> From: Vasiliki Kalavri
> Sent: February 9, 2016 10:54:51am PST
> To: dev@flink.apache.org
> Cc: Martin Junghanns
> Subject: Re: User Feedback
>
> Hi Martin,
>
> thank you for the feedback. Let me try to answer some of your concerns.
>
>
> On 9 February 2016 at 15:35, Martin Neumann
Hi Martin,
thank you for the feedback. Let me try to answer some of your concerns.
On 9 February 2016 at 15:35, Martin Neumann wrote:
> During this year's FOSDEM Martin Junghans and I set together and gathered
> some feedback for the Flink project. It is based on our
Not the DataSet but the JoinDataSet and the CoGroupDataSet do in the form
of an apply function.
On Tue, Feb 9, 2016 at 11:09 AM, Stefano Baghino <
stefano.bagh...@radicalbit.io> wrote:
> Sure, it was just a draft. I agree that filter and mapPartition make sense,
> but coGroup and join don't
This looks like a good design to me :-) The only thing is that it is not
complete. For example, the filter, mapPartition, coGroup and join functions
are missing.
Cheers,
Till
On Tue, Feb 9, 2016 at 1:18 AM, Stefano Baghino <
stefano.bagh...@radicalbit.io> wrote:
> What do you think of
Sure, it was just a draft. I agree that filter and mapPartition make sense,
but coGroup and join don't look like they take a function.
On Tue, Feb 9, 2016 at 10:08 AM, Till Rohrmann wrote:
> This looks like a good design to me :-) The only thing is that it is not
>
Stephan Ewen created FLINK-3376:
---
Summary: Add an illustration of Event Time and Watermarks to the
docsq
Key: FLINK-3376
URL: https://issues.apache.org/jira/browse/FLINK-3376
Project: Flink
Stephan Ewen created FLINK-3375:
---
Summary: Allow Watermark Generation in the Kafka Source
Key: FLINK-3375
URL: https://issues.apache.org/jira/browse/FLINK-3375
Project: Flink
Issue Type:
Ufuk Celebi created FLINK-3374:
--
Summary: CEPITCase testSimplePatternEventTime fails
Key: FLINK-3374
URL: https://issues.apache.org/jira/browse/FLINK-3374
Project: Flink
Issue Type: Test
Jakob Sultan Ericsson created FLINK-3373:
Summary: Using a newer library of Apache HttpClient than 4.2.6
will get class loading problems
Key: FLINK-3373
URL:
Maximilian Michels created FLINK-3378:
-
Summary: Consolidate TestingCluster and FokableFlinkMiniCluster
Key: FLINK-3378
URL: https://issues.apache.org/jira/browse/FLINK-3378
Project: Flink
I would also like to keep this RC. 0.10-SNAPSHOT will be equally stable.
On Tue, Feb 9, 2016 at 2:57 PM, Stephan Ewen wrote:
> I would be in favor of continuing the release candidate.
> The change is more a feature improvement than a bug, and these should come
> before release
During this year's FOSDEM Martin Junghans and I set together and gathered
some feedback for the Flink project. It is based on our personal experience
as well as the feedback and questions from People we taught the system.
This is going to be a longer email therefore I have split things into
I would be in favor of continuing the release candidate.
The change is more a feature improvement than a bug, and these should come
before release candidates (otherwise we never get one done).
The 1.0 release is also very close, so this improvement will be available
anyways pretty soon.
I
I see, thanks for the tip! I'll work on it; meanwhile, I've added some
functions and Scaladoc:
https://github.com/radicalbit/flink/blob/1159-implicit/flink-scala/src/main/scala/org/apache/flink/api/scala/extensions/package.scala
On Tue, Feb 9, 2016 at 12:01 PM, Till Rohrmann
Hey Nick,
I agree that this can be problematic when running multiple jobs on
YARN. Since there is a chance that this might be the last release 0.10
release, I would be OK to cancel the vote for your fix.
Still, let's hear the opinion of others before doing this. What do you think?
– Ufuk
On
I agree with you, but I acknowledge that there may be concerns regarding
the stability of the API. Perhaps the rationale behind the proposal of
Stephan and Till is to provide it as an extension to test how the
developers feel about it. It would be ideal to have a larger feedback from
the
Thanks Ufuk,
+1 for a new RC
2016-02-09 13:49 GMT+01:00 Ufuk Celebi :
> Hey Nick,
>
> I agree that this can be problematic when running multiple jobs on
> YARN. Since there is a chance that this might be the last release 0.10
> release, I would be OK to cancel the vote for your
Thanks for bringing this up Stefano, it would a very welcome addition
indeed.
I like the approach of having extensions through implicits as well. IMHO
though this should be the default
behavior, without the need to add another import.
On Tue, Feb 9, 2016 at 1:29 PM, Stefano Baghino <
Hi Ufuk,
This is nice to have but not a blocker.
So unless we find blocker for the current RC I prefer to continue evaluate
and VOTE current RC.
- Henry
On Tuesday, February 9, 2016, Ufuk Celebi wrote:
> Hey Nick,
>
> I agree that this can be problematic when running
Stephan Ewen created FLINK-3379:
---
Summary: Refactor TimestampExtractor
Key: FLINK-3379
URL: https://issues.apache.org/jira/browse/FLINK-3379
Project: Flink
Issue Type: Improvement
+1 (binding)
Re-ran the tests from the previous RC with the new RC
- Built the code (Linux x64, Scala 2.10, Hadoop 2.3.0)
- Ran all tests
- Checked that there are no cross-version dependencies
- Checked that LICENSE and NOTICE are up to date
- Check planVisualizer util
- ran examples
Given the initial feedback, I would like to keep the RC going as well. The
upcoming 1.0 release will include the fix and if users complain about it for
0.10 we can do another release in the future.
—
+1 for this release.
Forwarding my tests from the last vote.
Additionally, I’ve verified
Hi everyone!
There are two remaining issues right now pending for 1.0 that will cause
breaking API changes in the Streaming API.
1)
[FLINK-3379]
The Timestamp Extractor needs to be changed. That really seems necessary,
based on the user feedback, because a lot of people mentioned that they are
Klou created FLINK-3380:
---
Summary: Unstable Test: JobSubmissionFailsITCase
Key: FLINK-3380
URL: https://issues.apache.org/jira/browse/FLINK-3380
Project: Flink
Issue Type: Bug
Reporter:
Greg Hogan created FLINK-3382:
-
Summary: Improve clarity of object reuse in
ReusingMutableToRegularIteratorWrapper
Key: FLINK-3382
URL: https://issues.apache.org/jira/browse/FLINK-3382
Project: Flink
Hi Till,
I do agree with your point, so much so so that at the time being I'd
suggest to keeping these additions as optional, up to the end-user to
opt-in.
Adding them by default would effectively be an addition to the DataSet API
(despite being separated at a source file level).
I think your
I think that if we make the class part of the default ".api.scala" package
object, we effectively add these methods to DataSet.scala, because they
will be always be available on the data set.
If we want to retain the liberty to not commit to this change now, then we
should probably ask users to
I can elaborate in the project(...) method:
".returns()" is there to supply TypeInformation in cases where the system
cannot determine it. In the case of "project()", the system can perfectly
determine the output type info from the input and the projection.
For just getting a typed result, I
What we could do is to add the implicit class to the package object of
org.apache.flink.api.scala. Since one always has to import this package in
order to have the proper TypeInformations, you wouldn’t have to import the
extension explicitly.
The thing with the API is that you don’t want to break
Hi Nick,
Thanks for bringing up the issue. I believe some people (including me) will
try to run Apache Flink on YARN =)
If more 0.10.x releases needed before 1.0 definitely would love to include
this in.
- Henry
On Tue, Feb 9, 2016 at 9:33 AM, Nick Dimiduk wrote:
> Agree
Maximilian Michels created FLINK-3383:
-
Summary: Separate Maven deployment from CI testing
Key: FLINK-3383
URL: https://issues.apache.org/jira/browse/FLINK-3383
Project: Flink
Issue
Agree it's not a blocker. I can manage my own build with this patch if
necessary. Just thought I might not be the only one running streaming on
yarn.
On Tuesday, February 9, 2016, Henry Saputra wrote:
> Hi Ufuk,
>
> This is nice to have but not a blocker.
>
> So unless
32 matches
Mail list logo