[Math] Toward releasing 3.0 ?

classic Classic list List threaded Threaded
49 messages Options
123
Reply | Threaded
Open this post in threaded view
|

RE: [Math] Toward releasing 3.0 ?

Patrick Meyer
I can document the StorelessCovariance addition. What is the best way to add
documentation? Do I just add comments to the file and create a patch?


-----Original Message-----
From: Thomas Neidhart [mailto:[hidden email]]
Sent: Monday, February 13, 2012 9:21 AM
To: [hidden email]
Subject: Re: [Math] Toward releasing 3.0 ?

Hi,

I have seen that there are several classes that are almost undocumented:

 - PivotingQRDecomposition (linear)
 - StorelessCovariance (stat.correlation)
 - StorelessBivariateCovariance (stat.correlation)

both seem to be quite new contributions, is somebody willing to help here?

Thomas

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]



---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Thomas Neidhart
On 02/14/2012 12:50 PM, Patrick Meyer wrote:
> I can document the StorelessCovariance addition. What is the best way to add
> documentation? Do I just add comments to the file and create a patch?

Yes this would be fine. I have seen that there is an open issue
regarding this contribution, please attach the patch to it (MATH-449).

Thomas

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Gilles Sadowski
In reply to this post by Patrick Meyer
On Tue, Feb 14, 2012 at 06:50:34AM -0500, Patrick Meyer wrote:
> I can document the StorelessCovariance addition. What is the best way to add
> documentation? Do I just add comments to the file and create a patch?
>

Yes. Then please attach the diff to a new JIRA ticket (with target fix set
to 3.0). Thanks!

Gilles

> -----Original Message-----
> From: Thomas Neidhart [mailto:[hidden email]]
> Sent: Monday, February 13, 2012 9:21 AM
> To: [hidden email]
> Subject: Re: [Math] Toward releasing 3.0 ?
>
> Hi,
>
> I have seen that there are several classes that are almost undocumented:
>
>  - PivotingQRDecomposition (linear)
>  - StorelessCovariance (stat.correlation)
>  - StorelessBivariateCovariance (stat.correlation)
>
> both seem to be quite new contributions, is somebody willing to help here?
>
> Thomas

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Gilles Sadowski
In reply to this post by Thomas Neidhart
On Tue, Feb 14, 2012 at 01:12:50PM +0100, Thomas Neidhart wrote:
> On 02/14/2012 12:50 PM, Patrick Meyer wrote:
> > I can document the StorelessCovariance addition. What is the best way to add
> > documentation? Do I just add comments to the file and create a patch?
>
> Yes this would be fine. I have seen that there is an open issue
> regarding this contribution, please attach the patch to it (MATH-449).

Yes, of course, that's better than creating a new one as I suggested in my
previous reply!
I've just updated the target version of MATH-449.

Regards,
Gilles

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Sébastien Brisard
Dear all,
do we need to clear all Findbugs/PMD warning prior to releasing?
Sébastien


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Luc Maisonobe
Le 15/02/2012 07:32, Sébastien Brisard a écrit :
> Dear all,
> do we need to clear all Findbugs/PMD warning prior to releasing?

Yes, and checkstyle too.
In some cases, there are false positive which must be handled by adding
the appropriate filter using specific filtering comments in the code for
checkstyle, and using findbugs-exclude.xml file for findbugs. I don't
known about PMD.

Luc

> Sébastien
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Gilles Sadowski
Hi.

> > do we need to clear all Findbugs/PMD warning prior to releasing?
>
> Yes, and checkstyle too.

We'll have to make an exception for "BOBYQAOptimizer": I don't want to
eliminate all those potential clues pointing at the needed improvements
towards a clean Java implementation.
If the CheckStyle report must be empty for the released source code, I
propose to first create the release branch[1], and there, disable the
scanning of the entire class.

The same might be going for "PivotingQRDecomposition", although my preferred
solution would, again, be to not include it in the release, given that it
seems unsupported.

> [...]

Is it useful to create a JIRA ticket to list all the tasks needed to be
completed for the release?


Best,
Gilles

[1] When can we do that?

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

RE: [Math] Toward releasing 3.0 ?

Patrick Meyer
In reply to this post by Gilles Sadowski
OK, I submited a patch that includes comments and documentation. Let me know
if I need to write more, but I think I've covered the functionality of the
classes.

Patrick

-----Original Message-----
From: Gilles Sadowski [mailto:[hidden email]]
Sent: Tuesday, February 14, 2012 7:46 AM
To: [hidden email]
Subject: Re: [Math] Toward releasing 3.0 ?

On Tue, Feb 14, 2012 at 01:12:50PM +0100, Thomas Neidhart wrote:
> On 02/14/2012 12:50 PM, Patrick Meyer wrote:
> > I can document the StorelessCovariance addition. What is the best
> > way to add documentation? Do I just add comments to the file and create
a patch?
>
> Yes this would be fine. I have seen that there is an open issue
> regarding this contribution, please attach the patch to it (MATH-449).

Yes, of course, that's better than creating a new one as I suggested in my
previous reply!
I've just updated the target version of MATH-449.

Regards,
Gilles

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]



---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Thomas Neidhart
On 02/15/2012 02:41 PM, Patrick Meyer wrote:
> OK, I submited a patch that includes comments and documentation. Let me know
> if I need to write more, but I think I've covered the functionality of the
> classes.

Hi Patrick,

thanks for the patch. I have applied it together with additional code
cleanup and javadoc in r1244667.

What I have done:

 - applied your javadoc patch
 - added missing javadoc for variables
 - moved var initialization to constructor
 - changed exceptions to more specific ones
 - simplified getCovarianceMatrix (uses getData() to construct the
   RealMatrix)
 - changed deltaX / deltaY to local variables

Could you please take a look and give feedback about the changes made?

@all: from my point of view this would resolve issue MATH-449 so far as
it is fully documented and has proper unit tests.

Thanks Thomas

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

RE: [Math] Toward releasing 3.0 ?

Patrick Meyer
Looks great, thanks!

-----Original Message-----
From: Thomas Neidhart [mailto:[hidden email]]
Sent: Wednesday, February 15, 2012 2:28 PM
To: [hidden email]
Subject: Re: [Math] Toward releasing 3.0 ?

On 02/15/2012 02:41 PM, Patrick Meyer wrote:
> OK, I submited a patch that includes comments and documentation. Let
> me know if I need to write more, but I think I've covered the
> functionality of the classes.

Hi Patrick,

thanks for the patch. I have applied it together with additional code
cleanup and javadoc in r1244667.

What I have done:

 - applied your javadoc patch
 - added missing javadoc for variables
 - moved var initialization to constructor
 - changed exceptions to more specific ones
 - simplified getCovarianceMatrix (uses getData() to construct the
   RealMatrix)
 - changed deltaX / deltaY to local variables

Could you please take a look and give feedback about the changes made?

@all: from my point of view this would resolve issue MATH-449 so far as it
is fully documented and has proper unit tests.

Thanks Thomas

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]



---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Thomas Neidhart
In reply to this post by Sébastien Brisard
Hi @all,

just some status/feedback on some still open issues:

 - MATH-449: I have implemented (almost) all suggestions from
   Phil, and the code is documented and tested, so imho the issue
   can be resolved unless somebody has still reservations

 - MATH-431: the two tests were contributed by Mikkel Meyer Andersen
   (is he still active?) and I have cleaned up the exceptions while
   working on another issue. There are still things to do, as can be
   seen in the comments to the issue (mainly the results are not 100%
   equivalent to R due to some specific corrections). I do not have
   the expertise to work on them myself straight away. How do we proceed
   in such a case? Keep them as they are and note the differences
   in the javadoc, or remove them unless the issues are completely
   resolved?

Cheers,

Thomas

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Mikkel Meyer Andersen-2
2012/2/17 Thomas Neidhart <[hidden email]>:

> Hi @all,
>
> just some status/feedback on some still open issues:
>
>  - MATH-449: I have implemented (almost) all suggestions from
>   Phil, and the code is documented and tested, so imho the issue
>   can be resolved unless somebody has still reservations
>
>  - MATH-431: the two tests were contributed by Mikkel Meyer Andersen
>   (is he still active?) and I have cleaned up the exceptions while
>   working on another issue. There are still things to do, as can be
>   seen in the comments to the issue (mainly the results are not 100%
>   equivalent to R due to some specific corrections). I do not have
>   the expertise to work on them myself straight away. How do we proceed
>   in such a case? Keep them as they are and note the differences
>   in the javadoc, or remove them unless the issues are completely
>   resolved?
>
> Cheers,
>
> Thomas

Dear all,

I am still active in theory, but in practise I unfortunately haven't
contributed much lately due to high workload on my PhD. I am of course
sorry for this and hope to be able to highen my activity.

Regarding MATH-431, we need to discuss if we should adopt some of the
changes R makes and find references for all these. I am on holiday
next week, but I will be back in one week and would love to
participate in the discussion. Meanwhile I would love some views on
whether to adopt the changes or not. Personally, I am fine with
noticing in the code for 3.0 release that some corrections can be made
and refer to some (e.g. from R) but none are implemented yet.

Cheers, Mikkel.

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Gilles Sadowski
In reply to this post by Thomas Neidhart
Hello.

> just some status/feedback on some still open issues:
>
>  - MATH-449: I have implemented (almost) all suggestions from
>    Phil, and the code is documented and tested, so imho the issue
>    can be resolved unless somebody has still reservations

You probably know best.

>  - MATH-431: the two tests were contributed by Mikkel Meyer Andersen
>    (is he still active?) and I have cleaned up the exceptions while
>    working on another issue. There are still things to do, as can be
>    seen in the comments to the issue (mainly the results are not 100%
>    equivalent to R due to some specific corrections). I do not have
>    the expertise to work on them myself straight away. How do we proceed
>    in such a case? Keep them as they are and note the differences
>    in the javadoc, or remove them unless the issues are completely
>    resolved?

It's categorized as "New feature", so resolution can be postponed to 3.1.


Best,
Gilles

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Thomas Neidhart
In reply to this post by Mikkel Meyer Andersen-2
On 02/17/2012 09:04 PM, Mikkel Meyer Andersen wrote:

> I am still active in theory, but in practise I unfortunately haven't
> contributed much lately due to high workload on my PhD. I am of course
> sorry for this and hope to be able to highen my activity.
>
> Regarding MATH-431, we need to discuss if we should adopt some of the
> changes R makes and find references for all these. I am on holiday
> next week, but I will be back in one week and would love to
> participate in the discussion. Meanwhile I would love some views on
> whether to adopt the changes or not. Personally, I am fine with
> noticing in the code for 3.0 release that some corrections can be made
> and refer to some (e.g. from R) but none are implemented yet.

Dear Mikkel,

good to hear from you. I will set the fix version for the issue to 3.1
as Gilles suggested.

Looking forward to discussion on this topic.

Cheers,

Thomas

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Sébastien Brisard
Dear all,
I think 3.0 is very close at hand. Just a quick note to let you know
that unfortunately, I'll be away for a week, and won't be able to help
with the final brush up (if release occurs before next weekend).
Best wishes,
Sébastien


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Gilles Sadowski
Hi.

How do we proceed from here in order to release 3.0? Cf. ticket MATH-746,
"Things to do before releasing 3.0".

Can we start to talk about an expected release date?


Thanks,
Gilles

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Luc Maisonobe
Le 20/02/2012 23:26, Gilles Sadowski a écrit :
> Hi.

Hi Gilles,

>
> How do we proceed from here in order to release 3.0? Cf. ticket MATH-746,
> "Things to do before releasing 3.0".

Sorry for being late on this.

>
> Can we start to talk about an expected release date?

I guess you did a wonderful job for closing everything. As it is clean
enough, I think we could even skip the step of using a release branch
and we could simply tag the release candidates from the trunk. This
would simply imply refraining from any change which is not related to
the release for a few days.

Someone has to volunteer to act as the release manager. The task is
simply to perform the few commands described for example here:
<http://wiki.apache.org/commons/UsingNexus>. The release manager also
signs the packages using a gpg key, which should be put in the global
KEYS file. This file can be retrieved using the following svn command:

svn checkout --depth=immediates \
  https://[your-commiter-id]@svn.apache.org/repos/asf/commons/trunks-proper

The artifacts for the release candidate must be made available and a
VOTE thread must be started on the dev list for at least 72 hours (see
<http://www.apache.org/foundation/voting.html>). There can be several
release candidate before a version finally goes out (when I release
version 2.0 I think, we needed 6 candidates ...). When the vote passes,
the exact artifacts which were used for voting will be published by
uploading the source and binary zip and tar files and by promoting the
maven artifacts with Nexus. Not a single bit is changed (this would
change the gpg signatures). This means that for example the release date
which appears in the release notes must be estimated before the vote
taking the voting delay into account (plus one or two days as a safety
margin) and it must be updated as each release candidate is cut.

So there is no predefined release date until the vote finally passes.

At the pace at which you go now, I would say we could target a first
release candidate early next week.

Any volunteer as release manager ?

Luc

>
>
> Thanks,
> Gilles
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Gilles Sadowski
Hello.

> >
> > How do we proceed from here in order to release 3.0? Cf. ticket MATH-746,
> > "Things to do before releasing 3.0".
>
> Sorry for being late on this.
>
> >
> > Can we start to talk about an expected release date?
>
> I guess you did a wonderful job for closing everything. As it is clean
> enough, I think we could even skip the step of using a release branch
> and we could simply tag the release candidates from the trunk. This
> would simply imply refraining from any change which is not related to
> the release for a few days.
>
> Someone has to volunteer to act as the release manager. The task is
> simply to perform the few commands described for example here:
> <http://wiki.apache.org/commons/UsingNexus>. The release manager also
> signs the packages using a gpg key, which should be put in the global
> KEYS file. This file can be retrieved using the following svn command:
>
> svn checkout --depth=immediates \
>   https://[your-commiter-id]@svn.apache.org/repos/asf/commons/trunks-proper
>
> The artifacts for the release candidate must be made available and a
> VOTE thread must be started on the dev list for at least 72 hours (see
> <http://www.apache.org/foundation/voting.html>). There can be several
> release candidate before a version finally goes out (when I release
> version 2.0 I think, we needed 6 candidates ...). When the vote passes,
> the exact artifacts which were used for voting will be published by
> uploading the source and binary zip and tar files and by promoting the
> maven artifacts with Nexus. Not a single bit is changed (this would
> change the gpg signatures). This means that for example the release date
> which appears in the release notes must be estimated before the vote
> taking the voting delay into account (plus one or two days as a safety
> margin) and it must be updated as each release candidate is cut.
>
> So there is no predefined release date until the vote finally passes.
>
> At the pace at which you go now, I would say we could target a first
> release candidate early next week.
>
> Any volunteer as release manager ?

OK, I started to try the commands listed in the "UsingNexus" file. Not
everything works directly... [maven2 could not find a plugin, which led me
to upgrade to maven3, which printed a warning about "parent" being a broken
project, etc.]

I don't know maven (apart from the basics to build CM) so, it is not always
obvious which are the mandatory steps and what result must be observed in
order to check that everything went fine...

For the encryption key: I was always advised against writing a passphrase in
clear in a file; maven seems to support asking for the passphrase but when
it prints:
---CUT---
Enter passphrase: gpg: gpg-agent is not available in this session
---CUT---
When I enter the passphrase, it just prints that same message again...
[I guess I'll create a dummy key and store the passphrase in "settings.xml"
just for this to work...]


Regards,
Gilles

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

sebb-2-2
On 25 February 2012 09:59, Gilles Sadowski <[hidden email]> wrote:

> Hello.
>
>> >
>> > How do we proceed from here in order to release 3.0? Cf. ticket MATH-746,
>> > "Things to do before releasing 3.0".
>>
>> Sorry for being late on this.
>>
>> >
>> > Can we start to talk about an expected release date?
>>
>> I guess you did a wonderful job for closing everything. As it is clean
>> enough, I think we could even skip the step of using a release branch
>> and we could simply tag the release candidates from the trunk. This
>> would simply imply refraining from any change which is not related to
>> the release for a few days.
>>
>> Someone has to volunteer to act as the release manager. The task is
>> simply to perform the few commands described for example here:
>> <http://wiki.apache.org/commons/UsingNexus>. The release manager also
>> signs the packages using a gpg key, which should be put in the global
>> KEYS file. This file can be retrieved using the following svn command:
>>
>> svn checkout --depth=immediates \
>>   https://[your-commiter-id]@svn.apache.org/repos/asf/commons/trunks-proper
>>
>> The artifacts for the release candidate must be made available and a
>> VOTE thread must be started on the dev list for at least 72 hours (see
>> <http://www.apache.org/foundation/voting.html>). There can be several
>> release candidate before a version finally goes out (when I release
>> version 2.0 I think, we needed 6 candidates ...). When the vote passes,
>> the exact artifacts which were used for voting will be published by
>> uploading the source and binary zip and tar files and by promoting the
>> maven artifacts with Nexus. Not a single bit is changed (this would
>> change the gpg signatures). This means that for example the release date
>> which appears in the release notes must be estimated before the vote
>> taking the voting delay into account (plus one or two days as a safety
>> margin) and it must be updated as each release candidate is cut.
>>
>> So there is no predefined release date until the vote finally passes.
>>
>> At the pace at which you go now, I would say we could target a first
>> release candidate early next week.
>>
>> Any volunteer as release manager ?
>
> OK, I started to try the commands listed in the "UsingNexus" file. Not
> everything works directly... [maven2 could not find a plugin, which led me

Which plugin?

> to upgrade to maven3, which printed a warning about "parent" being a broken
> project, etc.]
>
> I don't know maven (apart from the basics to build CM) so, it is not always
> obvious which are the mandatory steps and what result must be observed in
> order to check that everything went fine...
>
> For the encryption key: I was always advised against writing a passphrase in
> clear in a file; maven seems to support asking for the passphrase but when
> it prints:
> ---CUT---
> Enter passphrase: gpg: gpg-agent is not available in this session
> ---CUT---
> When I enter the passphrase, it just prints that same message again...

Works for me using Maven 2.2.1 and 3.0.4

Which version of gpg have you installed locally?

To test it out, just use

mvn gpg:sign

It will fail later as it needs package first.

> [I guess I'll create a dummy key and store the passphrase in "settings.xml"
> just for this to work...]

You can use encrypted passwords:

http://maven.apache.org/guides/mini/guide-encryption.html

Better than plain text, but still not ideal if your host is not
physically secure.

Can also store the master key on a removable USB stick.

>
> Regards,
> Gilles
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [Math] Toward releasing 3.0 ?

Gilles Sadowski
On Sat, Feb 25, 2012 at 12:25:49PM +0000, sebb wrote:

> On 25 February 2012 09:59, Gilles Sadowski <[hidden email]> wrote:
> > Hello.
> >
> >> >
> >> > How do we proceed from here in order to release 3.0? Cf. ticket MATH-746,
> >> > "Things to do before releasing 3.0".
> >>
> >> Sorry for being late on this.
> >>
> >> >
> >> > Can we start to talk about an expected release date?
> >>
> >> I guess you did a wonderful job for closing everything. As it is clean
> >> enough, I think we could even skip the step of using a release branch
> >> and we could simply tag the release candidates from the trunk. This
> >> would simply imply refraining from any change which is not related to
> >> the release for a few days.
> >>
> >> Someone has to volunteer to act as the release manager. The task is
> >> simply to perform the few commands described for example here:
> >> <http://wiki.apache.org/commons/UsingNexus>. The release manager also
> >> signs the packages using a gpg key, which should be put in the global
> >> KEYS file. This file can be retrieved using the following svn command:
> >>
> >> svn checkout --depth=immediates \
> >>   https://[your-commiter-id]@svn.apache.org/repos/asf/commons/trunks-proper
> >>
> >> The artifacts for the release candidate must be made available and a
> >> VOTE thread must be started on the dev list for at least 72 hours (see
> >> <http://www.apache.org/foundation/voting.html>). There can be several
> >> release candidate before a version finally goes out (when I release
> >> version 2.0 I think, we needed 6 candidates ...). When the vote passes,
> >> the exact artifacts which were used for voting will be published by
> >> uploading the source and binary zip and tar files and by promoting the
> >> maven artifacts with Nexus. Not a single bit is changed (this would
> >> change the gpg signatures). This means that for example the release date
> >> which appears in the release notes must be estimated before the vote
> >> taking the voting delay into account (plus one or two days as a safety
> >> margin) and it must be updated as each release candidate is cut.
> >>
> >> So there is no predefined release date until the vote finally passes.
> >>
> >> At the pace at which you go now, I would say we could target a first
> >> release candidate early next week.
> >>
> >> Any volunteer as release manager ?
> >
> > OK, I started to try the commands listed in the "UsingNexus" file. Not
> > everything works directly... [maven2 could not find a plugin, which led me
>
> Which plugin?

I couldn't tell you now because installing maven3 implied desinstalling
maven2.

>
> > to upgrade to maven3, which printed a warning about "parent" being a broken
> > project, etc.]

This is the warning from maven3:
---CUT---
[WARNING]
[WARNING] Some problems were encountered while building the effective model for org.apache.commons:commons-math3:jar:3.0-SNAPSHOT
[WARNING] 'build.plugins.plugin.version' for org.apache.maven.plugins:maven-idea-plugin is missing. @org.apache.commons:commons-parent:20, /home/eran/.m2/repository/org/apache/commons/commons-parent/20/commons-parent-20.pom, line 316, column 15
[WARNING]
[WARNING] It is highly recommended to fix these problems because they threaten the stability of your build.
[WARNING]
[WARNING] For this reason, future Maven versions might no longer support building such malformed projects.
[WARNING]
---CUT---

> >
> > I don't know maven (apart from the basics to build CM) so, it is not always
> > obvious which are the mandatory steps and what result must be observed in
> > order to check that everything went fine...
> >
> > For the encryption key: I was always advised against writing a passphrase in
> > clear in a file; maven seems to support asking for the passphrase but when
> > it prints:
> > ---CUT---
> > Enter passphrase: gpg: gpg-agent is not available in this session
> > ---CUT---
> > When I enter the passphrase, it just prints that same message again...
>
> Works for me using Maven 2.2.1 and 3.0.4

Maven version is also 3.0.4 here.

>
> Which version of gpg have you installed locally?

---CUT---
$ gpg --version
gpg (GnuPG) 1.4.11
Copyright (C) 2010 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

Home: ~/.gnupg
Supported algorithms:
Pubkey: RSA, RSA-E, RSA-S, ELG-E, DSA
Cipher: 3DES, CAST5, BLOWFISH, AES, AES192, AES256, TWOFISH, CAMELLIA128,
        CAMELLIA192, CAMELLIA256
Hash: MD5, SHA1, RIPEMD160, SHA256, SHA384, SHA512, SHA224
Compression: Uncompressed, ZIP, ZLIB, BZIP2
---CUT---

>
> To test it out, just use
>
> mvn gpg:sign
>
> It will fail later as it needs package first.

This seems to work (if this is where you expected it to fail); it produces:
---CUT---
[INFO] Scanning for projects...
[WARNING]
[WARNING] Some problems were encountered while building the effective model for org.apache.commons:commons-math3:jar:3.0-SNAPSHOT
[WARNING] 'build.plugins.plugin.version' for org.apache.maven.plugins:maven-idea-plugin is missing. @org.apache.commons:commons-parent:20, /home/eran/.m2/repository/org/apache/commons/commons-parent/20/commons-parent-20.pom, line 316, column 15
[WARNING]
[WARNING] It is highly recommended to fix these problems because they threaten the stability of your build.
[WARNING]
[WARNING] For this reason, future Maven versions might no longer support building such malformed projects.
[WARNING]
[INFO]
[INFO] ------------------------------------------------------------------------
[INFO] Building Commons Math 3.0-SNAPSHOT
[INFO] ------------------------------------------------------------------------
[INFO]
[INFO] --- maven-gpg-plugin:1.1:sign (default-cli) @ commons-math3 ---
GPG Passphrase: *******************************
[INFO] ------------------------------------------------------------------------
[INFO] BUILD FAILURE
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 11.345s
[INFO] Finished at: Mon Feb 27 13:05:15 CET 2012
[INFO] Final Memory: 9M/105M
[INFO] ------------------------------------------------------------------------
[ERROR] Failed to execute goal
org.apache.maven.plugins:maven-gpg-plugin:1.1:sign (default-cli) on project
commons-math3: The project artifact has not been assembled yet. Please do
not invoke this goal before the lifecycle phase "package". -> [Help 1]
[ERROR]
[ERROR] To see the full stack trace of the errors, re-run Maven with the -e
switch.
[ERROR] Re-run Maven using the -X switch to enable full debug logging.
[ERROR]
[ERROR] For more information about the errors and possible solutions, please read the following articles:
[ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/MojoFailureException
---CUT---

However, when I run

  $ mvn clean deploy -Papache-release -Ptest-deploy

I get:

---CUT---
[INFO] Parent project loaded from repository.
[INFO]
[INFO] --- maven-gpg-plugin:1.1:sign (default) @ commons-math3 ---

You need a passphrase to unlock the secret key for
user: "Gilles Sadowski (ASF code signing) <[hidden email]>"
1024-bit DSA key, ID 51D05641, created 2003-09-28

Enter passphrase: gpg: gpg-agent is not available in this session
                 
You need a passphrase to unlock the secret key for
user: "Gilles Sadowski (ASF code signing) <[hidden email]>"
1024-bit DSA key, ID 51D05641, created 2003-09-28

Enter passphrase: gpg: Invalid passphrase; please try again ...
                 
You need a passphrase to unlock the secret key for
user: "Gilles Sadowski (ASF code signing) <[hidden email]>"
1024-bit DSA key, ID 51D05641, created 2003-09-28

Enter passphrase: gpg: gpg-agent is not available in this session
[... and so on ...]
---CUT---

 
> > [I guess I'll create a dummy key and store the passphrase in "settings.xml"
> > just for this to work...]
>
> You can use encrypted passwords:
>
> http://maven.apache.org/guides/mini/guide-encryption.html

I had read it, but didn't think it would work for the
  <gpg.passphrase></gpg.passphrase>
tag.

Anyway, I encrypted the pass phrase using

 $ mvn --encrypt-password "my pass phrase"

put the result in the above tag, and got:
---CUT---
INFO] --- maven-gpg-plugin:1.1:sign (default) @ commons-math3 ---
gpg: skipped "Gilles Sadowski (ASF code signing) <[hidden email]>": bad passphrase
gpg: signing failed: bad passphrase
[INFO] ------------------------------------------------------------------------
[INFO] BUILD FAILURE
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 2:20.088s
[INFO] Finished at: Mon Feb 27 13:15:10 CET 2012
[INFO] Final Memory: 36M/370M
[INFO] ------------------------------------------------------------------------
[ERROR] Failed to execute goal
org.apache.maven.plugins:maven-gpg-plugin:1.1:sign (default) on project commons-math3: Exit code: 2 -> [Help 1]
---CUT---

>
> Better than plain text, but still not ideal if your host is not
> physically secure.

It would have been good enough if it worked.
I must be missing some additional configuration...

>
> Can also store the master key on a removable USB stick.

I'm not that paranoid ;-). It is encrypted, and stored in
"settings-security.xml", only readable by me. And it serves only to run
maven.
It's just that storing the pass phrase of a general-purpose encrypting key,
in clear text does not seem right.


Thanks for any enlightenment as to what could cause this problem,
Gilles

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

123