You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Gilles Sadowski <gi...@harfang.homelinux.org> on 2013/01/09 01:22:11 UTC
[Math][ALL] Preparing 3.1.1
Hi.
I got another report that CM 3.1 is unusable due to
https://issues.apache.org/jira/browse/MATH-924
I'm going to prepare 3.1.1 ASAP.
(1)
However, the new "DiagonalMatrix" class is a skeleton currently, with
just enough functionality to fix the bug (hopefully).
[It even lacks unit tests. Help appreciated to fill that void.]
Even with unit tests, I'm tempted to mark this class as "deprecated" since
we did not seriously decide which way to go with the design of matrix
implementations. OK?
(2)
>From what I read on the ML, I guess the svn/CMS/site does not work smoothly
yet.
For this bug-fix release, with no API change, can we skip the web site
generation and vote only on the tag and JAR artefacts?
[I wouldn't even know how people will be able to download the source
distribution...]
Thanks,
Gilles
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [Math][ALL] Preparing 3.1.1
Posted by sebb <se...@gmail.com>.
On 9 January 2013 00:34, Phil Steitz <ph...@gmail.com> wrote:
> On 1/8/13 4:22 PM, Gilles Sadowski wrote:
>> Hi.
>>
>> I got another report that CM 3.1 is unusable due to
>> https://issues.apache.org/jira/browse/MATH-924
>>
>> I'm going to prepare 3.1.1 ASAP.
>>
>> (1)
>> However, the new "DiagonalMatrix" class is a skeleton currently, with
>> just enough functionality to fix the bug (hopefully).
>> [It even lacks unit tests. Help appreciated to fill that void.]
>> Even with unit tests, I'm tempted to mark this class as "deprecated" since
>> we did not seriously decide which way to go with the design of matrix
>> implementations. OK?
>
> I would not release it new marked deprecated. One option would be
> to move it to the package where it is used for now and give it
> package scope (so not part of public API). I would be OK just
> releasing as is, personally. I will take a closer look tonight; but
> the API looks OK to me, so I don't see big risk releasing it.
>>
>> (2)
>> >From what I read on the ML, I guess the svn/CMS/site does not work smoothly
>> yet.
>> For this bug-fix release, with no API change, can we skip the web site
>> generation and vote only on the tag and JAR artefacts?
>> [I wouldn't even know how people will be able to download the source
>> distribution...]
> I suspect replication from p.a.o still works, as we did open the
> requested JIRA and we are waiting on infra.
/www/www.apache.org/dist/commons is still group owned by commons (it
will be svnwc when svnpubsub is active).
Likewise for the site, /www/commons.apache.org/math
> We can ask for help
> getting the site replicated if it is still not working when the
> release is ready for the mirrors. We should vote on full artifacts.
>
> Phil
>>
>>
>> Thanks,
>> Gilles
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
>> For additional commands, e-mail: dev-help@commons.apache.org
>>
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [Math][ALL] Preparing 3.1.1
Posted by Phil Steitz <ph...@gmail.com>.
On 1/8/13 4:22 PM, Gilles Sadowski wrote:
> Hi.
>
> I got another report that CM 3.1 is unusable due to
> https://issues.apache.org/jira/browse/MATH-924
>
> I'm going to prepare 3.1.1 ASAP.
>
> (1)
> However, the new "DiagonalMatrix" class is a skeleton currently, with
> just enough functionality to fix the bug (hopefully).
> [It even lacks unit tests. Help appreciated to fill that void.]
> Even with unit tests, I'm tempted to mark this class as "deprecated" since
> we did not seriously decide which way to go with the design of matrix
> implementations. OK?
I would not release it new marked deprecated. One option would be
to move it to the package where it is used for now and give it
package scope (so not part of public API). I would be OK just
releasing as is, personally. I will take a closer look tonight; but
the API looks OK to me, so I don't see big risk releasing it.
>
> (2)
> >From what I read on the ML, I guess the svn/CMS/site does not work smoothly
> yet.
> For this bug-fix release, with no API change, can we skip the web site
> generation and vote only on the tag and JAR artefacts?
> [I wouldn't even know how people will be able to download the source
> distribution...]
I suspect replication from p.a.o still works, as we did open the
requested JIRA and we are waiting on infra. We can ask for help
getting the site replicated if it is still not working when the
release is ready for the mirrors. We should vote on full artifacts.
Phil
>
>
> Thanks,
> Gilles
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org