ael Armbrust <mich...@databricks.com>,
Tathagata Das <tathagata.das1...@gmail.com>, "user @spark"
<user@spark.apache.org>
Subject: Re: can we use mapGroupsWithState in raw sql?
This is cool! Looks to me this works too
select data.* from (SELECT max(struct(my_timestamp,*)) as
t;)
>>
>>
>> Thanks,
>> Arun
>>
>> From: Jungtaek Lim <kabh...@gmail.com>
>> Date: Wednesday, April 18, 2018 at 4:54 PM
>> To: Michael Armbrust <mich...@databricks.com>
>> Cc: kant kodali <kanth...@gmail.com>, Arun Iyer <ar
uot;,
> $"my_timestamp")).as("data")).select($"id", $"data.*")
>
>
> Thanks,
> Arun
>
> From: Jungtaek Lim <kabh...@gmail.com>
> Date: Wednesday, April 18, 2018 at 4:54 PM
> To: Michael Armbrust <mich...@databricks.com>
&
Michael Armbrust <mich...@databricks.com>
Cc: kant kodali <kanth...@gmail.com>, Arun Iyer <ar...@apache.org>, Tathagata
Das <tathagata.das1...@gmail.com>, "user @spark" <user@spark.apache.org>
Subject: Re: can we use mapGroupsWithState in raw sql?
Thanks M
tMode(“complete”/“update")….
>>>
>>> Unless the “stream” is already a grouped stream, in which case the above
>>> would not work since the support for multiple aggregate operations is not
>>> there yet.
>>>
>>> Thanks,
>>> Arun
>
ot;amount").writeStream.outputMode(“
>> complete”/“update")….
>>
>> Unless the “stream” is already a grouped stream, in which case the above
>> would not work since the support for multiple aggregate operations is not
>> there yet.
>>
>> Thanks,
>>
ate operations is not
> there yet.
>
> Thanks,
> Arun
>
> From: kant kodali <kanth...@gmail.com>
> Date: Tuesday, April 17, 2018 at 11:41 AM
> To: Tathagata Das <tathagata.das1...@gmail.com>
> Cc: "user @spark" <user@spark.apache.org>
> Subject
erations is not there yet.
Thanks,
Arun
From: kant kodali <kanth...@gmail.com>
Date: Tuesday, April 17, 2018 at 11:41 AM
To: Tathagata Das <tathagata.das1...@gmail.com>
Cc: "user @spark" <user@spark.apache.org>
Subject: Re: can we use mapGroupsWithState in raw sql?
;> wouldn't want to get different result according to the micro-batch, and
>> then you always want to deal with event time window.
>>
>> Thanks,
>> Jungtaek Lim (HeartSaVioR)
>>
>> 2018년 4월 18일 (수) 오전 3:42, kant kodali <kanth...@gmail.com>님이 작성:
>>
>>> Hi TD,
>>>
>>> Thanks for that. The only reason I ask is I don't see any alternative
>>> solution to solve the problem below using raw sql.
>>>
>>>
>>> How to select the max row for every group in spark structured streaming
>>> 2.3.0 without using order by since it requires complete mode or
>>> mapGroupWithState?
>>>
>>> *Input:*
>>>
>>> id | amount | my_timestamp
>>> ---
>>> 1 | 5 | 2018-04-01T01:00:00.000Z
>>> 1 | 10 | 2018-04-01T01:10:00.000Z
>>> 2 | 20 | 2018-04-01T01:20:00.000Z
>>> 2 | 30 | 2018-04-01T01:25:00.000Z
>>> 2 | 40 | 2018-04-01T01:30:00.000Z
>>>
>>> *Expected Output:*
>>>
>>> id | amount | my_timestamp
>>> ---
>>> 1 | 10 | 2018-04-01T01:10:00.000Z
>>> 2 | 40 | 2018-04-01T01:30:00.000Z
>>>
>>> Looking for a streaming solution using either raw sql like
>>> sparkSession.sql("sql
>>> query") or similar to raw sql but not something like mapGroupWithState
>>>
>>> On Mon, Apr 16, 2018 at 8:32 PM, Tathagata Das <
>>> tathagata.das1...@gmail.com> wrote:
>>>
>>>> Unfortunately no. Honestly it does not make sense as for type-aware
>>>> operations like map, mapGroups, etc., you have to provide an actual JVM
>>>> function. That does not fit in with the SQL language structure.
>>>>
>>>> On Mon, Apr 16, 2018 at 7:34 PM, kant kodali <kanth...@gmail.com>
>>>> wrote:
>>>>
>>>>> Hi All,
>>>>>
>>>>> can we use mapGroupsWithState in raw SQL? or is it in the roadmap?
>>>>>
>>>>> Thanks!
>>>>>
>>>>>
>>>>>
>>>>
>>>
>> 1 | 5 | 2018-04-01T01:00:00.000Z
>> 1 | 10 | 2018-04-01T01:10:00.000Z
>> 2 | 20 | 2018-04-01T01:20:00.000Z
>> 2 | 30 | 2018-04-01T01:25:00.000Z
>> 2 | 40 | 2018-04-01T01:30:00.000Z
>>
>> *Expected Output:*
>>
>> id | amount | my_timestamp
>> ---
>> 1 | 10 | 2018-04-01T01:10:00.000Z
>> 2 | 40 | 2018-04-01T01:30:00.000Z
>>
>> Looking for a streaming solution using either raw sql like
>> sparkSession.sql("sql
>> query") or similar to raw sql but not something like mapGroupWithState
>>
>> On Mon, Apr 16, 2018 at 8:32 PM, Tathagata Das <
>> tathagata.das1...@gmail.com> wrote:
>>
>>> Unfortunately no. Honestly it does not make sense as for type-aware
>>> operations like map, mapGroups, etc., you have to provide an actual JVM
>>> function. That does not fit in with the SQL language structure.
>>>
>>> On Mon, Apr 16, 2018 at 7:34 PM, kant kodali <kanth...@gmail.com> wrote:
>>>
>>>> Hi All,
>>>>
>>>> can we use mapGroupsWithState in raw SQL? or is it in the roadmap?
>>>>
>>>> Thanks!
>>>>
>>>>
>>>>
>>>
>>
0:00.000Z
> 2 | 40 | 2018-04-01T01:30:00.000Z
>
> Looking for a streaming solution using either raw sql like
> sparkSession.sql("sql
> query") or similar to raw sql but not something like mapGroupWithState
>
> On Mon, Apr 16, 2018 at 8:32 PM, Tathagata Das <
> tathagata.das1...@gmail.com> wrote:
>
>> Unfortunately no. Honestly it does not make sense as for type-aware
>> operations like map, mapGroups, etc., you have to provide an actual JVM
>> function. That does not fit in with the SQL language structure.
>>
>> On Mon, Apr 16, 2018 at 7:34 PM, kant kodali <kanth...@gmail.com> wrote:
>>
>>> Hi All,
>>>
>>> can we use mapGroupsWithState in raw SQL? or is it in the roadmap?
>>>
>>> Thanks!
>>>
>>>
>>>
>>
>
nction. That does not fit in with the SQL language structure.
>
> On Mon, Apr 16, 2018 at 7:34 PM, kant kodali <kanth...@gmail.com> wrote:
>
>> Hi All,
>>
>> can we use mapGroupsWithState in raw SQL? or is it in the roadmap?
>>
>> Thanks!
>>
>>
>>
>
,
>
> can we use mapGroupsWithState in raw SQL? or is it in the roadmap?
>
> Thanks!
>
>
>
Hi All,
can we use mapGroupsWithState in raw SQL? or is it in the roadmap?
Thanks!
14 matches
Mail list logo