You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hama.apache.org by Chanwit Kaewkasi <ch...@gmail.com> on 2009/01/06 05:34:13 UTC
Ivy for jar dependency management
Hello,
I'm thinking of using Ivy (http://ant.apache.org/ivy/) to help us
manage .jar files, as currently a lot of .jars have been put into
trunk/lib directory (including findbug stuff).
It would be better if we can only have source files in the repo, and
all jar are automatically resolved and downloaded when the project is
built.
What do you think? (If you all agree, I'll start learning Ivy configuration).
Regards,
Chanwit
--
Chanwit Kaewkasi
PhD Candidate,
Centre for Novel Computing
School of Computer Science
The University of Manchester
Oxford Road
Manchester
M13 9PL, UK
Re: Ivy for jar dependency management
Posted by "Edward J. Yoon" <ed...@apache.org>.
Cool.
On Wed, Feb 18, 2009 at 11:43 PM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
> I've got an external branch for testing here.
>
> http://github.com/chanwit/hama/tree/master
>
> Will notify you again when it's ready (and before getting merged).
>
> Chanwit
>
> 2009/2/17 Edward J. Yoon <ed...@apache.org>:
>> Hi chanwit, How goes it?
>>
>> On Mon, Jan 19, 2009 at 9:59 AM, Edward J. Yoon <ed...@apache.org> wrote:
>>>> It seems there is no maven repository for HBase or Hadoop that I can use.
>>>
>>> IMO, It would be nice If we can handle them (hadoop/hbase) manually
>>> and use Ivy for other dependencies at same time.
>>>
>>>> And after checking HBase artefact on Hudson, I've found that it
>>>> contains all dependency jars.
>>>
>>> Good idea. BTW, then how can we set up dev environment?
>>>
>>> On Sun, Jan 18, 2009 at 1:28 PM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
>>>> Some updates.
>>>>
>>>
>>>> And after checking HBase artefact on Hudson, I've found that it
>>>> contains all dependency jars.
>>>> I think I may write a script to fetch it to prepare the build
>>>> environment rather than using ivy.
>>>>
>>>> - or -
>>>>
>>>> I may write a script to build a private maven repo for HBase and
>>>> Hadoop, then we can use ivy.
>>>>
>>>> So the concept are still as follows:
>>>> - SVN may contain only "src" and some other documents.
>>>> - HBase will be downloaded from Hudson into "lib". (We probably needs
>>>> different configuration for "dev" and "production")
>>>> - Findbug may be downloaded to run separately.
>>>>
>>>> Regards,
>>>>
>>>> Chanwit
>>>>
>>>> 2009/1/7 Ian Holsman <li...@holsman.net>:
>>>>> Edward J. Yoon wrote:
>>>>>>
>>>>>> That's great.
>>>>>>
>>>>>
>>>>> It will make life easier
>>>>> +1
>>>>>
>>>>>> On Wed, Jan 7, 2009 at 6:31 AM, Chanwit Kaewkasi <ch...@gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>>
>>>>>>> Some more update.
>>>>>>> I've just found that Hadoop's using Ivy in its build script as well.
>>>>>>> So it's not that hard to follow:
>>>>>>>
>>>>>>> http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
>>>>>>>
>>>>>>> Chanwit
>>>>>>>
>>>>>>> 2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
>>>>>>>
>>>>>>>>
>>>>>>>> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
>>>>>>>>
>>>>>>>>>
>>>>>>>>> I like this idea. +1
>>>>>>>>>
>>>>>>>>> Then, We don't need to update periodically?
>>>>>>>>>
>>>>>>>>
>>>>>>>> Yes, I think so. Just change the XML configuration, and rebuild, then
>>>>>>>> it should resolve a new set of jars for us without putting them into
>>>>>>>> the repo. Let me go through its document more.
>>>>>>>>
>>>>>>>> Chanwit
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Chanwit Kaewkasi
>>>>>>> PhD Candidate,
>>>>>>> Centre for Novel Computing
>>>>>>> School of Computer Science
>>>>>>> The University of Manchester
>>>>>>> Oxford Road
>>>>>>> Manchester
>>>>>>> M13 9PL, UK
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Chanwit Kaewkasi
>>>> PhD Candidate,
>>>> Centre for Novel Computing
>>>> School of Computer Science
>>>> The University of Manchester
>>>> Oxford Road
>>>> Manchester
>>>> M13 9PL, UK
>>>>
>>>
>>>
>>>
>>> --
>>> Best Regards, Edward J. Yoon @ NHN, corp.
>>> edwardyoon@apache.org
>>> http://blog.udanax.org
>>>
>>
>>
>>
>> --
>> Best Regards, Edward J. Yoon
>> edwardyoon@apache.org
>> http://blog.udanax.org
>>
>
>
>
> --
> Chanwit Kaewkasi
> PhD Candidate,
> Centre for Novel Computing
> School of Computer Science
> The University of Manchester
> Oxford Road
> Manchester
> M13 9PL, UK
>
--
Best Regards, Edward J. Yoon
edwardyoon@apache.org
http://blog.udanax.org
Re: Ivy for jar dependency management
Posted by Chanwit Kaewkasi <ch...@gmail.com>.
Hi,
2009/2/19 pi song <pi...@gmail.com>:
> Hi,
> From my understanding, Ivy also relies on maven repository. Then why don't
> you go for Maven2 directly?
>
> Do you have things like parser generator or any kind of code generator in
> your compile process?
I was thinking about using Ivy instead of Maven because I just want to
keep thing simple.
Now, I'm writing a script to fetch HBase directly.
No Ivy or Maven anymore as there is no maven repo for HBase at the moment.
Best,
Chanwit
> Pi
>
> On Thu, Feb 19, 2009 at 1:43 AM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
>
>> I've got an external branch for testing here.
>>
>> http://github.com/chanwit/hama/tree/master
>>
>> Will notify you again when it's ready (and before getting merged).
>>
>> Chanwit
>>
>> 2009/2/17 Edward J. Yoon <ed...@apache.org>:
>> > Hi chanwit, How goes it?
>> >
>> > On Mon, Jan 19, 2009 at 9:59 AM, Edward J. Yoon <ed...@apache.org>
>> wrote:
>> >>> It seems there is no maven repository for HBase or Hadoop that I can
>> use.
>> >>
>> >> IMO, It would be nice If we can handle them (hadoop/hbase) manually
>> >> and use Ivy for other dependencies at same time.
>> >>
>> >>> And after checking HBase artefact on Hudson, I've found that it
>> >>> contains all dependency jars.
>> >>
>> >> Good idea. BTW, then how can we set up dev environment?
>> >>
>> >> On Sun, Jan 18, 2009 at 1:28 PM, Chanwit Kaewkasi <ch...@gmail.com>
>> wrote:
>> >>> Some updates.
>> >>>
>> >>
>> >>> And after checking HBase artefact on Hudson, I've found that it
>> >>> contains all dependency jars.
>> >>> I think I may write a script to fetch it to prepare the build
>> >>> environment rather than using ivy.
>> >>>
>> >>> - or -
>> >>>
>> >>> I may write a script to build a private maven repo for HBase and
>> >>> Hadoop, then we can use ivy.
>> >>>
>> >>> So the concept are still as follows:
>> >>> - SVN may contain only "src" and some other documents.
>> >>> - HBase will be downloaded from Hudson into "lib". (We probably needs
>> >>> different configuration for "dev" and "production")
>> >>> - Findbug may be downloaded to run separately.
>> >>>
>> >>> Regards,
>> >>>
>> >>> Chanwit
>> >>>
>> >>> 2009/1/7 Ian Holsman <li...@holsman.net>:
>> >>>> Edward J. Yoon wrote:
>> >>>>>
>> >>>>> That's great.
>> >>>>>
>> >>>>
>> >>>> It will make life easier
>> >>>> +1
>> >>>>
>> >>>>> On Wed, Jan 7, 2009 at 6:31 AM, Chanwit Kaewkasi <ch...@gmail.com>
>> >>>>> wrote:
>> >>>>>
>> >>>>>>
>> >>>>>> Some more update.
>> >>>>>> I've just found that Hadoop's using Ivy in its build script as well.
>> >>>>>> So it's not that hard to follow:
>> >>>>>>
>> >>>>>>
>> http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
>> >>>>>>
>> >>>>>> Chanwit
>> >>>>>>
>> >>>>>> 2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
>> >>>>>>
>> >>>>>>>
>> >>>>>>> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
>> >>>>>>>
>> >>>>>>>>
>> >>>>>>>> I like this idea. +1
>> >>>>>>>>
>> >>>>>>>> Then, We don't need to update periodically?
>> >>>>>>>>
>> >>>>>>>
>> >>>>>>> Yes, I think so. Just change the XML configuration, and rebuild,
>> then
>> >>>>>>> it should resolve a new set of jars for us without putting them
>> into
>> >>>>>>> the repo. Let me go through its document more.
>> >>>>>>>
>> >>>>>>> Chanwit
>> >>>>>>>
>> >>>>>>>
>> >>>>>>
>> >>>>>> --
>> >>>>>> Chanwit Kaewkasi
>> >>>>>> PhD Candidate,
>> >>>>>> Centre for Novel Computing
>> >>>>>> School of Computer Science
>> >>>>>> The University of Manchester
>> >>>>>> Oxford Road
>> >>>>>> Manchester
>> >>>>>> M13 9PL, UK
>> >>>>>>
>> >>>>>>
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>
>> >>>>
>> >>>
>> >>>
>> >>>
>> >>> --
>> >>> Chanwit Kaewkasi
>> >>> PhD Candidate,
>> >>> Centre for Novel Computing
>> >>> School of Computer Science
>> >>> The University of Manchester
>> >>> Oxford Road
>> >>> Manchester
>> >>> M13 9PL, UK
>> >>>
>> >>
>> >>
>> >>
>> >> --
>> >> Best Regards, Edward J. Yoon @ NHN, corp.
>> >> edwardyoon@apache.org
>> >> http://blog.udanax.org
>> >>
>> >
>> >
>> >
>> > --
>> > Best Regards, Edward J. Yoon
>> > edwardyoon@apache.org
>> > http://blog.udanax.org
>> >
>>
>>
>>
>> --
>> Chanwit Kaewkasi
>> PhD Candidate,
>> Centre for Novel Computing
>> School of Computer Science
>> The University of Manchester
>> Oxford Road
>> Manchester
>> M13 9PL, UK
>>
>
--
Chanwit Kaewkasi
PhD Candidate,
Centre for Novel Computing
School of Computer Science
The University of Manchester
Oxford Road
Manchester
M13 9PL, UK
Re: Ivy for jar dependency management
Posted by pi song <pi...@gmail.com>.
Hi,
>From my understanding, Ivy also relies on maven repository. Then why don't
you go for Maven2 directly?
Do you have things like parser generator or any kind of code generator in
your compile process?
Pi
On Thu, Feb 19, 2009 at 1:43 AM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
> I've got an external branch for testing here.
>
> http://github.com/chanwit/hama/tree/master
>
> Will notify you again when it's ready (and before getting merged).
>
> Chanwit
>
> 2009/2/17 Edward J. Yoon <ed...@apache.org>:
> > Hi chanwit, How goes it?
> >
> > On Mon, Jan 19, 2009 at 9:59 AM, Edward J. Yoon <ed...@apache.org>
> wrote:
> >>> It seems there is no maven repository for HBase or Hadoop that I can
> use.
> >>
> >> IMO, It would be nice If we can handle them (hadoop/hbase) manually
> >> and use Ivy for other dependencies at same time.
> >>
> >>> And after checking HBase artefact on Hudson, I've found that it
> >>> contains all dependency jars.
> >>
> >> Good idea. BTW, then how can we set up dev environment?
> >>
> >> On Sun, Jan 18, 2009 at 1:28 PM, Chanwit Kaewkasi <ch...@gmail.com>
> wrote:
> >>> Some updates.
> >>>
> >>
> >>> And after checking HBase artefact on Hudson, I've found that it
> >>> contains all dependency jars.
> >>> I think I may write a script to fetch it to prepare the build
> >>> environment rather than using ivy.
> >>>
> >>> - or -
> >>>
> >>> I may write a script to build a private maven repo for HBase and
> >>> Hadoop, then we can use ivy.
> >>>
> >>> So the concept are still as follows:
> >>> - SVN may contain only "src" and some other documents.
> >>> - HBase will be downloaded from Hudson into "lib". (We probably needs
> >>> different configuration for "dev" and "production")
> >>> - Findbug may be downloaded to run separately.
> >>>
> >>> Regards,
> >>>
> >>> Chanwit
> >>>
> >>> 2009/1/7 Ian Holsman <li...@holsman.net>:
> >>>> Edward J. Yoon wrote:
> >>>>>
> >>>>> That's great.
> >>>>>
> >>>>
> >>>> It will make life easier
> >>>> +1
> >>>>
> >>>>> On Wed, Jan 7, 2009 at 6:31 AM, Chanwit Kaewkasi <ch...@gmail.com>
> >>>>> wrote:
> >>>>>
> >>>>>>
> >>>>>> Some more update.
> >>>>>> I've just found that Hadoop's using Ivy in its build script as well.
> >>>>>> So it's not that hard to follow:
> >>>>>>
> >>>>>>
> http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
> >>>>>>
> >>>>>> Chanwit
> >>>>>>
> >>>>>> 2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
> >>>>>>
> >>>>>>>
> >>>>>>> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
> >>>>>>>
> >>>>>>>>
> >>>>>>>> I like this idea. +1
> >>>>>>>>
> >>>>>>>> Then, We don't need to update periodically?
> >>>>>>>>
> >>>>>>>
> >>>>>>> Yes, I think so. Just change the XML configuration, and rebuild,
> then
> >>>>>>> it should resolve a new set of jars for us without putting them
> into
> >>>>>>> the repo. Let me go through its document more.
> >>>>>>>
> >>>>>>> Chanwit
> >>>>>>>
> >>>>>>>
> >>>>>>
> >>>>>> --
> >>>>>> Chanwit Kaewkasi
> >>>>>> PhD Candidate,
> >>>>>> Centre for Novel Computing
> >>>>>> School of Computer Science
> >>>>>> The University of Manchester
> >>>>>> Oxford Road
> >>>>>> Manchester
> >>>>>> M13 9PL, UK
> >>>>>>
> >>>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>
> >>>>
> >>>
> >>>
> >>>
> >>> --
> >>> Chanwit Kaewkasi
> >>> PhD Candidate,
> >>> Centre for Novel Computing
> >>> School of Computer Science
> >>> The University of Manchester
> >>> Oxford Road
> >>> Manchester
> >>> M13 9PL, UK
> >>>
> >>
> >>
> >>
> >> --
> >> Best Regards, Edward J. Yoon @ NHN, corp.
> >> edwardyoon@apache.org
> >> http://blog.udanax.org
> >>
> >
> >
> >
> > --
> > Best Regards, Edward J. Yoon
> > edwardyoon@apache.org
> > http://blog.udanax.org
> >
>
>
>
> --
> Chanwit Kaewkasi
> PhD Candidate,
> Centre for Novel Computing
> School of Computer Science
> The University of Manchester
> Oxford Road
> Manchester
> M13 9PL, UK
>
Re: Ivy for jar dependency management
Posted by Chanwit Kaewkasi <ch...@gmail.com>.
I've got an external branch for testing here.
http://github.com/chanwit/hama/tree/master
Will notify you again when it's ready (and before getting merged).
Chanwit
2009/2/17 Edward J. Yoon <ed...@apache.org>:
> Hi chanwit, How goes it?
>
> On Mon, Jan 19, 2009 at 9:59 AM, Edward J. Yoon <ed...@apache.org> wrote:
>>> It seems there is no maven repository for HBase or Hadoop that I can use.
>>
>> IMO, It would be nice If we can handle them (hadoop/hbase) manually
>> and use Ivy for other dependencies at same time.
>>
>>> And after checking HBase artefact on Hudson, I've found that it
>>> contains all dependency jars.
>>
>> Good idea. BTW, then how can we set up dev environment?
>>
>> On Sun, Jan 18, 2009 at 1:28 PM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
>>> Some updates.
>>>
>>
>>> And after checking HBase artefact on Hudson, I've found that it
>>> contains all dependency jars.
>>> I think I may write a script to fetch it to prepare the build
>>> environment rather than using ivy.
>>>
>>> - or -
>>>
>>> I may write a script to build a private maven repo for HBase and
>>> Hadoop, then we can use ivy.
>>>
>>> So the concept are still as follows:
>>> - SVN may contain only "src" and some other documents.
>>> - HBase will be downloaded from Hudson into "lib". (We probably needs
>>> different configuration for "dev" and "production")
>>> - Findbug may be downloaded to run separately.
>>>
>>> Regards,
>>>
>>> Chanwit
>>>
>>> 2009/1/7 Ian Holsman <li...@holsman.net>:
>>>> Edward J. Yoon wrote:
>>>>>
>>>>> That's great.
>>>>>
>>>>
>>>> It will make life easier
>>>> +1
>>>>
>>>>> On Wed, Jan 7, 2009 at 6:31 AM, Chanwit Kaewkasi <ch...@gmail.com>
>>>>> wrote:
>>>>>
>>>>>>
>>>>>> Some more update.
>>>>>> I've just found that Hadoop's using Ivy in its build script as well.
>>>>>> So it's not that hard to follow:
>>>>>>
>>>>>> http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
>>>>>>
>>>>>> Chanwit
>>>>>>
>>>>>> 2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
>>>>>>
>>>>>>>
>>>>>>> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
>>>>>>>
>>>>>>>>
>>>>>>>> I like this idea. +1
>>>>>>>>
>>>>>>>> Then, We don't need to update periodically?
>>>>>>>>
>>>>>>>
>>>>>>> Yes, I think so. Just change the XML configuration, and rebuild, then
>>>>>>> it should resolve a new set of jars for us without putting them into
>>>>>>> the repo. Let me go through its document more.
>>>>>>>
>>>>>>> Chanwit
>>>>>>>
>>>>>>>
>>>>>>
>>>>>> --
>>>>>> Chanwit Kaewkasi
>>>>>> PhD Candidate,
>>>>>> Centre for Novel Computing
>>>>>> School of Computer Science
>>>>>> The University of Manchester
>>>>>> Oxford Road
>>>>>> Manchester
>>>>>> M13 9PL, UK
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>
>>>
>>>
>>> --
>>> Chanwit Kaewkasi
>>> PhD Candidate,
>>> Centre for Novel Computing
>>> School of Computer Science
>>> The University of Manchester
>>> Oxford Road
>>> Manchester
>>> M13 9PL, UK
>>>
>>
>>
>>
>> --
>> Best Regards, Edward J. Yoon @ NHN, corp.
>> edwardyoon@apache.org
>> http://blog.udanax.org
>>
>
>
>
> --
> Best Regards, Edward J. Yoon
> edwardyoon@apache.org
> http://blog.udanax.org
>
--
Chanwit Kaewkasi
PhD Candidate,
Centre for Novel Computing
School of Computer Science
The University of Manchester
Oxford Road
Manchester
M13 9PL, UK
Re: Ivy for jar dependency management
Posted by "Edward J. Yoon" <ed...@apache.org>.
Hi chanwit, How goes it?
On Mon, Jan 19, 2009 at 9:59 AM, Edward J. Yoon <ed...@apache.org> wrote:
>> It seems there is no maven repository for HBase or Hadoop that I can use.
>
> IMO, It would be nice If we can handle them (hadoop/hbase) manually
> and use Ivy for other dependencies at same time.
>
>> And after checking HBase artefact on Hudson, I've found that it
>> contains all dependency jars.
>
> Good idea. BTW, then how can we set up dev environment?
>
> On Sun, Jan 18, 2009 at 1:28 PM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
>> Some updates.
>>
>
>> And after checking HBase artefact on Hudson, I've found that it
>> contains all dependency jars.
>> I think I may write a script to fetch it to prepare the build
>> environment rather than using ivy.
>>
>> - or -
>>
>> I may write a script to build a private maven repo for HBase and
>> Hadoop, then we can use ivy.
>>
>> So the concept are still as follows:
>> - SVN may contain only "src" and some other documents.
>> - HBase will be downloaded from Hudson into "lib". (We probably needs
>> different configuration for "dev" and "production")
>> - Findbug may be downloaded to run separately.
>>
>> Regards,
>>
>> Chanwit
>>
>> 2009/1/7 Ian Holsman <li...@holsman.net>:
>>> Edward J. Yoon wrote:
>>>>
>>>> That's great.
>>>>
>>>
>>> It will make life easier
>>> +1
>>>
>>>> On Wed, Jan 7, 2009 at 6:31 AM, Chanwit Kaewkasi <ch...@gmail.com>
>>>> wrote:
>>>>
>>>>>
>>>>> Some more update.
>>>>> I've just found that Hadoop's using Ivy in its build script as well.
>>>>> So it's not that hard to follow:
>>>>>
>>>>> http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
>>>>>
>>>>> Chanwit
>>>>>
>>>>> 2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
>>>>>
>>>>>>
>>>>>> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
>>>>>>
>>>>>>>
>>>>>>> I like this idea. +1
>>>>>>>
>>>>>>> Then, We don't need to update periodically?
>>>>>>>
>>>>>>
>>>>>> Yes, I think so. Just change the XML configuration, and rebuild, then
>>>>>> it should resolve a new set of jars for us without putting them into
>>>>>> the repo. Let me go through its document more.
>>>>>>
>>>>>> Chanwit
>>>>>>
>>>>>>
>>>>>
>>>>> --
>>>>> Chanwit Kaewkasi
>>>>> PhD Candidate,
>>>>> Centre for Novel Computing
>>>>> School of Computer Science
>>>>> The University of Manchester
>>>>> Oxford Road
>>>>> Manchester
>>>>> M13 9PL, UK
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>>
>> --
>> Chanwit Kaewkasi
>> PhD Candidate,
>> Centre for Novel Computing
>> School of Computer Science
>> The University of Manchester
>> Oxford Road
>> Manchester
>> M13 9PL, UK
>>
>
>
>
> --
> Best Regards, Edward J. Yoon @ NHN, corp.
> edwardyoon@apache.org
> http://blog.udanax.org
>
--
Best Regards, Edward J. Yoon
edwardyoon@apache.org
http://blog.udanax.org
Re: Ivy for jar dependency management
Posted by "Edward J. Yoon" <ed...@apache.org>.
> It seems there is no maven repository for HBase or Hadoop that I can use.
IMO, It would be nice If we can handle them (hadoop/hbase) manually
and use Ivy for other dependencies at same time.
> And after checking HBase artefact on Hudson, I've found that it
> contains all dependency jars.
Good idea. BTW, then how can we set up dev environment?
On Sun, Jan 18, 2009 at 1:28 PM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
> Some updates.
>
> And after checking HBase artefact on Hudson, I've found that it
> contains all dependency jars.
> I think I may write a script to fetch it to prepare the build
> environment rather than using ivy.
>
> - or -
>
> I may write a script to build a private maven repo for HBase and
> Hadoop, then we can use ivy.
>
> So the concept are still as follows:
> - SVN may contain only "src" and some other documents.
> - HBase will be downloaded from Hudson into "lib". (We probably needs
> different configuration for "dev" and "production")
> - Findbug may be downloaded to run separately.
>
> Regards,
>
> Chanwit
>
> 2009/1/7 Ian Holsman <li...@holsman.net>:
>> Edward J. Yoon wrote:
>>>
>>> That's great.
>>>
>>
>> It will make life easier
>> +1
>>
>>> On Wed, Jan 7, 2009 at 6:31 AM, Chanwit Kaewkasi <ch...@gmail.com>
>>> wrote:
>>>
>>>>
>>>> Some more update.
>>>> I've just found that Hadoop's using Ivy in its build script as well.
>>>> So it's not that hard to follow:
>>>>
>>>> http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
>>>>
>>>> Chanwit
>>>>
>>>> 2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
>>>>
>>>>>
>>>>> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
>>>>>
>>>>>>
>>>>>> I like this idea. +1
>>>>>>
>>>>>> Then, We don't need to update periodically?
>>>>>>
>>>>>
>>>>> Yes, I think so. Just change the XML configuration, and rebuild, then
>>>>> it should resolve a new set of jars for us without putting them into
>>>>> the repo. Let me go through its document more.
>>>>>
>>>>> Chanwit
>>>>>
>>>>>
>>>>
>>>> --
>>>> Chanwit Kaewkasi
>>>> PhD Candidate,
>>>> Centre for Novel Computing
>>>> School of Computer Science
>>>> The University of Manchester
>>>> Oxford Road
>>>> Manchester
>>>> M13 9PL, UK
>>>>
>>>>
>>>
>>>
>>>
>>>
>>
>>
>
>
>
> --
> Chanwit Kaewkasi
> PhD Candidate,
> Centre for Novel Computing
> School of Computer Science
> The University of Manchester
> Oxford Road
> Manchester
> M13 9PL, UK
>
--
Best Regards, Edward J. Yoon @ NHN, corp.
edwardyoon@apache.org
http://blog.udanax.org
Re: Ivy for jar dependency management
Posted by Chanwit Kaewkasi <ch...@gmail.com>.
Some updates.
It seems there is no maven repository for HBase or Hadoop that I can use.
And after checking HBase artefact on Hudson, I've found that it
contains all dependency jars.
I think I may write a script to fetch it to prepare the build
environment rather than using ivy.
- or -
I may write a script to build a private maven repo for HBase and
Hadoop, then we can use ivy.
So the concept are still as follows:
- SVN may contain only "src" and some other documents.
- HBase will be downloaded from Hudson into "lib". (We probably needs
different configuration for "dev" and "production")
- Findbug may be downloaded to run separately.
Regards,
Chanwit
2009/1/7 Ian Holsman <li...@holsman.net>:
> Edward J. Yoon wrote:
>>
>> That's great.
>>
>
> It will make life easier
> +1
>
>> On Wed, Jan 7, 2009 at 6:31 AM, Chanwit Kaewkasi <ch...@gmail.com>
>> wrote:
>>
>>>
>>> Some more update.
>>> I've just found that Hadoop's using Ivy in its build script as well.
>>> So it's not that hard to follow:
>>>
>>> http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
>>>
>>> Chanwit
>>>
>>> 2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
>>>
>>>>
>>>> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
>>>>
>>>>>
>>>>> I like this idea. +1
>>>>>
>>>>> Then, We don't need to update periodically?
>>>>>
>>>>
>>>> Yes, I think so. Just change the XML configuration, and rebuild, then
>>>> it should resolve a new set of jars for us without putting them into
>>>> the repo. Let me go through its document more.
>>>>
>>>> Chanwit
>>>>
>>>>
>>>
>>> --
>>> Chanwit Kaewkasi
>>> PhD Candidate,
>>> Centre for Novel Computing
>>> School of Computer Science
>>> The University of Manchester
>>> Oxford Road
>>> Manchester
>>> M13 9PL, UK
>>>
>>>
>>
>>
>>
>>
>
>
--
Chanwit Kaewkasi
PhD Candidate,
Centre for Novel Computing
School of Computer Science
The University of Manchester
Oxford Road
Manchester
M13 9PL, UK
Re: Ivy for jar dependency management
Posted by Ian Holsman <li...@holsman.net>.
Edward J. Yoon wrote:
> That's great.
>
It will make life easier
+1
> On Wed, Jan 7, 2009 at 6:31 AM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
>
>> Some more update.
>> I've just found that Hadoop's using Ivy in its build script as well.
>> So it's not that hard to follow:
>>
>> http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
>>
>> Chanwit
>>
>> 2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
>>
>>> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
>>>
>>>> I like this idea. +1
>>>>
>>>> Then, We don't need to update periodically?
>>>>
>>> Yes, I think so. Just change the XML configuration, and rebuild, then
>>> it should resolve a new set of jars for us without putting them into
>>> the repo. Let me go through its document more.
>>>
>>> Chanwit
>>>
>>>
>>
>> --
>> Chanwit Kaewkasi
>> PhD Candidate,
>> Centre for Novel Computing
>> School of Computer Science
>> The University of Manchester
>> Oxford Road
>> Manchester
>> M13 9PL, UK
>>
>>
>
>
>
>
Re: Ivy for jar dependency management
Posted by "Edward J. Yoon" <ed...@apache.org>.
That's great.
On Wed, Jan 7, 2009 at 6:31 AM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
> Some more update.
> I've just found that Hadoop's using Ivy in its build script as well.
> So it's not that hard to follow:
>
> http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
>
> Chanwit
>
> 2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
>> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
>>> I like this idea. +1
>>>
>>> Then, We don't need to update periodically?
>>
>> Yes, I think so. Just change the XML configuration, and rebuild, then
>> it should resolve a new set of jars for us without putting them into
>> the repo. Let me go through its document more.
>>
>> Chanwit
>>
>
>
>
> --
> Chanwit Kaewkasi
> PhD Candidate,
> Centre for Novel Computing
> School of Computer Science
> The University of Manchester
> Oxford Road
> Manchester
> M13 9PL, UK
>
--
Best Regards, Edward J. Yoon @ NHN, corp.
edwardyoon@apache.org
http://blog.udanax.org
Re: Ivy for jar dependency management
Posted by Chanwit Kaewkasi <ch...@gmail.com>.
Some more update.
I've just found that Hadoop's using Ivy in its build script as well.
So it's not that hard to follow:
http://svn.apache.org/viewvc/hadoop/core/trunk/build.xml?view=markup
Chanwit
2009/1/6 Chanwit Kaewkasi <ch...@gmail.com>:
> 2009/1/6 Edward J. Yoon <ed...@apache.org>:
>> I like this idea. +1
>>
>> Then, We don't need to update periodically?
>
> Yes, I think so. Just change the XML configuration, and rebuild, then
> it should resolve a new set of jars for us without putting them into
> the repo. Let me go through its document more.
>
> Chanwit
>
--
Chanwit Kaewkasi
PhD Candidate,
Centre for Novel Computing
School of Computer Science
The University of Manchester
Oxford Road
Manchester
M13 9PL, UK
Re: Ivy for jar dependency management
Posted by Chanwit Kaewkasi <ch...@gmail.com>.
2009/1/6 Edward J. Yoon <ed...@apache.org>:
> I like this idea. +1
>
> Then, We don't need to update periodically?
Yes, I think so. Just change the XML configuration, and rebuild, then
it should resolve a new set of jars for us without putting them into
the repo. Let me go through its document more.
Chanwit
Re: Ivy for jar dependency management
Posted by "Edward J. Yoon" <ed...@apache.org>.
I like this idea. +1
Then, We don't need to update periodically?
On Tue, Jan 6, 2009 at 1:34 PM, Chanwit Kaewkasi <ch...@gmail.com> wrote:
> Hello,
>
> I'm thinking of using Ivy (http://ant.apache.org/ivy/) to help us
> manage .jar files, as currently a lot of .jars have been put into
> trunk/lib directory (including findbug stuff).
> It would be better if we can only have source files in the repo, and
> all jar are automatically resolved and downloaded when the project is
> built.
>
> What do you think? (If you all agree, I'll start learning Ivy configuration).
>
> Regards,
>
> Chanwit
>
> --
> Chanwit Kaewkasi
> PhD Candidate,
> Centre for Novel Computing
> School of Computer Science
> The University of Manchester
> Oxford Road
> Manchester
> M13 9PL, UK
>
--
Best Regards, Edward J. Yoon @ NHN, corp.
edwardyoon@apache.org
http://blog.udanax.org