Re: Question about Commit Policy

But thats very long, and together with the issue tag I almost always have I lose a lot of my precious 80 characters. On Tue, Jan 27, 2015 at 1:17 PM, Fabian Hueske fhue...@gmail.com wrote: I know I argued against enforcing commit tags, but how about we make two tags mandatory, i.e.,

RE: Gather a distributed dataset

Hi everyone, I just wanted to give you the pointer FLINK-1038 https://github.com/apache/flink/pull/94 This is an output format that can send DataSet contents via Java RMI to, e.g., the driver. I am currently using it a lot and it seems to scale pretty well. Cheers, Sebastian -Original

[jira] [Created] (FLINK-1458) Interfaces and abstract classes are not valid types

John Sandiford created FLINK-1458: - Summary: Interfaces and abstract classes are not valid types Key: FLINK-1458 URL: https://issues.apache.org/jira/browse/FLINK-1458 Project: Flink Issue

Re: flink build issue

Hi, thank you for trying out Flink. I'm sorry that you ran into this issue. Flink does not have support for Hadoop YARN before 2.2.0. The reason for that is that Hadoop has changed the YARN APIs with the 2.2.0 release. (the pre 2.2.0 APIs are marked as alpha) If you take a closer look into the

flink build issue

Hello : I want build flink 0.8 on yarn ,but we use version : 2.0.0-cdh4.3.0 ,it failed how can i build it . command : mvn clean install -DskipTests -Pvendor-repos -Dhadoop.version=2.0.0-cdh4.3.0 The document say , not support before hadoop 2.2.0 ,but i want to run on yarn like spark on yarn

Re: Warning message in Scala type analysis

Ok, if it is only a testing dependency it should be alright. The problem seems to be that some of the Joda classes have annotations that reside in joda-convert. The java compiler seems to be OK with missing annotations while the Scala compiler gives a warning. At least that's how I understood it

Re: Question about Commit Policy

Originally I was going to support Fabian's suggestion - I still like it, but I have to admit that it can become a slight overkill like in de8e066. [1] [1] https://git-wip-us.apache.org/repos/asf/flink/commit/de8e066 On Tue, Jan 27, 2015 at 2:03 PM, Max Michels m...@data-artisans.com wrote: I

Re: Question about Commit Policy

I did not propose to pick the 5 most suitable tags. ;-) Just to mark commits that break or extend the API. If [api-breaking] and [api-extending] is considered to waste too much space, I would also be happy with [api-] and [api+] (or any other meaningful marker). 2015-01-28 2:37 GMT+01:00 Márton

[jira] [Created] (FLINK-1457) RAT check fails on Windows

Sebastian Kruse created FLINK-1457: -- Summary: RAT check fails on Windows Key: FLINK-1457 URL: https://issues.apache.org/jira/browse/FLINK-1457 Project: Flink Issue Type: Bug

Re: Tagging Flink classes with InterfaceAudience and InterfaceStability

I don't get the difference between Private and LimitedPrivate, but otherwise seems like quite a nice idea. It will be also good if we can agree upon what these tags actually mean and add this meaning to the documentation. 2015-01-27 15:46 GMT+01:00 Robert Metzger rmetz...@apache.org: Hi,

Re: Implementing a list accumulator

Thank you for your help, Ufuk and Stephan. I made some changes to immediately serialize the stored objects. On Thu, Jan 22, 2015 at 2:58 AM, Stephan Ewen se...@apache.org wrote: True, that is tricky. The user code does not necessarily respect the non-reuse mode. That may be true for any user

Re: API behavior with data sinks (lazy) and eager operations

Let's make it clear that count/collection type of actions execute the plan up till that point (including the data sinks). From a user perspective, this seems most logic to me. The user might even rely on the data generated by the sinks. On Mon, Jan 19, 2015 at 11:46 AM, Fabian Hueske