从气流中的 BigQueryOperator 获取结果

Question

我正在尝试BigQueryOperator使用气流获取结果，但找不到方法。我尝试调用成员next()中的方法bq_cursor（在 1.10 中可用），但它返回None. 这就是我尝试做的方式

import datetime
import logging

from airflow import models
from airflow.contrib.operators import bigquery_operator
from airflow.operators import python_operator


yesterday = datetime.datetime.combine(
    datetime.datetime.today() - datetime.timedelta(1),
    datetime.datetime.min.time()
)

def MyChequer(**kwargs):
    big_query_count = bigquery_operator.BigQueryOperator(
        task_id='my_bq_query',
        sql='select count(*) from mydataset.mytable'
    )

    big_query_count.execute(context=kwargs)

    logging.info(big_query_count)
    logging.info(big_query_count.__dict__)
    logging.info(big_query_count.bq_cursor.next())

default_dag_args = {
    'start_date': yesterday,
    'email_on_failure': False,
    'email_on_retry': False,
    'project_id': 'myproject'
}

with models.DAG(
        'bigquery_results_execution',
        # Continue to run DAG once per day
        schedule_interval=datetime.timedelta(days=1),
        default_args=default_dag_args) as dag:

    myoperator = python_operator.PythonOperator(
        task_id='threshold_operator',
        provide_context=True,
        python_callable=MyChequer
    )

    # Define DAG
    myoperator

查看bigquery_hook.py和bigquery_operator.py它似乎是获取结果的唯一可用方法。

Answer 1

每当我需要从 BigQuery 查询获取数据并将其用于某些用途时，我都会使用 BigQuery 挂钩创建自己的运算符。我通常将其称为 BigQueryToXOperator，我们有一堆用于将 BigQuery 数据发送到其他内部系统的运算符。

例如，我有一个 BigQueryToPubSub 运算符，您可能会发现它作为示例非常有用，用于说明如何查询 BigQuery，然后逐行处理结果，将它们发送到 Google PubSub。请考虑以下通用示例代码，了解如何自行执行此操作：

class BigQueryToXOperator(BaseOperator):
    template_fields = ['sql']
    ui_color = '#000000'

    @apply_defaults
    def __init__(
            self,
            sql,
            keys,
            bigquery_conn_id='bigquery_default',
            delegate_to=None,
            *args,
            **kwargs):
        super(BigQueryToXOperator, self).__init__(*args, **kwargs)
        self.sql = sql
        self.keys = keys # A list of keys for the columns in the result set of sql
        self.bigquery_conn_id = bigquery_conn_id
        self.delegate_to = delegate_to


    def execute(self, context):
        """
        Run query and handle results row by row.
        """
        cursor = self._query_bigquery()
        for row in cursor.fetchall():
            # Zip keys and row together because the cursor returns a list of list (not list of dicts)
            row_dict = dumps(dict(zip(self.keys,row))).encode('utf-8')

            # Do what you want with the row...
            handle_row(row_dict)


    def _query_bigquery(self):
        """
        Queries BigQuery and returns a cursor to the results.
        """
        bq = BigQueryHook(bigquery_conn_id=self.bigquery_conn_id,
                          use_legacy_sql=False)
        conn = bq.get_conn()
        cursor = conn.cursor()
        cursor.execute(self.sql)
        return cursor

Answer 2

感谢@kaxil 和@Mike 的回答。我发现了问题。有一种错误（在我看来） BigQueryCursor。作为 的一部分run_with_configuration，running_job_id被返回但从未分配给job_id用于在next方法中提取结果的。一种解决方法（不是很优雅，但如果您不想重新实现所有内容，则很好），是job_id根据running_job_id钩子中的 进行分配，如下所示

big_query_count.execute(context=kwargs)
#workaround
big_query_count.bq_cursor.job_id = big_query_count.bq_cursor.running_job_id
logging.info(big_query_count.bq_cursor.next())