You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Niklas Gustavsson <ni...@protocol7.com> on 2005/08/05 10:32:34 UTC
[exec][patch] commons-exec initial contrib
Brett Porter wrote:
>> * Would it be a good starting point (if so I'll create a patch for the
>> code I've cleaned up and removed the Ant specifics from)?
>
>
> I think so.
(This email with a .zip attachment does not seem to get through to the
list. And since there's no component for exec in bugzilla, I've uploaded
the file to
http://www.protocol7.com/tmp/commons-exec-initial-20050802.zip instead.
If I should send it somehow different, please enlighten me.)
Attached is the initial contribution, I've zipped it since it's largish.
Note that there are tons of problems with this code (e.g. design issues,
dependency on JCL, lack of unit tests/javadoc) but IMHO it's a good
enough starting point. It uses maven2 for building and builds with the
latest SVN version of maven2.
If you have any issues with the code, maven2 setup, patch or anything,
please tell me and I'll get it fixed!
/niklas
------------------
Niklas Gustavsson
niklas@protocol7.com
http://www.protocol7.com
--
------------------
Niklas Gustavsson
niklas@protocol7.com
http://www.protocol7.com
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Brett Porter <br...@apache.org>.
Thanks! I'll add it to the sandbox shortly and see if there is anything
I can do.
- Brett
Niklas Gustavsson wrote:
> Brett Porter wrote:
>
>>> * Would it be a good starting point (if so I'll create a patch for the
>>> code I've cleaned up and removed the Ant specifics from)?
>>
>>
>>
>> I think so.
>
>
>
> (This email with a .zip attachment does not seem to get through to the
> list. And since there's no component for exec in bugzilla, I've
> uploaded the file to
> http://www.protocol7.com/tmp/commons-exec-initial-20050802.zip
> instead. If I should send it somehow different, please enlighten me.)
>
> Attached is the initial contribution, I've zipped it since it's
> largish. Note that there are tons of problems with this code (e.g.
> design issues, dependency on JCL, lack of unit tests/javadoc) but IMHO
> it's a good enough starting point. It uses maven2 for building and
> builds with the latest SVN version of maven2.
>
> If you have any issues with the code, maven2 setup, patch or anything,
> please tell me and I'll get it fixed!
>
> /niklas
>
> ------------------
> Niklas Gustavsson
> niklas@protocol7.com
> http://www.protocol7.com
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Kev Jackson <ke...@it.fts-vn.com>.
Niklas Gustavsson wrote:
> Kev Jackson wrote:
>
>> I'm not a fan of maven, so I'll write up a quick build file for it,
>> I'll also look into javadoc/unit tests as soon as I have some time
>> (perhaps tomorrow)
>
>
> Great! With regards for the unit tests I personally think that the
> first step should be to refactor the launcher so that they do not
> directly call Runtime.exec(). That currently makes unit testing hard.
Ok, when I have time I'll look into it - I'm currently bogged down with
some very dull development for 10+ hours a day, so I've neither the time
nor inclination to spend on re-factoring etc right now (I thought I'd
get time this weekend - I was wrong :( ) Sorry for no contributions,
I'll try and get something to you as soon as I've reduced my workload a
little
Kev
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Niklas Gustavsson <ni...@protocol7.com>.
Kev Jackson wrote:
> I'm not a fan of maven, so I'll write up a quick build file for it, I'll
> also look into javadoc/unit tests as soon as I have some time (perhaps
> tomorrow)
Great! With regards for the unit tests I personally think that the first
step should be to refactor the launcher so that they do not directly
call Runtime.exec(). That currently makes unit testing hard.
/niklas
------------------
Niklas Gustavsson
niklas@protocol7.com
http://www.protocol7.com
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Niklas Gustavsson <ni...@protocol7.com>.
Brett Porter wrote:
> Niklas, I made a couple of small structural changes, fleshed out the
> POM, and added a front page on the site:
> http://jakarta.apache.org/commons/sandbox/exec/
Yeah, the POM was a bit... limited :-)
Excellent, I'll sync up right away so that I can get going on the
development.
/niklas
------------------
Niklas Gustavsson
niklas@protocol7.com
http://www.protocol7.com
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Brett Porter <br...@apache.org>.
Kev Jackson wrote:
> I'm not a fan of maven, so I'll write up a quick build file for it,
> I'll also look into javadoc/unit tests as soon as I have some time
> (perhaps tomorrow)
>
I've autogenerated the Ant build script for you. You'll need m2 to do
the site at this point.
Niklas, I made a couple of small structural changes, fleshed out the
POM, and added a front page on the site:
http://jakarta.apache.org/commons/sandbox/exec/
Pushed up build #1:
http://cvs.apache.org/repository/org.apache.commons/jars/
Ok, I'll take a closer look and provide more feedback.
Thanks!
- Brett
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Kev Jackson <ke...@it.fts-vn.com>.
>
>
> (This email with a .zip attachment does not seem to get through to the
> list. And since there's no component for exec in bugzilla, I've
> uploaded the file to
> http://www.protocol7.com/tmp/commons-exec-initial-20050802.zip
> instead. If I should send it somehow different, please enlighten me.)
>
> Attached is the initial contribution, I've zipped it since it's
> largish. Note that there are tons of problems with this code (e.g.
> design issues, dependency on JCL, lack of unit tests/javadoc) but IMHO
> it's a good enough starting point. It uses maven2 for building and
> builds with the latest SVN version of maven2.
>
> If you have any issues with the code, maven2 setup, patch or anything,
> please tell me and I'll get it fixed!
>
I'm not a fan of maven, so I'll write up a quick build file for it, I'll
also look into javadoc/unit tests as soon as I have some time (perhaps
tomorrow)
Kev
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Niklas Gustavsson <ni...@protocol7.com>.
Noel J. Bergman wrote:
> Niklas Gustavsson wrote:
>
>
>>Should a CLA/grant be needed, I would of course be happy to work that out.
>>Would a CLA be needed for future patches or would attaching them to a
>>Bugzilla entry be acceptable?
>
>
> If you are going to contribute on an on-going basis, as CLA would be best,
> yes.
Happy to oblige :-)
I'll sign the agreement (http://www.apache.org/licenses/icla.txt) and
fax it to ASF. Should I also get the CCLA signed while I'm at it?
Contributions I make to Apache projects are not part of my job, but I
understand that the CCLA might be required anyway.
/niklas
------------------
Niklas Gustavsson
niklas@protocol7.com
http://www.protocol7.com
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Brett Porter <br...@apache.org>.
I can confirm that Niklas' CLA has been received and recorded by the ASF.
Cheers,
Brett
Noel J. Bergman wrote:
>Niklas Gustavsson wrote:
>
>
>
>>Should a CLA/grant be needed, I would of course be happy to work that out.
>>Would a CLA be needed for future patches or would attaching them to a
>>Bugzilla entry be acceptable?
>>
>>
>
>If you are going to contribute on an on-going basis, as CLA would be best,
>yes. Otherwise, a Software Grant would cover substantial contributions.
>The route of attaching patches to bugzilla/mailing list/JIRA is best for
>small patches to existing code, not substantial changes or entire new
>packages.
>
> --- Noel
>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
>For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
RE: [exec][patch] commons-exec initial contrib
Posted by "Noel J. Bergman" <no...@devtech.com>.
Niklas Gustavsson wrote:
> Should a CLA/grant be needed, I would of course be happy to work that out.
> Would a CLA be needed for future patches or would attaching them to a
> Bugzilla entry be acceptable?
If you are going to contribute on an on-going basis, as CLA would be best,
yes. Otherwise, a Software Grant would cover substantial contributions.
The route of attaching patches to bugzilla/mailing list/JIRA is best for
small patches to existing code, not substantial changes or entire new
packages.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Niklas Gustavsson <ni...@protocol7.com>.
Hi
I was was writing a reply to Noel's message but Brett beat me to it.
Brett Porter wrote:
> What have I missed and how do we need to correct it? In the worst case,
> I can svn cp from Ant (as they are soon going to SVN), and stepwise
> refine it to include these improvements. Do we still need a CLA/grant
> from Niklas?
Should a CLA/grant be needed, I would of course be happy to work that out.
Would a CLA be needed for future patches or would attaching them to a
Bugzilla entry be acceptable?
/niklas
------------------
Niklas Gustavsson
niklas@protocol7.com
http://www.protocol7.com
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Niklas Gustavsson <ni...@protocol7.com>.
Brett Porter wrote:
>>Grey areas, but the larger the contributions that more important that we get
>>a CLA. And, as you noted, the o.a.c.e.launcher package was created by him,
>>and is not just a set of patches.
>
> Actually, I thought originally this was an original contribution, but on
> closer inspection I've found that it is also derived from another set of
> Ant classes.
That's true. The only new code in my patch are the unit tests and the
pom.xml. Everything else is more or less refactored Ant code.
/niklas
------------------
Niklas Gustavsson
niklas@protocol7.com
http://www.protocol7.com
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Brett Porter <br...@apache.org>.
Noel J. Bergman wrote:
>Brett Porter wrote:
>
>
>>"A copy of several classes from the Apache Ant codebase, restructured
>>into a reusable library without dependence on Ant itself."
>>
>>
>
>Sounds fine. Now that we know the history, even the original commit comment
>makes sense, but when you put the original against the e-mail archives
>(which talk about "initial contribution") and you don't know the history, it
>isn't as clear as your revision.
>
>
>
Done (yay SVN - would have required an admin in CVS).
- Brett
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
RE: [exec][patch] commons-exec initial contrib
Posted by "Noel J. Bergman" <no...@devtech.com>.
Brett Porter wrote:
> "A copy of several classes from the Apache Ant codebase, restructured
> into a reusable library without dependence on Ant itself."
Sounds fine. Now that we know the history, even the original commit comment
makes sense, but when you put the original against the e-mail archives
(which talk about "initial contribution") and you don't know the history, it
isn't as clear as your revision.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Brett Porter <br...@apache.org>.
Noel J. Bergman wrote:
>>Ok, do I need to edit the commit message?
>>
>>
>
>If you don't mind, yes thanks.
>
>
Ok, I could use some guidance on this. The original message was:
r230452 | brett | 2005-08-05 22:21:46 +1000 (Fri, 05 Aug 2005) | 3 lines
Submitted by: Niklas Gustavsson
initial import of refactored code from the Ant Exec tasks.
------------------------------------------------------------------------
What areas did you find this deficient that I can improve it?
How about:
"A copy of several classes from the Apache Ant codebase, restructured
into a reusable library without dependence on Ant itself."
>
>>Is a CLA from Niklas (and CCLA if necessary) plus modification
>>of the original commit message to indicate origin sufficient?
>>
>>
>
>Seems so. From what you seem to have tracked down, everything is turning
>out to be really patches to a variety of Ant packages.
>
>
Thanks, I'll follow up on this, and post to Jakarta PMC when it is
received in iclas.txt.
- Brett
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
RE: [exec][patch] commons-exec initial contrib
Posted by "Noel J. Bergman" <no...@devtech.com>.
Brett Porter wrote:
> Noel J. Bergman wrote:
> > I re-read the commit messages, and no it was not entirely
> > clear, at least not in terms of the clarity I'd want for
> > IP clearance. The concern is tracking provenance.
> Ok, do I need to edit the commit message?
If you don't mind, yes thanks.
> Is [re-doing everything starting with svn cp] necessary
> if Niklas has a CLA on file?
Probably not worth it. Just make the comments clear. :-)
> > as you noted, the o.a.c.e.launcher package was created by him,
> > and is not just a set of patches.
> Actually, I thought originally this was an original contribution,
> but on closer inspection I've found that it is also derived from
> another set of Ant classes.
:-) See, this is why we really want to make provenance clear.
> Is a CLA from Niklas (and CCLA if necessary) plus modification
> of the original commit message to indicate origin sufficient?
Seems so. From what you seem to have tracked down, everything is turning
out to be really patches to a variety of Ant packages.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Brett Porter <br...@apache.org>.
Noel J. Bergman wrote:
>I re-read the commit messages, and no it was not entirely clear, at least
>not in terms of the clarity I'd want for IP clearance. The concern is
>tracking provenance.
>
>
Ok, do I need to edit the commit message?
>Ant is in Subversion. If these are patched versions of Ant code, why didn't
>you use `svn cp` to initially populate the tree, with appropriate comments,
>then apply the patches?
>
>
Only some new components are in there, but according to their dev list
they have voted on moving the main repository (that includes this code)
in the next couple of weeks.
Is this necessary if Niklas has a CLA on file?
>
>Grey areas, but the larger the contributions that more important that we get
>a CLA. And, as you noted, the o.a.c.e.launcher package was created by him,
>and is not just a set of patches.
>
>
Actually, I thought originally this was an original contribution, but on
closer inspection I've found that it is also derived from another set of
Ant classes.
So, the question now is what is necessary to clear this up? Is a CLA
from Niklas (and CCLA if necessary) plus modification of the original
commit message to indicate origin sufficient? On reading the IP
clearance document, it appears so.
- Brett
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
RE: [exec][patch] commons-exec initial contrib
Posted by "Noel J. Bergman" <no...@devtech.com>.
Brett Porter wrote:
> The "contribution" comes almost entirely from the Ant codebase. Sorry if
> I didn't make that clear from the commit message.
I re-read the commit messages, and no it was not entirely clear, at least
not in terms of the clarity I'd want for IP clearance. The concern is
tracking provenance.
Ant is in Subversion. If these are patched versions of Ant code, why didn't
you use `svn cp` to initially populate the tree, with appropriate comments,
then apply the patches?
> Niklas is not a yet a committer, so doesn't have a CLA. I was under the
> impression his contribution was goverened by the same rules as any patch
> submitted to the mailing list.
Grey areas, but the larger the contributions that more important that we get
a CLA. And, as you noted, the o.a.c.e.launcher package was created by him,
and is not just a set of patches.
> I was also under the impression that the sandbox was equipped to house
> code in the same way as the incubator as official releases are not made
> from there.
Absolutely not. The sandbox is not a substitute incubator.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
Re: [exec][patch] commons-exec initial contrib
Posted by Brett Porter <br...@apache.org>.
The "contribution" comes almost entirely from the Ant codebase. Sorry if
I didn't make that clear from the commit message.
The exceptions:
- changed structure (merged classes such as ExecTask and Exec,
refactored out environment).
- addition of the o.a.c.e.launcher package which was created by Niklas.
I gave it a once over before committing it.
Niklas is not a yet a committer, so doesn't have a CLA. I was under the
impression his contribution was goverened by the same rules as any patch
submitted to the mailing list. I have asked that he use bugzilla in future.
I was also under the impression that the sandbox was equipped to house
code in the same way as the incubator as official releases are not made
from there.
What have I missed and how do we need to correct it? In the worst case,
I can svn cp from Ant (as they are soon going to SVN), and stepwise
refine it to include these improvements. Do we still need a CLA/grant
from Niklas?
- Brett
Noel J. Bergman wrote:
>Re: http://svn.apache.org/repos/asf/jakarta/commons/sandbox/exec/trunk/
>
>Whoa, hello!
>
>This thing is labeled as an initial contribution, and the SVN tree was just
>created for it in July. It doesn't matter if it is a sandbox project or
>not. New codebases *MUST* be imported through the Incubator, at least in
>terms of recording the IP clearance.
>
>Where is the CLA for Niklas Gustavsson? Where is the Software Grant for the
>codebase? What are the origins of this codebase, since that isn't clear
>from the e-mail archives.
>
>Please explain what I am missing or correct this situation immediately.
>
> --- Noel
>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
>For additional commands, e-mail: commons-dev-help@jakarta.apache.org
>
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org
RE: [exec][patch] commons-exec initial contrib
Posted by "Noel J. Bergman" <no...@devtech.com>.
Re: http://svn.apache.org/repos/asf/jakarta/commons/sandbox/exec/trunk/
Whoa, hello!
This thing is labeled as an initial contribution, and the SVN tree was just
created for it in July. It doesn't matter if it is a sandbox project or
not. New codebases *MUST* be imported through the Incubator, at least in
terms of recording the IP clearance.
Where is the CLA for Niklas Gustavsson? Where is the Software Grant for the
codebase? What are the origins of this codebase, since that isn't clear
from the e-mail archives.
Please explain what I am missing or correct this situation immediately.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org