Source code for airflow.providers.apache.hive.transfers.mssql_to_hive
#
# 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.
"""This module contains an operator to move data from MSSQL to Hive."""
from __future__ import annotations
import csv
from collections.abc import Sequence
from tempfile import NamedTemporaryFile
from typing import TYPE_CHECKING
import pymssql
from airflow.models import BaseOperator
from airflow.providers.apache.hive.hooks.hive import HiveCliHook
from airflow.providers.microsoft.mssql.hooks.mssql import MsSqlHook
if TYPE_CHECKING:
from airflow.utils.context import Context
[docs]class MsSqlToHiveOperator(BaseOperator):
"""
Moves data from Microsoft SQL Server to Hive.
The operator runs your query against Microsoft SQL Server, stores
the file locally before loading it into a Hive table. If the
``create`` or ``recreate`` arguments are set to ``True``, a
``CREATE TABLE`` and ``DROP TABLE`` statements are generated.
Hive data types are inferred from the cursor's metadata.
Note that the table generated in Hive uses ``STORED AS textfile``
which isn't the most efficient serialization format. If a
large amount of data is loaded and/or if the table gets
queried considerably, you may want to use this operator only to
stage the data into a temporary table before loading it into its
final destination using a ``HiveOperator``.
:param sql: SQL query to execute against the Microsoft SQL Server
database. (templated)
:param hive_table: target Hive table, use dot notation to target a specific
database. (templated)
:param create: whether to create the table if it doesn't exist
:param recreate: whether to drop and recreate the table at every execution
:param partition: target partition as a dict of partition columns and
values. (templated)
:param delimiter: field delimiter in the file
:param mssql_conn_id: source Microsoft SQL Server connection
:param hive_cli_conn_id: Reference to the
:ref:`Hive CLI connection id <howto/connection:hive_cli>`.
:param hive_auth: optional authentication option passed for the Hive connection
:param tblproperties: TBLPROPERTIES of the hive table being created
"""
[docs] template_fields: Sequence[str] = ("sql", "partition", "hive_table")
[docs] template_ext: Sequence[str] = (".sql",)
[docs] template_fields_renderers = {"sql": "tsql"}
def __init__(
self,
*,
sql: str,
hive_table: str,
create: bool = True,
recreate: bool = False,
partition: dict | None = None,
delimiter: str = chr(1),
mssql_conn_id: str = "mssql_default",
hive_cli_conn_id: str = "hive_cli_default",
hive_auth: str | None = None,
tblproperties: dict | None = None,
**kwargs,
) -> None:
super().__init__(**kwargs)
self.sql = sql
self.hive_table = hive_table
self.partition = partition
self.create = create
self.recreate = recreate
self.delimiter = delimiter
self.mssql_conn_id = mssql_conn_id
self.hive_cli_conn_id = hive_cli_conn_id
self.partition = partition or {}
self.tblproperties = tblproperties
self.hive_auth = hive_auth
@classmethod
[docs] def type_map(cls, mssql_type: int) -> str:
"""Map MsSQL type to Hive type."""
map_dict = {
pymssql.BINARY.value: "INT",
pymssql.DECIMAL.value: "FLOAT",
pymssql.NUMBER.value: "INT",
}
return map_dict.get(mssql_type, "STRING")
[docs] def execute(self, context: Context):
mssql = MsSqlHook(mssql_conn_id=self.mssql_conn_id)
self.log.info("Dumping Microsoft SQL Server query results to local file")
with mssql.get_conn() as conn:
with conn.cursor() as cursor:
cursor.execute(self.sql)
with NamedTemporaryFile(mode="w", encoding="utf-8") as tmp_file:
csv_writer = csv.writer(tmp_file, delimiter=self.delimiter)
field_dict = {}
for col_count, field in enumerate(cursor.description, start=1):
col_position = f"Column{col_count}"
field_dict[col_position if field[0] == "" else field[0]] = self.type_map(field[1])
csv_writer.writerows(cursor)
tmp_file.flush()
hive = HiveCliHook(hive_cli_conn_id=self.hive_cli_conn_id, auth=self.hive_auth)
self.log.info("Loading file into Hive")
hive.load_file(
tmp_file.name,
self.hive_table,
field_dict=field_dict,
create=self.create,
partition=self.partition,
delimiter=self.delimiter,
recreate=self.recreate,
tblproperties=self.tblproperties,
)