Re: [math] math 1.1 RCs and FCS
Phil Steitz wrote: I just published an RC4-SNAPSHOT jar to cvs.apache.org/repostory -- commons-math-1.1-RC4-SNAPSHOT.jar This is not (yet) an RC, as I am still working on BZ 37086. It does include other bug fixes after RC3, though, so I thought it best to publish a snapshot rather than pusing the RC3 jar. One correction below. References to java-repository mean the mirrored, ibiblio-synched official release repo at http://www.apache.org/dist/java-repository/. This is all described in the Repository Structure section in http://jakarta.apache.org/commons/building.html Thanks. - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
I just published an RC4-SNAPSHOT jar to cvs.apache.org/repostory -- commons-math-1.1-RC4-SNAPSHOT.jar This is not (yet) an RC, as I am still working on BZ 37086. It does include other bug fixes after RC3, though, so I thought it best to publish a snapshot rather than pusing the RC3 jar. One correction below. References to java-repository mean the mirrored, ibiblio-synched official release repo at http://www.apache.org/dist/java-repository/. This is all described in the Repository Structure section in http://jakarta.apache.org/commons/building.html Phil On 10/15/05, Mauro Talevi [EMAIL PROTECTED] wrote: Phil Steitz wrote: There is general consensus that it is OK to publish snapshots to our internal repository at cvs.apache.org/repository (aka java-repository), but not to ibibilio. The snaps on cvs.apache.org are not guaranteed to remain there and users should not be encouraged to use them in production applications. Others can jump in here and state different opinions, but I think at least within commons we are in agreement that we publish only released jars to ibiblio. Some apache projects vote on and release formal betas and some separate release votes from stability votes. At this point, in commons, we follow the process described here: http://jakarta.apache.org/commons/releases, which is to fix contents of the release via a release plan, then cut RCs for download / inspection from people.apache.org, iterating until a full release vote passes and we publish to apache mirros and ibibilio. This process and the Jakarta voting / release guidelines really predate maven. It may be reasonable to add a push RC jars to java-repository to the RC process. If other commons committers are OK with pushing RC jars to java-repository, I will do this for the current [math] RC and update http://jakarta.apache.org/commons/releases to add a publish to java-repository step to the RC step. This creates a little more work for RMs, but if it means the RCs get more / better testing and we are all OK with it from support / oversight perspective, I am OK with it That'd be great! java-repository is just as easy to add it to the list of maven remote repos. Thanks! Cheers - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
Phil Steitz wrote: I understand, unfortunately this is against our release policy (at least publishing to ibiblio). I will try to get the actual release out ASAP. Just to understand - this release policy seems specific to the commons math project, since other commons projects seem to deploy RCs and snapshots. If so - is there any specific reason for it? If not on ibiblio - perhaps another URL on ASF where to deploy the jars? They would be not be considered public releases - just like the zip and tar.gz distributions. IMO it would make it a lot easier to test the RCs in real production projects and would benefit the commons math project. Cheers - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
On 10/15/05, Mauro Talevi [EMAIL PROTECTED] wrote: Phil Steitz wrote: I understand, unfortunately this is against our release policy (at least publishing to ibiblio). I will try to get the actual release out ASAP. Just to understand - this release policy seems specific to the commons math project, since other commons projects seem to deploy RCs and snapshots. If so - is there any specific reason for it? If not on ibiblio - perhaps another URL on ASF where to deploy the jars? snip/ IIUC, http://cvs.apache.org/dist/jakarta/ may be used for this purpose. -Rahul They would be not be considered public releases - just like the zip and tar.gz distributions. IMO it would make it a lot easier to test the RCs in real production projects and would benefit the commons math project. Cheers - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
On 10/15/05, Mauro Talevi [EMAIL PROTECTED] wrote: Phil Steitz wrote: I understand, unfortunately this is against our release policy (at least publishing to ibiblio). I will try to get the actual release out ASAP. Just to understand - this release policy seems specific to the commons math project, since other commons projects seem to deploy RCs and snapshots. If so - is there any specific reason for it? If not on ibiblio - perhaps another URL on ASF where to deploy the jars? We do have an internal maven repo at cvs.apache.org/repository. Assuming others are OK (see below), I will push the RC3 jars there. They would be not be considered public releases - just like the zip and tar.gz distributions. IMO it would make it a lot easier to test the RCs in real production projects and would benefit the commons math project. There is general consensus that it is OK to publish snapshots to our internal repository at cvs.apache.org/repository (aka java-repository), but not to ibibilio. The snaps on cvs.apache.org are not guaranteed to remain there and users should not be encouraged to use them in production applications. Others can jump in here and state different opinions, but I think at least within commons we are in agreement that we publish only released jars to ibiblio. Some apache projects vote on and release formal betas and some separate release votes from stability votes. At this point, in commons, we follow the process described here: http://jakarta.apache.org/commons/releases, which is to fix contents of the release via a release plan, then cut RCs for download / inspection from people.apache.org, iterating until a full release vote passes and we publish to apache mirros and ibibilio. This process and the Jakarta voting / release guidelines really predate maven. It may be reasonable to add a push RC jars to java-repository to the RC process. If other commons committers are OK with pushing RC jars to java-repository, I will do this for the current [math] RC and update http://jakarta.apache.org/commons/releases to add a publish to java-repository step to the RC step. This creates a little more work for RMs, but if it means the RCs get more / better testing and we are all OK with it from support / oversight perspective, I am OK with it Phil - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: [math] math 1.1 RCs and FCS
You can publish snapshot on this repository which isn't synchronized with
ibiblio.
It's where we put all our snapshots (comes from project.properties.sample in
commons-build)
# Repository to deploy snapshots
maven.repo.apache.snapshots=scp://cvs.apache.org
maven.repo.apache.snapshots.directory=/www/cvs.apache.org/repository
maven.repo.apache.snapshots.username=${maven.repo.apache.releases.username}
maven.repo.apache.snapshots.privatekey=${maven.repo.apache.releases.privatekey}
maven.repo.apache.snapshots.passphrase=${maven.repo.apache.releases.passphrase}
maven.repo.apache.snapshots.group=${maven.repo.apache.releases.group}
Arnaud
-Message d'origine-
De : news [mailto:[EMAIL PROTECTED] De la part de Mauro Talevi
Envoyé : samedi 15 octobre 2005 14:10
À : commons-dev@jakarta.apache.org
Objet : Re: [math] math 1.1 RCs and FCS
Phil Steitz wrote:
I understand, unfortunately this is against our release policy (at
least publishing to ibiblio). I will try to get the actual release
out ASAP.
Just to understand - this release policy seems specific to
the commons math project, since other commons projects seem
to deploy RCs and snapshots. If so - is there any specific
reason for it?
If not on ibiblio - perhaps another URL on ASF where to
deploy the jars?
They would be not be considered public releases - just like
the zip and tar.gz distributions.
IMO it would make it a lot easier to test the RCs in real
production projects and would benefit the commons math project.
Cheers
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
On 10/15/05, Arnaud HERITIER [EMAIL PROTECTED] wrote: You can publish snapshot on this repository which isn't synchronized with ibiblio. It's where we put all our snapshots (comes from project.properties.sample in commons-build) # Repository to deploy snapshots maven.repo.apache.snapshots=scp://cvs.apache.org maven.repo.apache.snapshots.directory=/www/cvs.apache.org/repository snip/ So, whats /www/cvs.apache.org/dist/jakarta for? I see snapshots there, and I remember it being referred to as the place to put snapshots. Thanks! -Rahul - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
Phil Steitz wrote: There is general consensus that it is OK to publish snapshots to our internal repository at cvs.apache.org/repository (aka java-repository), but not to ibibilio. The snaps on cvs.apache.org are not guaranteed to remain there and users should not be encouraged to use them in production applications. Others can jump in here and state different opinions, but I think at least within commons we are in agreement that we publish only released jars to ibiblio. Some apache projects vote on and release formal betas and some separate release votes from stability votes. At this point, in commons, we follow the process described here: http://jakarta.apache.org/commons/releases, which is to fix contents of the release via a release plan, then cut RCs for download / inspection from people.apache.org, iterating until a full release vote passes and we publish to apache mirros and ibibilio. This process and the Jakarta voting / release guidelines really predate maven. It may be reasonable to add a push RC jars to java-repository to the RC process. If other commons committers are OK with pushing RC jars to java-repository, I will do this for the current [math] RC and update http://jakarta.apache.org/commons/releases to add a publish to java-repository step to the RC step. This creates a little more work for RMs, but if it means the RCs get more / better testing and we are all OK with it from support / oversight perspective, I am OK with it That'd be great! java-repository is just as easy to add it to the list of maven remote repos. Thanks! Cheers - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: [math] math 1.1 RCs and FCS
I think that http://cvs.apache.org/dist/ is used to distribute packages. It doesn't follow the maven repository layout. The repository to store snapshots for maven is http://cvs.apache.org/repository/ Arnaud -Message d'origine- De : Rahul Akolkar [mailto:[EMAIL PROTECTED] Envoyé : samedi 15 octobre 2005 17:53 À : Jakarta Commons Developers List Objet : Re: [math] math 1.1 RCs and FCS On 10/15/05, Arnaud HERITIER [EMAIL PROTECTED] wrote: You can publish snapshot on this repository which isn't synchronized with ibiblio. It's where we put all our snapshots (comes from project.properties.sample in commons-build) # Repository to deploy snapshots maven.repo.apache.snapshots=scp://cvs.apache.org maven.repo.apache.snapshots.directory=/www/cvs.apache.org/repository snip/ So, whats /www/cvs.apache.org/dist/jakarta for? I see snapshots there, and I remember it being referred to as the place to put snapshots. Thanks! -Rahul - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[ALL] Maven snapshots repository (was Re: [math] math 1.1 RCs and FCS)
On 10/15/05, Arnaud HERITIER [EMAIL PROTECTED] wrote: I think that http://cvs.apache.org/dist/ is used to distribute packages. It doesn't follow the maven repository layout. The repository to store snapshots for maven is http://cvs.apache.org/repository/ snip/ Thanks Arnaud. I have the need to make snapshots available for one of the sandbox projects, so this was a timely bit of clarification for me. -Rahul - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
Phil, Phil Steitz wrote: There are two open bugs holding up the release: http://issues.apache.org/bugzilla/show_bug.cgi?id=37019 (trivial) http://issues.apache.org/bugzilla/show_bug.cgi?id=37086 (requires some work) I will try to get these sorted this weekend. Patches, feedback welcome! Any [math] committers out there, pls review mails and bug report on Complex classes and weigh in so we can resolve and get the release out. Thanks for the update. Bugzilla server is down ATM. But I've been following the issue 37019. It's still not 100% clear what the correct scenario is: should or should the NaNs have an impact on the result? Two comments you made seem slightly inconsistent. One: The general policy in [math] is to use NaN consistently with floating point arithemtic specs (IEEE 754 for real, C99x Annex G for complex arithmetic) so for statistics such as Sum, Mean, etc. that involve arithmetic computations, NaN values force the return value to be NaN. Two: This is definitely a bug, as the javadoc indicates clearly that NaN values should have no impact on the result: The result is NaN iff all values are NaN (i.e. NaN values have no impact on the value of the statistic). Could you clarify please? Is the javadoc wrong or the impl? We created a release branch once the contents of the 1.1 release were fixed. That branch is MATH_1_1. The svn trunk version includes some numerical analysis enhancements that will not be included in 1.1. That's why the trunk version says 1.2 That's fine, but I guess what I was also asking is whether you have a policy of releasing jars labeled 1.1-rc-* (I haven't been able to find them) or only the final release. Cheers - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
On 10/14/05, Mauro Talevi [EMAIL PROTECTED] wrote:
Phil,
Phil Steitz wrote:
There are two open bugs holding up the release:
http://issues.apache.org/bugzilla/show_bug.cgi?id=37019 (trivial)
http://issues.apache.org/bugzilla/show_bug.cgi?id=37086 (requires some work)
I will try to get these sorted this weekend. Patches, feedback
welcome! Any [math] committers out there, pls review mails and bug
report on Complex classes and weigh in so we can resolve and get the
release out.
Thanks for the update. Bugzilla server is down ATM.
But I've been following the issue 37019.
It's still not 100% clear what the correct scenario is:
should or should the NaNs have an impact on the result?
Two comments you made seem slightly inconsistent.
One:
The general policy in [math] is to use NaN consistently with floating
point arithemtic specs (IEEE 754 for real, C99x Annex G for complex
arithmetic) so for statistics such as Sum, Mean, etc. that involve
arithmetic computations, NaN values force the return value to be NaN.
Two:
This is definitely a bug, as the javadoc indicates clearly that NaN
values should have no impact on the result:
The result is NaN iff all values are NaN (i.e. NaN values have no
impact on the value of the statistic).
Could you clarify please? Is the javadoc wrong or the impl?
What is incorrect is the impl, assuming the test results presented in
the bug report are correct (I have not turned these into Junit test
cases yet to verify). The iff in the javadoc means if and only if
- so, e.g. the min of {NaN, 0} should be 0, not NaN.
We created a release branch once the contents of the 1.1 release were
fixed. That branch is MATH_1_1. The svn trunk version includes some
numerical analysis enhancements that will not be included in 1.1.
That's why the trunk version says 1.2
That's fine, but I guess what I was also asking is whether you have
a policy of releasing jars labeled 1.1-rc-* (I haven't been able to find
them) or only the final release.
What you are looking for should be in the tars/zips here
http://people.apache.org/~psteitz/commons-math/
Note that these are not public apache releases, but release
candidates for internal evaluation.
Thanks for the feedback!
Phil
Cheers
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
Phil Steitz wrote:
What is incorrect is the impl, assuming the test results presented in
the bug report are correct (I have not turned these into Junit test
cases yet to verify). The iff in the javadoc means if and only if
- so, e.g. the min of {NaN, 0} should be 0, not NaN.
attached patch
http://issues.apache.org/bugzilla/attachment.cgi?id=16699
What you are looking for should be in the tars/zips here
http://people.apache.org/~psteitz/commons-math/
Note that these are not public apache releases, but release
candidates for internal evaluation.
Publishing RCs jars as well (aside from the distributions)
so that maven can look them up (eg ibiblio or somewhere else at ASF)
would also be very useful.
It would enable users to quickly upgrade their maven projects
without needing to download, unpack, etc ...
Cheers
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
On 10/14/05, Mauro Talevi [EMAIL PROTECTED] wrote:
Phil Steitz wrote:
What is incorrect is the impl, assuming the test results presented in
the bug report are correct (I have not turned these into Junit test
cases yet to verify). The iff in the javadoc means if and only if
- so, e.g. the min of {NaN, 0} should be 0, not NaN.
attached patch
http://issues.apache.org/bugzilla/attachment.cgi?id=16699
Thanks!
What you are looking for should be in the tars/zips here
http://people.apache.org/~psteitz/commons-math/
Note that these are not public apache releases, but release
candidates for internal evaluation.
Publishing RCs jars as well (aside from the distributions)
so that maven can look them up (eg ibiblio or somewhere else at ASF)
would also be very useful.
It would enable users to quickly upgrade their maven projects
without needing to download, unpack, etc ...
I understand, unfortunately this is against our release policy (at
least publishing to ibiblio). I will try to get the actual release
out ASAP.
Phil
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: [math] math 1.1 RCs and FCS
On 10/13/05, Mauro Talevi [EMAIL PROTECTED] wrote: Hello, is there any other place where the 1.1 RCs can be found? http://people.apache.org/~psteitz/commons-math/ http://www.apache.org/dist/jakarta/commons/math/binaries/ only contains 1.0. Also, I was looking to get the latest on the 1.1 release. Is there anything holding up the 1.1 final? There are two open bugs holding up the release: http://issues.apache.org/bugzilla/show_bug.cgi?id=37019 (trivial) http://issues.apache.org/bugzilla/show_bug.cgi?id=37086 (requires some work) I will try to get these sorted this weekend. Patches, feedback welcome! Any [math] committers out there, pls review mails and bug report on Complex classes and weigh in so we can resolve and get the release out. Phil There was a vote about a month ago, which only resulted in a minor build issue being fixed. Since then no other news - either pro or con release. Chasing down findbugs complaints led to review of Complex equals and arithmetic, which brought up 37086. Oddly, the current version is 1.2-SNAPSHOT. Have I missed something? We created a release branch once the contents of the 1.1 release were fixed. That branch is MATH_1_1. The svn trunk version includes some numerical analysis enhancements that will not be included in 1.1. That's why the trunk version says 1.2 Phil - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
