Hi Dongoo,
You are right that plan advisor doesn't have ability to block user request.
Actually, we add a new REST API in the gateway side to validate the users
request is legal or not in our inner branch. Therefore, user can validate
their statements before execution or execute their statement di
Hi Shengkai,
Thank you for introducing an interesting concept of plan advice.
After reviewing the FLIP, I thought that plan advice cannot fully replace
the original intention of the OperationValidator.
Similar to the pluggable ResourceValidator[1] in flink k8s operator, the
OperationValidator aims
Hi Dongwoo,
Thanks for your proposal!
I think the idea about OperationValidator is similar to the plan advice[1],
which can give advice if the statement has problems. I wonder if it would
be more beneficial to develop the concept within the existing
infrastructure, rather than introducing a new c
Hi Martijn,
Thanks for the guidance on making a FLIP.
Before proceeding, I’d like to gather some initial thoughts from the
community to see if we agree on the usefulness of this feature.
Could anyone share their opinions on the custom query operation validator
for Flink SQL Gateway?
Thanks in adv
Hi Dongwoo,
I think this should first have been a FLIP [1], given that it proposes to
make something pluggable with custom interfaces.
Best regards,
Martijn
[1]
https://cwiki.apache.org/confluence/display/FLINK/Flink+Improvement+Proposals
On Thu, Jul 18, 2024 at 2:50 PM Dongwoo Kim wrote:
>
Hi devs,
A month ago, I created a jira issue and pr for a custom query operation
validator in Flink SQL Gateway to block unsafe queries.
Since there has been no feedback yet, I would appreciate it if someone
could review it or provide some feedback.
I'm curious if this feature can be added and if
Hi everyone,
Is there anyone available to review this PR (1) that I opened 1,5 month
ago ? People I've pinged seem to be unavailable at the moment.
Thanks
[1] https://github.com/apache/flink/pull/23443
Best
Etienne
Could anyone help me review the changes？Thank you~
Here is the JIRA: https://issues.apache.org/jira/browse/FLINK-28910
Here is the PR: https://github.com/apache/flink/pull/20542
Please ignore me.
I originally wanted to send it to calcite's dev mail list, but I sent it to
the wrong mail list.
I'm terribly sorry.😥
Jing Zhang 于2022年1月26日周三 14:55写道：
> Hi community,
> My apologies for interrupting.
> Anyone could help to review the pr
> https://github.com/apache/calcite/pull
Hi community,
My apologies for interrupting.
Anyone could help to review the pr
https://github.com/apache/calcite/pull/2606?
Thanks a lot.
CALCITE-4865 is the first sub-task of CALCITE-4864. This Jira aims to
extend existing Table function in order to support Polymorphic Table
Function which is in
Thank you, Kostas, for reviewing this.
Although points 1 and 3 are something which I was planning to address in
the actual implementation, #2 would still be a show stopper.
I'll spend some more time on this and maybe come up with a better way to
achieve the same use case without mixing the two no
Hi Shailesh,
Your solution may fit your use case, but as Dawid mentioned earlier, it makes a
lot of
assumptions about the input.
From a look at your PoC:
1) You assume no late data (you do not drop anything) and no out-of-orderness.
2) You mix the two notions of time (event and processing).
3)
Bump.
On Thu, Mar 22, 2018 at 7:54 PM, Shailesh Jain
wrote:
> To trigger the computations for each batch, I'll have to use the
> processing time timer in the abstract keyed cep operator, right?
>
> The reason why I'm avoiding the watermarks is that it is not possible to
> generate watermarks per
To trigger the computations for each batch, I'll have to use the processing
time timer in the abstract keyed cep operator, right?
The reason why I'm avoiding the watermarks is that it is not possible to
generate watermarks per key.
Thanks for the 'within' remark.
A couple of questions:
1. Given
If you do the buffering you can emit watermark for each such batch (equal to
highest timestamp in such batch). This way you won’t need to sort. CEP library
will do it for you.
The within clause will work in EventTime then.
One more remark also the within clause always work for whole pattern not
Thanks for your reply, Dawid.
I understand that the approach I've tried out is not generic enough, and
would need a lot more thought to be put into w.r.t parallelism
considerations, out of order events, effects on downstream operators etc.
The intention was to do a quick implementation to check th
Hi Shailesh,
Thanks for your interest in the CEP library and sorry for late response. I must
say I am not fun of this approach.
After this change, the Processing time is no longer a processing time, plus it
will work differently in any other place of Flink. It will also not sort the
events etc.
Thanks Aljoscha.
Bump.
I understand everyone would be busy with 1.5.0, but would really appreciate
slight help in unblocking us here.
Thanks,
Shailesh
On Thu, Mar 15, 2018 at 1:47 AM, Aljoscha Krettek
wrote:
> Hi,
>
> I think this should have been sent to the dev mailing list because in the
>
Hi,
I think this should have been sent to the dev mailing list because in the user
mailing list it might disappear among a lot of other mail.
Forwarding...
Best,
Aljoscha
> On 14. Mar 2018, at 06:20, Shailesh Jain wrote:
>
> Hi,
>
> We've been facing issues* w.r.t watermarks not supported p
19 matches
Mail list logo
