import pandas as pd
import sqlalchemy as sa
from flask import Flask
from pytest_mock import MockFixture
from sqlalchemy.sql import text
from sqlalchemy.sql.elements import TextClause

from superset import db
from superset.connectors.sqla.models import SqlaTable, TableColumn, SqlMetric
from superset.constants import EMPTY_STRING, NULL_STRING
from superset.db_engine_specs.bigquery import BigQueryEngineSpec
from superset.db_engine_specs.druid import DruidEngineSpec
from superset.exceptions import QueryObjectValidationError, SupersetSecurityException
from superset.models.core import Database
from superset.utils.core import (
    AdhocMetricExpressionType,
    FilterOperator,
    GenericDataType,
    TemporalType,
    backend,
)
from superset.utils.database import get_example_database
from tests.integration_tests.fixtures.birth_names_dashboard import (
    load_birth_names_dashboard_with_slices,
    load_birth_names_data,
)
from tests.integration_tests.test_app import app

from .base_tests import SupersetTestCase


VIRTUAL_TABLE_INT_TYPES: Dict[str, Pattern[str]] = {
    "hive": re.compile(r"^INT_TYPE$"),
    "mysql": re.compile("^LONGLONG$"),
    "postgresql": re.compile(r"^INT$"),
    "presto": re.compile(r"^INTEGER$"),
    "sqlite": re.compile(r"^INT$"),
}

VIRTUAL_TABLE_STRING_TYPES: Dict[str, Pattern[str]] = {
    "hive": re.compile(r"^STRING_TYPE$"),
    "mysql": re.compile(r"^VAR_STRING$"),
    "postgresql": re.compile(r"^STRING$"),
    "presto": re.compile(r"^VARCHAR*"),
    "sqlite": re.compile(r"^STRING$"),
}


class FilterTestCase(NamedTuple):
    operator: str
    value: Union[float, int, List[Any], str]
    expected: Union[str, List[str]]


class TestDatabaseModel(SupersetTestCase):
    def test_is_time_druid_time_col(self):
        """Druid has a special __time column"""

        database = Database(database_name="druid_db", sqlalchemy_uri="druid://db")
        tbl = SqlaTable(table_name="druid_tbl", database=database)
        col = TableColumn(column_name="__time", type="INTEGER", table=tbl)
        self.assertEqual(col.is_dttm, None)
        DruidEngineSpec.alter_new_orm_column(col)
        self.assertEqual(col.is_dttm, True)

        col = TableColumn(column_name="__not_time", type="INTEGER", table=tbl)
        self.assertEqual(col.is_temporal, False)

    def test_temporal_varchar(self):
        """Ensure a column with is_dttm set to true evaluates to is_temporal == True"""

        database = get_example_database()
        tbl = SqlaTable(table_name="test_tbl", database=database)
        col = TableColumn(column_name="ds", type="VARCHAR", table=tbl)
        # by default, VARCHAR should not be assumed to be temporal
        assert col.is_temporal is False
        # changing to `is_dttm = True`, calling `is_temporal` should return True
        col.is_dttm = True
        assert col.is_temporal is True

    def test_db_column_types(self):
        test_cases: Dict[str, GenericDataType] = {
            # string
            "CHAR": GenericDataType.STRING,
            "VARCHAR": GenericDataType.STRING,
            "NVARCHAR": GenericDataType.STRING,
            "STRING": GenericDataType.STRING,
            "TEXT": GenericDataType.STRING,
            "NTEXT": GenericDataType.STRING,
            # numeric
            "INTEGER": GenericDataType.NUMERIC,
            "BIGINT": GenericDataType.NUMERIC,
            "DECIMAL": GenericDataType.NUMERIC,
            # temporal
            "DATE": GenericDataType.TEMPORAL,
            "DATETIME": GenericDataType.TEMPORAL,
            "TIME": GenericDataType.TEMPORAL,
            "TIMESTAMP": GenericDataType.TEMPORAL,
        }

        tbl = SqlaTable(table_name="col_type_test_tbl", database=get_example_database())
        for str_type, db_col_type in test_cases.items():
            col = TableColumn(column_name="foo", type=str_type, table=tbl)
            self.assertEqual(col.is_temporal, db_col_type == GenericDataType.TEMPORAL)
            self.assertEqual(col.is_numeric, db_col_type == GenericDataType.NUMERIC)
            self.assertEqual(col.is_string, db_col_type == GenericDataType.STRING)

        for str_type, db_col_type in test_cases.items():
            col = TableColumn(column_name="foo", type=str_type, table=tbl, is_dttm=True)
            self.assertTrue(col.is_temporal)

    @patch("superset.jinja_context.g")
    def test_extra_cache_keys(self, flask_g):
        flask_g.user.username = "abc"
        base_query_obj = {
            "granularity": None,
            "from_dttm": None,
            "to_dttm": None,
            "groupby": ["user"],
            "metrics": [],
            "is_timeseries": False,
            "filter": [],
        }

        # Table with Jinja callable.
        table1 = SqlaTable(
            table_name="test_has_extra_cache_keys_table",
            sql="SELECT '{{ current_username() }}' as user",
            database=get_example_database(),
        )

        query_obj = dict(**base_query_obj, extras={})
        extra_cache_keys = table1.get_extra_cache_keys(query_obj)
        self.assertTrue(table1.has_extra_cache_key_calls(query_obj))
        assert extra_cache_keys == ["abc"]

        # Table with Jinja callable disabled.
        table2 = SqlaTable(
            table_name="test_has_extra_cache_keys_disabled_table",
            sql="SELECT '{{ current_username(False) }}' as user",
            database=get_example_database(),
        )
        query_obj = dict(**base_query_obj, extras={})
        extra_cache_keys = table2.get_extra_cache_keys(query_obj)
        self.assertTrue(table2.has_extra_cache_key_calls(query_obj))
        self.assertListEqual(extra_cache_keys, [])

        # Table with no Jinja callable.
        query = "SELECT 'abc' as user"
        table3 = SqlaTable(
            table_name="test_has_no_extra_cache_keys_table",
            sql=query,
            database=get_example_database(),
        )

        query_obj = dict(**base_query_obj, extras={"where": "(user != 'abc')"})
        extra_cache_keys = table3.get_extra_cache_keys(query_obj)
        self.assertFalse(table3.has_extra_cache_key_calls(query_obj))
        self.assertListEqual(extra_cache_keys, [])

        # With Jinja callable in SQL expression.
        query_obj = dict(
            **base_query_obj, extras={"where": "(user != '{{ current_username() }}')"}
        )
        extra_cache_keys = table3.get_extra_cache_keys(query_obj)
        self.assertTrue(table3.has_extra_cache_key_calls(query_obj))
        assert extra_cache_keys == ["abc"]

        # Cleanup
        for table in [table1, table2, table3]:
            db.session.delete(table)
        db.session.commit()

    @patch("superset.jinja_context.g")
    def test_jinja_metrics_and_calc_columns(self, flask_g):
        flask_g.user.username = "abc"
        base_query_obj = {
            "granularity": None,
            "from_dttm": None,
            "to_dttm": None,
            "groupby": ["user", "expr"],
            "metrics": [
                {
                    "expressionType": AdhocMetricExpressionType.SQL,
                    "sqlExpression": "SUM(case when user = '{{ current_username() }}' "
                    "then 1 else 0 end)",
                    "label": "SUM(userid)",
                }
            ],
            "is_timeseries": False,
            "filter": [],
        }

        table = SqlaTable(
            table_name="test_has_jinja_metric_and_expr",
            sql="SELECT '{{ current_username() }}' as user",
            database=get_example_database(),
        )
        TableColumn(
            column_name="expr",
            expression="case when '{{ current_username() }}' = 'abc' "
            "then 'yes' else 'no' end",
            type="VARCHAR(100)",
            table=table,
        )
        db.session.commit()

        sqla_query = table.get_sqla_query(**base_query_obj)
        query = table.database.compile_sqla_query(sqla_query.sqla_query)
        # assert expression
        assert "case when 'abc' = 'abc' then 'yes' else 'no' end" in query
        # assert metric
        assert "SUM(case when user = 'abc' then 1 else 0 end)" in query
        # Cleanup
        db.session.delete(table)
        db.session.commit()

    def test_adhoc_metrics_and_calc_columns(self):
        base_query_obj = {
            "granularity": None,
            "from_dttm": None,
            "to_dttm": None,
            "groupby": ["user", "expr"],
            "metrics": [
                {
                    "expressionType": AdhocMetricExpressionType.SQL,
                    "sqlExpression": "(SELECT (SELECT * from birth_names) "
                    "from test_validate_adhoc_sql)",
                    "label": "adhoc_metrics",
                }
            ],
            "is_timeseries": False,
            "filter": [],
        }

        table = SqlaTable(
            table_name="test_validate_adhoc_sql", database=get_example_database()
        )
        db.session.commit()

        with pytest.raises(SupersetSecurityException):
            table.get_sqla_query(**base_query_obj)
        # Cleanup
        db.session.delete(table)
        db.session.commit()

    @pytest.mark.usefixtures("load_birth_names_dashboard_with_slices")
    def test_where_operators(self):
        filters: Tuple[FilterTestCase, ...] = (
            FilterTestCase(FilterOperator.IS_NULL, "", "IS NULL"),
            FilterTestCase(FilterOperator.IS_NOT_NULL, "", "IS NOT NULL"),
            # Some db backends translate true/false to 1/0
            FilterTestCase(FilterOperator.IS_TRUE, "", ["IS 1", "IS true"]),
            FilterTestCase(FilterOperator.IS_FALSE, "", ["IS 0", "IS false"]),
            FilterTestCase(FilterOperator.GREATER_THAN, 0, "> 0"),
            FilterTestCase(FilterOperator.GREATER_THAN_OR_EQUALS, 0, ">= 0"),
            FilterTestCase(FilterOperator.LESS_THAN, 0, "< 0"),
            FilterTestCase(FilterOperator.LESS_THAN_OR_EQUALS, 0, "<= 0"),
            FilterTestCase(FilterOperator.EQUALS, 0, "= 0"),
            FilterTestCase(FilterOperator.NOT_EQUALS, 0, "!= 0"),
            FilterTestCase(FilterOperator.IN, ["1", "2"], "IN (1, 2)"),
            FilterTestCase(FilterOperator.NOT_IN, ["1", "2"], "NOT IN (1, 2)"),
        )
        table = self.get_table(name="birth_names")
        for filter_ in filters:
            query_obj = {
                "granularity": None,
                "from_dttm": None,
                "to_dttm": None,
                "groupby": ["gender"],
                "metrics": ["count"],
                "is_timeseries": False,
                "filter": [
                    {"col": "num", "op": filter_.operator, "val": filter_.value}
                ],
                "extras": {},
            }
            sqla_query = table.get_sqla_query(**query_obj)
            sql = table.database.compile_sqla_query(sqla_query.sqla_query)
            if isinstance(filter_.expected, list):
                self.assertTrue(
                    any([candidate in sql for candidate in filter_.expected])
                )
            else:
                self.assertIn(filter_.expected, sql)

    @pytest.mark.usefixtures("load_birth_names_dashboard_with_slices")
    def test_boolean_type_where_operators(self):
        table = self.get_table(name="birth_names")
        db.session.add(
            TableColumn(
                column_name="boolean_gender",
                expression="case when gender = 'boy' then True else False end",
                type="BOOLEAN",
                table=table,
            )
        )
        query_obj = {
            "granularity": None,
            "from_dttm": None,
            "to_dttm": None,
            "groupby": ["boolean_gender"],
            "metrics": ["count"],
            "is_timeseries": False,
            "filter": [
                {
                    "col": "boolean_gender",
                    "op": FilterOperator.IN,
                    "val": ["true", "false"],
                }
            ],
            "extras": {},
        }
        sqla_query = table.get_sqla_query(**query_obj)
        sql = table.database.compile_sqla_query(sqla_query.sqla_query)
        dialect = table.database.get_dialect()
        operand = "(true, false)"
        # override native_boolean=False behavior in MySQLCompiler
        # https://github.com/sqlalchemy/sqlalchemy/blob/master/lib/sqlalchemy/dialects/mysql/base.py
        if not dialect.supports_native_boolean and dialect.name != "mysql":
            operand = "(1, 0)"
        self.assertIn(f"IN {operand}", sql)

    def test_incorrect_jinja_syntax_raises_correct_exception(self):
        query_obj = {
            "granularity": None,
            "from_dttm": None,
            "to_dttm": None,
            "groupby": ["user"],
            "metrics": [],
            "is_timeseries": False,
            "filter": [],
            "extras": {},
        }

        # Table with Jinja callable.
        table = SqlaTable(
            table_name="test_table",
            sql="SELECT '{{ abcd xyz + 1 ASDF }}' as user",
            database=get_example_database(),
        )
        # TODO(villebro): make it work with presto
        if get_example_database().backend != "presto":
            with pytest.raises(QueryObjectValidationError):
                table.get_sqla_query(**query_obj)

    def test_query_format_strip_trailing_semicolon(self):
        query_obj = {
            "granularity": None,
            "from_dttm": None,
            "to_dttm": None,
            "groupby": ["user"],
            "metrics": [],
            "is_timeseries": False,
            "filter": [],
            "extras": {},
        }

        # Table with Jinja callable.
        table = SqlaTable(
            table_name="test_table",
            sql="SELECT * from test_table;",
            database=get_example_database(),
        )
        sqlaq = table.get_sqla_query(**query_obj)
        sql = table.database.compile_sqla_query(sqlaq.sqla_query)
        assert sql[-1] != ";"

    def test_multiple_sql_statements_raises_exception(self):
        base_query_obj = {
            "granularity": None,
            "from_dttm": None,
            "to_dttm": None,
            "groupby": ["grp"],
            "metrics": [],
            "is_timeseries": False,
            "filter": [],
        }

        table = SqlaTable(
            table_name="test_has_extra_cache_keys_table",
            sql="SELECT 'foo' as grp, 1 as num; SELECT 'bar' as grp, 2 as num",
            database=get_example_database(),
        )

        query_obj = dict(**base_query_obj, extras={})
        with pytest.raises(QueryObjectValidationError):
            table.get_sqla_query(**query_obj)

    def test_dml_statement_raises_exception(self):
        base_query_obj = {
            "granularity": None,
            "from_dttm": None,
            "to_dttm": None,
            "groupby": ["grp"],
            "metrics": [],
            "is_timeseries": False,
            "filter": [],
        }

        table = SqlaTable(
            table_name="test_has_extra_cache_keys_table",
            sql="DELETE FROM foo",
            database=get_example_database(),
        )

        query_obj = dict(**base_query_obj, extras={})
        with pytest.raises(QueryObjectValidationError):
            table.get_sqla_query(**query_obj)

    def test_fetch_metadata_for_updated_virtual_table(self):
        table = SqlaTable(
            table_name="updated_sql_table",
            database=get_example_database(),
            sql="select 123 as intcol, 'abc' as strcol, 'abc' as mycase",
        )
        TableColumn(column_name="intcol", type="FLOAT", table=table)
        TableColumn(column_name="oldcol", type="INT", table=table)
        TableColumn(
            column_name="expr",
            expression="case when 1 then 1 else 0 end",
            type="INT",
            table=table,
        )
        TableColumn(
            column_name="mycase",
            expression="case when 1 then 1 else 0 end",
            type="INT",
            table=table,
        )

        # make sure the columns have been mapped properly
        assert len(table.columns) == 4
        table.fetch_metadata()
        # assert that the removed column has been dropped and
        # the physical and calculated columns are present
        assert {col.column_name for col in table.columns} == {
            "intcol",
            "strcol",
            "mycase",
            "expr",
        }
        cols: Dict[str, TableColumn] = {col.column_name: col for col in table.columns}
        # assert that the type for intcol has been updated (asserting CI types)
        backend = get_example_database().backend
        assert VIRTUAL_TABLE_INT_TYPES[backend].match(cols["intcol"].type)
        # assert that the expression has been replaced with the new physical column
        assert cols["mycase"].expression == ""
        assert VIRTUAL_TABLE_STRING_TYPES[backend].match(cols["mycase"].type)
        assert cols["expr"].expression == "case when 1 then 1 else 0 end"

    @patch("superset.models.core.Database.db_engine_spec", BigQueryEngineSpec)
    def test_labels_expected_on_mutated_query(self):
        query_obj = {
            "granularity": None,
            "from_dttm": None,
            "to_dttm": None,
            "groupby": ["user"],
            "metrics": [
                {
                    "expressionType": "SIMPLE",
                    "column": {"column_name": "user"},
                    "aggregate": "COUNT_DISTINCT",
                    "label": "COUNT_DISTINCT(user)",
                }
            ],
            "is_timeseries": False,
            "filter": [],
            "extras": {},
        }

        database = Database(database_name="testdb", sqlalchemy_uri="sqlite://")
        table = SqlaTable(table_name="bq_table", database=database)
        db.session.add(database)
        db.session.add(table)
        db.session.commit()
        sqlaq = table.get_sqla_query(**query_obj)
        assert sqlaq.labels_expected == ["user", "COUNT_DISTINCT(user)"]
        sql = table.database.compile_sqla_query(sqlaq.sqla_query)
        assert "COUNT_DISTINCT_user__00db1" in sql
        db.session.delete(table)
        db.session.delete(database)
        db.session.commit()


@pytest.fixture
def text_column_table():
    with app.app_context():
        table = SqlaTable(
            table_name="text_column_table",
            sql=(
                "SELECT 'foo' as foo "
                "UNION SELECT '' "
                "UNION SELECT NULL "
                "UNION SELECT 'null' "
                "UNION SELECT '\"text in double quotes\"' "
                "UNION SELECT '''text in single quotes''' "
                "UNION SELECT 'double quotes \" in text' "
                "UNION SELECT 'single quotes '' in text' "
            ),
            database=get_example_database(),
        )
        TableColumn(column_name="foo", type="VARCHAR(255)", table=table)
        SqlMetric(metric_name="count", expression="count(*)", table=table)
        yield table


def test_values_for_column_on_text_column(text_column_table):
    # null value, empty string and text should be retrieved
    with_null = text_column_table.values_for_column(column_name="foo", limit=10000)
    assert None in with_null
    assert len(with_null) == 8


def test_filter_on_text_column(text_column_table):
    table = text_column_table
    # null value should be replaced
    result_object = table.query(
        {
            "metrics": ["count"],
            "filter": [{"col": "foo", "val": [NULL_STRING], "op": "IN"}],
            "is_timeseries": False,
        }
    )
    assert result_object.df["count"][0] == 1

    # also accept None value
    result_object = table.query(
        {
            "metrics": ["count"],
            "filter": [{"col": "foo", "val": [None], "op": "IN"}],
            "is_timeseries": False,
        }
    )
    assert result_object.df["count"][0] == 1

    # empty string should be replaced
    result_object = table.query(
        {
            "metrics": ["count"],
            "filter": [{"col": "foo", "val": [EMPTY_STRING], "op": "IN"}],
            "is_timeseries": False,
        }
    )
    assert result_object.df["count"][0] == 1

    # also accept "" string
    result_object = table.query(
        {
            "metrics": ["count"],
            "filter": [{"col": "foo", "val": [""], "op": "IN"}],
            "is_timeseries": False,
        }
    )
    assert result_object.df["count"][0] == 1

    # both replaced
    result_object = table.query(
        {
            "metrics": ["count"],
            "filter": [
                {
                    "col": "foo",
                    "val": [EMPTY_STRING, NULL_STRING, "null", "foo"],
                    "op": "IN",
                }
            ],
            "is_timeseries": False,
        }
    )
    assert result_object.df["count"][0] == 4

    # should filter text in double quotes
    result_object = table.query(
        {
            "metrics": ["count"],
            "filter": [{"col": "foo", "val": ['"text in double quotes"'], "op": "IN",}],
            "is_timeseries": False,
        }
    )
    assert result_object.df["count"][0] == 1

    # should filter text in single quotes
    result_object = table.query(
        {
            "metrics": ["count"],
            "filter": [{"col": "foo", "val": ["'text in single quotes'"], "op": "IN",}],
            "is_timeseries": False,
        }
    )
    assert result_object.df["count"][0] == 1

    # should filter text with double quote
    result_object = table.query(
        {
            "metrics": ["count"],
            "filter": [{"col": "foo", "val": ['double quotes " in text'], "op": "IN",}],
            "is_timeseries": False,
        }
    )
    assert result_object.df["count"][0] == 1

    # should filter text with single quote
    result_object = table.query(
        {
            "metrics": ["count"],
            "filter": [{"col": "foo", "val": ["single quotes ' in text"], "op": "IN",}],
            "is_timeseries": False,
        }
    )
    assert result_object.df["count"][0] == 1


def test_should_generate_closed_and_open_time_filter_range():
    with app.app_context():
        if backend() != "postgresql":
            pytest.skip(f"{backend()} has different dialect for datetime column")

        table = SqlaTable(
            table_name="temporal_column_table",
            sql=(
                "SELECT '2021-12-31'::timestamp as datetime_col "
                "UNION SELECT '2022-01-01'::timestamp "
                "UNION SELECT '2022-03-10'::timestamp "
                "UNION SELECT '2023-01-01'::timestamp "
                "UNION SELECT '2023-03-10'::timestamp "
            ),
            database=get_example_database(),
        )
        TableColumn(
            column_name="datetime_col", type="TIMESTAMP", table=table, is_dttm=True,
        )
        SqlMetric(metric_name="count", expression="count(*)", table=table)
        result_object = table.query(
            {
                "metrics": ["count"],
                "is_timeseries": False,
                "filter": [],
                "from_dttm": datetime(2022, 1, 1),
                "to_dttm": datetime(2023, 1, 1),
                "granularity": "datetime_col",
            }
        )
        """ >>> result_object.query
                SELECT count(*) AS count
                FROM
                  (SELECT '2021-12-31'::timestamp as datetime_col
                   UNION SELECT '2022-01-01'::timestamp
                   UNION SELECT '2022-03-10'::timestamp
                   UNION SELECT '2023-01-01'::timestamp
                   UNION SELECT '2023-03-10'::timestamp) AS virtual_table
                WHERE datetime_col >= TO_TIMESTAMP('2022-01-01 00:00:00.000000', 'YYYY-MM-DD HH24:MI:SS.US')
                  AND datetime_col < TO_TIMESTAMP('2023-01-01 00:00:00.000000', 'YYYY-MM-DD HH24:MI:SS.US')
        """
        assert result_object.df.iloc[0]["count"] == 2


@pytest.mark.parametrize(
    "row,dimension,result",
    [
        (pd.Series({"foo": "abc"}), "foo", "abc"),
        (pd.Series({"bar": True}), "bar", True),
        (pd.Series({"baz": 123}), "baz", 123),
        (pd.Series({"baz": np.int16(123)}), "baz", 123),
        (pd.Series({"baz": np.uint32(123)}), "baz", 123),
        (pd.Series({"baz": np.int64(123)}), "baz", 123),
        (pd.Series({"qux": 123.456}), "qux", 123.456),
        (pd.Series({"qux": np.float32(123.456)}), "qux", 123.45600128173828),
        (pd.Series({"qux": np.float64(123.456)}), "qux", 123.456),
        (pd.Series({"quux": "2021-01-01"}), "quux", "2021-01-01"),
        (
            pd.Series({"quuz": "2021-01-01T00:00:00"}),
            "quuz",
            text("TIME_PARSE('2021-01-01T00:00:00')"),
        ),
    ],
)
def test__normalize_prequery_result_type(
    app_context: Flask,
    mocker: MockFixture,
    row: pd.Series,
    dimension: str,
    result: Any,
) -> None:
    def _convert_dttm(
        target_type: str, dttm: datetime, db_extra: Optional[Dict[str, Any]] = None
    ) -> Optional[str]:
        if target_type.upper() == TemporalType.TIMESTAMP:
            return f"""TIME_PARSE('{dttm.isoformat(timespec="seconds")}')"""

        return None

    table = SqlaTable(table_name="foobar", database=get_example_database())
    mocker.patch.object(table.db_engine_spec, "convert_dttm", new=_convert_dttm)

    columns_by_name = {
        "foo": TableColumn(
            column_name="foo", is_dttm=False, table=table, type="STRING",
        ),
        "bar": TableColumn(
            column_name="bar", is_dttm=False, table=table, type="BOOLEAN",
        ),
        "baz": TableColumn(
            column_name="baz", is_dttm=False, table=table, type="INTEGER",
        ),
        "qux": TableColumn(
            column_name="qux", is_dttm=False, table=table, type="FLOAT",
        ),
        "quux": TableColumn(
            column_name="quuz", is_dttm=True, table=table, type="STRING",
        ),
        "quuz": TableColumn(
            column_name="quux", is_dttm=True, table=table, type="TIMESTAMP",
        ),
    }

    normalized = table._normalize_prequery_result_type(row, dimension, columns_by_name,)

    assert type(normalized) == type(result)

    if isinstance(normalized, TextClause):
        assert str(normalized) == str(result)
    else:
        assert normalized == result