hdt-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mirko Kämpf <mirko.kae...@gmail.com>
Subject Re: Hadoop-Eclipse project contribution
Date Wed, 03 Jul 2013 16:27:09 GMT
Hi Chris,

I just created an account @ reviews.apache.org.
How would the process for the review and code merge be now?

Srimanth would upload the contribution to a certain branch in the
incubator-hdt project.
>From there we start discussion and merge it patch after patch with the
main project, right?

Best wishes
Mirko




2013/7/3 Mattmann, Chris A (398J) <chris.a.mattmann@jpl.nasa.gov>:
> Hi Mirko,
>
> I would advise against review at Github. Review should be done here
> on the Apache lists where the HDT project exists. Why don't we use
> Review Board?
>
> http://reviews.apache.org/
>
> Cheers,
> Chris
>
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Chris Mattmann, Ph.D.
> Senior Computer Scientist
> NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
> Office: 171-266B, Mailstop: 171-246
> Email: chris.a.mattmann@nasa.gov
> WWW:  http://sunset.usc.edu/~mattmann/
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Adjunct Assistant Professor, Computer Science Department
> University of Southern California, Los Angeles, CA 90089 USA
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
>
>
>
>
>
> -----Original Message-----
> From: Mirko Kämpf <mirko.kaempf@gmail.com>
> Reply-To: "dev@hdt.incubator.apache.org" <dev@hdt.incubator.apache.org>
> Date: Wednesday, July 3, 2013 12:44 AM
> To: "dev@hdt.incubator.apache.org" <dev@hdt.incubator.apache.org>
> Subject: Re: Hadoop-Eclipse project contribution
>
>>Hi Srimanth,
>>
>>I am not sure if I get it right. The contribution is quite big, so
>>small patches will be
>>inconvinient and you have no chance for doing it in this way. But the ASF
>>roles
>>are against a large "one block" contribution.
>>
>>PLEASE CORRECT ME IF I AM WRONG.
>>
>>But in order do fit into the process I suggest to upload
>>your contribution to a github repository. From there we can
>>start to review and merge the code patch by patch.
>>I voluntere in this process.
>>
>>Thanks,
>>
>>Mirko
>>
>>
>>2013/7/3 Srimanth Gunturi <srimanth@gmail.com>:
>>> Hi Suresh,
>>> Sorry for the confusion. I meant to request for committer ability, not
>>> contributor.
>>> I am currently a committer with the Apache Ambari project.
>>>
>>> Imho, this work will require revision control before it can be made
>>> workable. With patches, I would have to do this privately, and cannot
>>> collaborate with anyone in the community. That was the reasoning behind
>>>the
>>> request.
>>>
>>> In any case, the size of contribution will depend on further discussion
>>> with committers on how to structure the code.
>>> I am hoping to start the discussion soon.
>>> Best regards,
>>> Srimanth
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> On Tue, Jun 18, 2013 at 7:32 PM, Suresh Marru <smarru@apache.org> wrote:
>>>
>>>> On Jun 18, 2013, at 1:24 PM, Srimanth Gunturi <srimanth@gmail.com>
>>>>wrote:
>>>>
>>>> > Hi Adam, Chris,
>>>> > I will take time to look into HDT code.
>>>> > In the meantime I have created Epic HDT-32: Merge the code base of
>>>> > Hadoop-Eclipse project into
>>>> > HDT<https://issues.apache.org/jira/browse/HDT-32> to
>>>> > track the merging process.
>>>> > It contains broad subtasks which need to be accomplished. Each of
>>>>them
>>>> will
>>>> > need good amount of discussions.
>>>> >
>>>> > I suggest the creation of a stream where the merge process can be
>>>>worked
>>>> on.
>>>> > I would like to request contributor access to work on the merging
>>>> process.
>>>>
>>>> Hi Srimanth,
>>>>
>>>> As you already know from being a committer on Ambari, there is no
>>>>formal
>>>> access for contributors. The asf roles [1] describe these in detail.
>>>>All
>>>> contributors have to submit the code as patches to JIRA issues. So
>>>>kindly
>>>> work with patches for now. Bulk contributions like this are favorably
>>>> considered and quickly get voted in as committers, but thats
>>>>subjective and
>>>> also there is a latency with such voting (especially for incubating
>>>> projects). So i suggest you not to wait and start sending with the
>>>>merge
>>>> and keep turning in patches. It will be little inconvenient, but
>>>>patches,
>>>> reviews and acceptance are the bets way to grab others attention and
>>>> interest.
>>>>
>>>> Suresh
>>>> [1] - http://www.apache.org/foundation/how-it-works.html#roles
>>>>
>>>> > The first goal would be to get to a point where the base
>>>>functionalities
>>>> of
>>>> > both projects are preserved for the community.
>>>> >
>>>> > With regards to the JIRA model, I am thinking we need 2 more
>>>>components
>>>> > under HDT - 'Common' and 'ZooKeeper'.
>>>> > 'Common' would be for issues with base code common to all services.
>>>> > Regards,
>>>> > Srimanth
>>>> >
>>>> >
>>>> >
>>>> >
>>>> > On Tue, Jun 18, 2013 at 8:17 AM, Suresh Marru <smarru@apache.org>
>>>>wrote:
>>>> >
>>>> >> On Jun 18, 2013, at 1:03 AM, "Mattmann, Chris A (398J)" <
>>>> >> chris.a.mattmann@jpl.nasa.gov> wrote:
>>>> >>
>>>> >>> Hey Adam,
>>>> >>>
>>>> >>> Suggestion: Have Srimanth file some JIRA tickets, and incrementally
>>>> >>> add the contribution, working with HDT committers. Srimanth
-- do
>>>> >>> you have an ICLA on file?
>>>> >>
>>>> >> Srimanth seem to be a committer and has ICLA on file. So should
be
>>>>ready
>>>> >> to proceed to JIRA's and contributions.
>>>> >>
>>>> >> Suresh
>>>> >>
>>>> >>>
>>>> >>> Cheers,
>>>> >>> Chris
>>>> >>>
>>>> >>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>> >>> Chris Mattmann, Ph.D.
>>>> >>> Senior Computer Scientist
>>>> >>> NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
>>>> >>> Office: 171-266B, Mailstop: 171-246
>>>> >>> Email: chris.a.mattmann@nasa.gov
>>>> >>> WWW:  http://sunset.usc.edu/~mattmann/
>>>> >>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>> >>> Adjunct Assistant Professor, Computer Science Department
>>>> >>> University of Southern California, Los Angeles, CA 90089 USA
>>>> >>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>> >>>
>>>> >>>
>>>> >>>
>>>> >>>
>>>> >>>
>>>> >>>
>>>> >>> -----Original Message-----
>>>> >>> From: Adam Berry <adamb@apache.org>
>>>> >>> Reply-To: "dev@hdt.incubator.apache.org"
>>>><dev@hdt.incubator.apache.org
>>>> >
>>>> >>> Date: Monday, June 17, 2013 9:31 PM
>>>> >>> To: "dev@hdt.incubator.apache.org" <dev@hdt.incubator.apache.org>
>>>> >>> Subject: Re: Hadoop-Eclipse project contribution
>>>> >>>
>>>> >>>> Hi Srimanth,
>>>> >>>>
>>>> >>>> Welcome!
>>>> >>>>
>>>> >>>> This looks great. I think this work is a really good fit
for the
>>>>HDT
>>>> and
>>>> >>>> its goals. In the next couple of days I will dive a little
deeper
>>>>into
>>>> >> the
>>>> >>>> code.
>>>> >>>>
>>>> >>>> I'm not sure what the process is for a larger contribution
like
>>>>this.
>>>> >>>> Mentors, what are the options for proceeding here.
>>>> >>>>
>>>> >>>> In the mean time Srimant, please take a look around, and
let us
>>>>know
>>>> if
>>>> >>>> you
>>>> >>>> have any more questions.
>>>> >>>>
>>>> >>>> Cheers,
>>>> >>>> Adam
>>>> >>>>
>>>> >>>>
>>>> >>>> On Mon, Jun 17, 2013 at 2:25 PM, Srimanth Gunturi
>>>> >>>> <srimanth@gmail.com>wrote:
>>>> >>>>
>>>> >>>>> Hi Mirko,
>>>> >>>>> The update site URL is to be put into Eclipse's install
new
>>>>software
>>>> /
>>>> >>>>> update site dialogs.
>>>> >>>>> There are files underneath there that Eclipse knows
relative
>>>>location
>>>> >>>>> of.
>>>> >>>>>
>>>> >>>>> I have removed the blank page, so you can see the contents
too.
>>>> >>>>> Hope that helps.
>>>> >>>>> Regards,
>>>> >>>>> Srimanth
>>>> >>>>>
>>>> >>>>>
>>>> >>>>>
>>>> >>>>>
>>>> >>>>> On Mon, Jun 17, 2013 at 12:09 PM, Mirko Kämpf <
>>>> mirko.kaempf@gmail.com>
>>>> >>>>> wrote:
>>>> >>>>>> Hello Srimnath,
>>>> >>>>>>
>>>> >>>>>> I think this sound great. I would realy like to
see your project
>>>> >>>>> merged
>>>> >>>>> in
>>>> >>>>>> to the HDT project.
>>>> >>>>>>
>>>> >>>>>> I just tried to install the plugins but the update
site :
>>>> >>>>>> http://people.apache.org/~srimanth/hadoop-eclipse/update-site
>>>> >>>>>> is empty?
>>>> >>>>>>
>>>> >>>>>> Do you have another URL for a first test?
>>>> >>>>>>
>>>> >>>>>> Thanks and best wishes
>>>> >>>>>> Mirko
>>>> >>>>>>
>>>> >>>>>>
>>>> >>>>>>
>>>> >>>>>> 2013/6/17 Srimanth Gunturi <srimanth@gmail.com>:
>>>> >>>>>>> Hello,
>>>> >>>>>>> My name is Srimanth, and am the creator of the
Hadoop-Eclipse
>>>> project
>>>> >>>>>>> (http://people.apache.org/~srimanth/hadoop-eclipse/)
released
>>>> under
>>>> >>>>>>> the Apache License.
>>>> >>>>>>> The project currently provides a HDFS Explorer
written on top
>>>>of
>>>> >>>>>>> Eclipse File System (EFS), and a ZooKeeper Explorer.
>>>> >>>>>>>
>>>> >>>>>>> During the creation of my project, I was unaware
of the
>>>>parallel
>>>> >>>>>>> creation of the Hadoop Development Tools Incubator
project in
>>>> Apache.
>>>> >>>>>>> I started the project to address some of the
deficiencies
>>>>found in
>>>> >>>>> the
>>>> >>>>>>> Eclipse functionality provided in Hadoop at
that point.
>>>> >>>>>>> One of them being that the HDFS/MR clients were
compiled
>>>>directly
>>>> >>>>> into
>>>> >>>>>>> the plugins - which meant that only one version
of the Eclipse
>>>> plugin
>>>> >>>>>>> could be active at any one point. Basically
it was not
>>>>allowing a
>>>> >>>>>>> client to connect to multiple versions of Hadoop
servers.
>>>> >>>>>>> Another was the HDFS functionality provided.
Eclipse provides
>>>>rich
>>>> >>>>>>> file system interaction which was not being
utilized. For
>>>>example,
>>>> >>>>>>> right-clicking on a HDFS file/folder was not
showing the
>>>>regular
>>>> >>>>>>> actions one would find (like new folder file/folder
actions).
>>>> >>>>>>>
>>>> >>>>>>> In Hadoop-Eclipse project I rewrote the HDFS
functionality on
>>>>top
>>>> of
>>>> >>>>>>> EFS so that we get for free all the rich functionality
that
>>>>Eclipse
>>>> >>>>>>> provides, like actions, drag&drop, Team
synchronizations,
>>>> >>>>> permissions,
>>>> >>>>>>> etc. You can also access HDFS just like any
Eclipse project -
>>>>write
>>>> >>>>>>> scripts, etc. This is like basic functionality
which can be
>>>>easily
>>>> >>>>>>> expanded upon. It is important to note here
that supporting new
>>>> >>>>>>> versions of HDFS is as easy as creating a fragment-plugin
and
>>>> >>>>>>> extending 1 base class.
>>>> >>>>>>> I have also implemented a ZooKeeper client which
provides basic
>>>> >>>>> client
>>>> >>>>>>> functionality. This helps in keeping an abstract
service
>>>>oriented
>>>> >>>>> view
>>>> >>>>>>> in underlying internals.
>>>> >>>>>>>
>>>> >>>>>>> If the community is interested, I would like
join the group as
>>>>a
>>>> >>>>>>> contributor and contribute my code to the HDT
project. I would
>>>>like
>>>> >>>>> to
>>>> >>>>>>> join our efforts to provide better tools in
the Hadoop
>>>>environment.
>>>> >>>>>>> Please let me know of any questions you have
regarding
>>>> functionality,
>>>> >>>>>>> code base, etc.
>>>> >>>>>>> Best regards,
>>>> >>>>>>> Srimanth
>>>> >>>>>
>>>> >>>
>>>> >>
>>>> >>
>>>>
>>>>
>

Mime
View raw message