下面这个例子从kafka读取json格式的数据， 然后做窗口聚合后写入es， 可以参考下代码结构， 修改相应数据字段。 这份代码你本地应该是不能运行的
from pyflink.datastream import StreamExecutionEnvironment, TimeCharacteristic
from pyflink.table import StreamTableEnvironment, DataTypes, EnvironmentSettings
from pyflink.table.udf import udf
@udf(input_types=[DataTypes.INT()], result_type=DataTypes.STRING())
def platform_code_to_name(code):
return "mobile" if code == 0 else "pc"
def log_processing():
env = StreamExecutionEnvironment.get_execution_environment()
env.set_stream_time_characteristic(TimeCharacteristic.EventTime)
env_settings = EnvironmentSettings.Builder().use_blink_planner().build()
t_env = StreamTableEnvironment.create(stream_execution_environment=env,
environment_settings=env_settings)
source_ddl = """
CREATE TABLE payment_msg(
createTime VARCHAR,
rt as TO_TIMESTAMP(createTime),
orderId BIGINT,
payAmount DOUBLE,
payPlatform INT,
paySource INT,
WATERMARK FOR rt as rt - INTERVAL '2' SECOND
) WITH (
'connector.type' = 'kafka',
'connector.version' = 'universal',
'connector.topic' = 'payment_msg_2',
'connector.properties.bootstrap.servers' = '0.0.0.0:9092',
'connector.properties.group.id' = 'test_3',
'connector.startup-mode' = 'latest-offset',
'format.type' = 'json'
)
"""
t_env.sql_update(source_ddl)
es_sink_ddl = """
CREATE TABLE es_sink (
platform VARCHAR,
pay_amount DOUBLE,
rowtime TIMESTAMP(3)
) with (
'connector.type' = 'elasticsearch',
'connector.version' = '7',
'connector.hosts' = 'http://localhost:9200',
'connector.index' = 'platform_pay_amount_1',
'connector.document-type' = 'payment',
'update-mode' = 'upsert',
'connector.flush-on-checkpoint' = 'true',
'connector.key-delimiter' = '$',
'connector.key-null-literal' = 'n/a',
'connector.bulk-flush.max-size' = '42mb',
'connector.bulk-flush.max-actions' = '32',
'connector.bulk-flush.interval' = '1000',
'connector.bulk-flush.backoff.delay' = '1000',
'format.type' = 'json'
)
"""
t_env.sql_update(es_sink_ddl)
t_env.register_function('platformcodetoname', platform_code_to_name)
query = """
select platformcodetoname(payPlatform) as platform, sum(payAmount)
as pay_amount, cast(tumble_start(rt, interval '5' seconds) as BIGINT)
as rowtime
from payment_msg
group by tumble(rt, interval '5' seconds), payPlatform
"""
count_result = t_env.sql_query(query)
t_env.create_temporary_view('windowed_values', count_result)
query2 = """
select platform, last_value(pay_amount), rowtime from
windowed_values group by platform, rowtime
"""
final_result = t_env.sql_query(query2)
final_result.execute_insert(table_path='es_sink')
if __name__ == '__main__':
log_processing()
奇怪的不朽琴师 <[email protected]> 于2020年7月15日周三 下午4:40写道：
> Shuiqiang，你好：
>
> hi，能否请求您贡献一下完整的代码的案例，我是初学者，官网的2-from_kafka_to_kafka.py这个没有窗口，我现在想要一个在此基础上有窗口的demo，尝试编了很久也未能解决。我在给这个demo加上窗口功能后总是有各种各样的问题，十分痛苦，如能帮助，感激不尽。
>
>
> 恳请所有看到此封邮件的大佬！
>
>
> 谢谢！
>
>
>
>
> ------------------ 原始邮件 ------------------
> 发件人:
> "user-zh"
> <
> [email protected]>;
> 发送时间: 2020年7月15日(星期三) 中午11:25
> 收件人: "user-zh"<[email protected]>;
>
> 主题: Re: pyflink1.11.0window
>
>
>
> 举个sql例子
> select platformcodetoname(payPlatform) as platform, sum(payAmount) as
> pay_amount, cast(tumble_start(rt, interval '5' seconds) as BIGINT) as
> rowtime
> from payment_msg group by tumble(rt, interval '5' seconds), payPlatform
> 这个query 对每5s的tumble窗口做统计。
>
> 奇怪的不朽琴师 <[email protected]> 于2020年7月15日周三 上午11:10写道：
>
> > Shuiqiang，你好：
> > &nbsp;
> &nbsp;我的目的是每间隔一段时间做一次汇总统计，比如每两秒做一下汇总，请问这个需求我改如何定义window？
> >
> >
> > ------------------&nbsp;原始邮件&nbsp;------------------
> > 发件人:
> >
> "user-zh"
> >
> <
> > [email protected]&gt;;
> > 发送时间:&nbsp;2020年7月15日(星期三) 上午10:51
> > 收件人:&nbsp;"user-zh"<[email protected]&gt;;
> >
> > 主题:&nbsp;Re: pyflink1.11.0window
> >
> >
> >
> > 琴师你好，
> > 异常栈信息org.apache.flink.table.api.ValidationException: A tumble window
> > expects a size value literal.
> > 看起来是接下tumble window定义的代码不太正确吧
> >
> > Best,
> > Shuiqiang
> >
> > 奇怪的不朽琴师 <[email protected]&gt; 于2020年7月15日周三 上午10:27写道：
> >
> > &gt; 你好：
> > &gt; &amp;nbsp; &amp;nbsp;
> > &gt;
> >
> &amp;nbsp;我按着你回复的建议改了source但是会报新的错误，请问这个是因为什么？我想调试一个window一直没有成功，请帮帮我，谢谢。
> > &gt; Traceback (most recent call last):
> > &gt; &amp;nbsp; File "tou.py", line 71, in <module&amp;gt;
> > &gt; &amp;nbsp; &amp;nbsp; from_kafka_to_kafka_demo()
> > &gt; &amp;nbsp; File "tou.py", line 21, in
> from_kafka_to_kafka_demo
> > &gt; &amp;nbsp; &amp;nbsp; .select(" id,&amp;nbsp;
> time1 , time1 ")\
> > &gt; &amp;nbsp; File
> > &gt;
> "/usr/local/lib/python3.7/site-packages/pyflink/table/table.py", line
> > 907,
> > &gt; in select
> > &gt; &amp;nbsp; &amp;nbsp; return
> Table(self._j_table.select(fields),
> > self._t_env)
> > &gt; &amp;nbsp; File
> > "/usr/local/lib/python3.7/site-packages/py4j/java_gateway.py",
> > &gt; line 1286, in __call__
> > &gt; &amp;nbsp; &amp;nbsp; answer, self.gateway_client,
> self.target_id,
> > self.name)
> > &gt; &amp;nbsp; File
> > &gt;
> "/usr/local/lib/python3.7/site-packages/pyflink/util/exceptions.py",
> > line
> > &gt; 147, in deco
> > &gt; &amp;nbsp; &amp;nbsp; return f(*a, **kw)
> > &gt; &amp;nbsp; File
> > "/usr/local/lib/python3.7/site-packages/py4j/protocol.py",
> > &gt; line 328, in get_return_value
> > &gt; &amp;nbsp; &amp;nbsp; format(target_id, ".", name),
> value)
> > &gt; py4j.protocol.Py4JJavaError: An error occurred while calling
> > o26.select.
> > &gt; : org.apache.flink.table.api.ValidationException: A tumble
> window
> > expects
> > &gt; a size value literal.
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.table.operations.utils.AggregateOperationFactory.getAsValueLiteral(AggregateOperationFactory.java:384)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.table.operations.utils.AggregateOperationFactory.validateAndCreateTumbleWindow(AggregateOperationFactory.java:302)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.table.operations.utils.AggregateOperationFactory.createResolvedWindow(AggregateOperationFactory.java:236)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.table.operations.utils.OperationTreeBuilder.windowAggregate(OperationTreeBuilder.java:250)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.table.api.internal.TableImpl$WindowGroupedTableImpl.select(TableImpl.java:794)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.table.api.internal.TableImpl$WindowGroupedTableImpl.select(TableImpl.java:781)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt; sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt; java.lang.reflect.Method.invoke(Method.java:498)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.api.python.shaded.py4j.reflection.MethodInvoker.invoke(MethodInvoker.java:244)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.api.python.shaded.py4j.reflection.ReflectionEngine.invoke(ReflectionEngine.java:357)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.api.python.shaded.py4j.Gateway.invoke(Gateway.java:282)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.api.python.shaded.py4j.commands.AbstractCommand.invokeMethod(AbstractCommand.java:132)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.api.python.shaded.py4j.commands.CallCommand.execute(CallCommand.java:79)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > &gt;
> >
> org.apache.flink.api.python.shaded.py4j.GatewayConnection.run(GatewayConnection.java:238)
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> at
> > java.lang.Thread.run(Thread.java:748)
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt; def register_rides_source(st_env):
> > &gt; &amp;nbsp; &amp;nbsp; source_ddl = \
> > &gt; &amp;nbsp; &amp;nbsp; """
> > &gt; &amp;nbsp; &amp;nbsp; create table source1(
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp;id int,
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp;time1 timestamp,
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp;type string,
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp;WATERMARK FOR
> time1 as time1 -
> > INTERVAL '2' SECOND
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp;) with (
> > &gt; &amp;nbsp; &amp;nbsp; 'connector.type' = 'kafka',
> > &gt; &amp;nbsp; &amp;nbsp; 'update-mode' = 'append',
> > &gt; &amp;nbsp; &amp;nbsp; 'connector.topic' = 'tp1',
> > &gt; &amp;nbsp; &amp;nbsp;
> 'connector.properties.bootstrap.servers' =
> > 'localhost:9092',
> > &gt; &amp;nbsp; &amp;nbsp;
> 'connector.properties.zookeeper.connect' =
> > 'localhost:2181',
> > &gt; &amp;nbsp; &amp;nbsp; 'format.type' = 'json',
> > &gt; &amp;nbsp; &amp;nbsp; 'format.derive-schema' =
> 'true',
> > &gt; &amp;nbsp; &amp;nbsp; 'connector.version' =
> 'universal'
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp;)
> > &gt; &amp;nbsp; &amp;nbsp; """
> > &gt; &amp;nbsp; &amp;nbsp; st_env.sql_update(source_ddl)
> > &gt;
> > &gt; &amp;nbsp; &amp;nbsp;&amp;nbsp;
> > &gt; &amp;nbsp; &amp;nbsp; s_env =
> > &gt; StreamExecutionEnvironment.get_execution_environment()
> > &gt; &amp;nbsp; &amp;nbsp; s_env.set_parallelism(1)
> > &gt;
> > &gt;
> > &gt; &amp;nbsp; &amp;nbsp; st_env =
> StreamTableEnvironment.create(s_env)
> > &gt;
> > &gt;
> > &gt; &amp;nbsp; &amp;nbsp; register_rides_source(st_env)
> > &gt; &amp;nbsp; &amp;nbsp; register_rides_sink(st_env)
> > &gt;
> > &gt;
> > &gt; &amp;nbsp; &amp;nbsp; st_env.from_path("source1")\
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> > &gt; .window(Tumble.over("2.secends").on("time1").alias("w")) \
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> .group_by("w") \
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> .select(" id,&amp;nbsp;
> > time1 , time1 ")\
> > &gt; &amp;nbsp; &amp;nbsp; &amp;nbsp; &amp;nbsp;
> .insert_into("sink1")
> > &gt; &amp;nbsp; &amp;nbsp;&amp;nbsp;
> > &gt; &amp;nbsp; &amp;nbsp;
> st_env.execute("2-from_kafka_to_kafka")
> > &gt;
> > &gt;
> > &gt; 代码如上
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> > &gt;
> ------------------&amp;nbsp;原始邮件&amp;nbsp;------------------
> > &gt; 发件人:
> >
> &gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
> > "user-zh"
> >
> &gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
> > <
> > &gt; [email protected]&amp;gt;;
> > &gt; 发送时间:&amp;nbsp;2020年7月10日(星期五) 上午9:17
> > &gt; 收件人:&amp;nbsp;"user-zh"<[email protected]
> &amp;gt;;
> > &gt;
> > &gt; 主题:&amp;nbsp;Re: pyflink1.11.0window
> > &gt;
> > &gt;
> > &gt;
> > &gt; 琴师你好，
> > &gt;
> > &gt; 你的source ddl里有指定time1为 time attribute吗？
> > &gt; create table source1(
> > &gt;
> >
> &amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;
> id
> > int,
> > &gt;
> >
> &amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;
> > time1 timestamp,
> > &gt;
> >
> &amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;
> type
> > string,
> > &gt;
> >
> &amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;
> > WATERMARK FOR time1 as time1 -
> > &gt; INTERVAL '2' SECOND
> > &gt; ) with (...)
> > &gt;
> > &gt; 奇怪的不朽琴师 <[email protected]&amp;gt; 于2020年7月10日周五
> 上午8:43写道：
> > &gt;
> > &gt; &amp;gt;
> >
> ------------------&amp;amp;nbsp;原始邮件&amp;amp;nbsp;------------------
> > &gt; &amp;gt; 发件人:
> > &gt;
> >
> &amp;gt;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;
> > &gt; "奇怪的不朽琴师"
> > &gt;
> >
> &amp;gt;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;&amp;nbsp;
> > &gt; <
> > &gt; &amp;gt; [email protected]&amp;amp;gt;;
> > &gt; &amp;gt; 发送时间:&amp;amp;nbsp;2020年7月9日(星期四) 下午5:08
> > &gt; &amp;gt; 收件人:&amp;amp;nbsp;"godfrey he"<
> [email protected]
> > &amp;amp;gt;;
> > &gt; &amp;gt;
> > &gt; &amp;gt; 主题:&amp;amp;nbsp;pyflink1.11.0window
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt; 你好：
> > &gt; &amp;gt; &amp;amp;nbsp;
> &amp;amp;nbsp;我在使用pyflink1.11版本时，window开窗仍会报错
> > &gt; &amp;gt; :
> org.apache.flink.table.api.ValidationException: A group
> > window
> > &gt; expects a
> > &gt; &amp;gt; time attribute for grouping in a stream
> environment.
> > &gt; &amp;gt;
> > &gt; &amp;gt; 请问这个问题没有修复么？或者是我使用的方式不对，如果是使用不对，能提供一个正确的案例么？
> > &gt; &amp;gt; 代码如下
> > &gt; &amp;gt; 谢谢
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt; def from_kafka_to_kafka_demo():
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; s_env =
> > &gt; &amp;gt;
> StreamExecutionEnvironment.get_execution_environment()
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> s_env.set_parallelism(1)
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; # use
> blink table planner
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; st_env =
> > StreamTableEnvironment.create(s_env)
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; #
> register source and sink
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> register_rides_source(st_env)
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> register_rides_sink(st_env)
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> st_env.from_path("source1")\
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp; &amp;amp;nbsp;
> > &gt; &amp;gt;
> .window(Tumble.over("1.secends").on("time1").alias("w")) \
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp; &amp;amp;nbsp;
> > .group_by("w") \
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp; &amp;amp;nbsp;
> > .select(" id,&amp;amp;nbsp;
> > &gt; time1 , time1 ")\
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp; &amp;amp;nbsp;
> > .insert_into("sink1")
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> > st_env.execute("2-from_kafka_to_kafka")
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt; def register_rides_source(st_env):
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> source_ddl = \
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; '''
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; create
> table source1(
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp; &amp;amp;nbsp;
> > id int,
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp;time1 timestamp,
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp;type string
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp;) with (
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> 'connector.type' = 'kafka',
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> 'update-mode' = 'append',
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> 'connector.topic' = 'tp1',
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> > 'connector.properties.bootstrap.servers' =
> > &gt; 'localhost:9092'
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp;)
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; '''
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> st_env.sql_update(source_ddl)
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt; def register_rides_sink(st_env):
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; sink_ddl
> = \
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; '''
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; create
> table sink1(
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp; &amp;amp;nbsp;
> > id int,
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp;time1 timestamp,
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp;time2 timestamp
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp;) with (
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> 'connector.type' = 'kafka',
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> 'update-mode' = 'append',
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> 'connector.topic' = 'tp3',
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> > 'connector.properties.bootstrap.servers' =
> > &gt; 'localhost:9092'
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> &amp;amp;nbsp;)
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp; '''
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> st_env.sql_update(sink_ddl)
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt; if __name__ == '__main__':
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
> from_kafka_to_kafka_demo()
> > &gt; &amp;gt;
> > &gt; &amp;gt;
> > &gt; &amp;gt; &amp;amp;nbsp; &amp;amp;nbsp;
