Re: [Math] Change on branch "develop": Commons Math DeveloperBranch - Build # 13 - Failure

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

Re: [Math] Change on branch "develop": Commons Math DeveloperBranch - Build # 13 - Failure

Gilles Sadowski
Hi.

On Sat, 14 May 2016 17:19:38 +0000 (UTC), Apache Jenkins Server wrote:
> The Apache Jenkins build system has built Commons Math
> DeveloperBranch (build #13)
>
> Status: Failure
>
> Check console output at
> https://builds.apache.org/job/Commons%20Math%20DeveloperBranch/13/ to
> view the results.

When checking the console output, one sees that this failure is not
caused by the commits.

Can Jenkins be configured to distinguish between human cause and
infrastructure cause?
[E.g. the failing to "deploy" should be reported with a "warning",
indicating that the build itself had been successful.]


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] Change on branch "develop": Commons Math DeveloperBranch - Build # 13 - Failure

Rob Tompkins
The problem here seems to be that https://repository.apache.org is unavailable (serving 503s). So this failure is not related to the commits on the branch. If it builds locally, then I would not be all that concerned.

-Rob

> On May 14, 2016, at 1:33 PM, Gilles <[hidden email]> wrote:
>
> Hi.
>
>> On Sat, 14 May 2016 17:19:38 +0000 (UTC), Apache Jenkins Server wrote:
>> The Apache Jenkins build system has built Commons Math
>> DeveloperBranch (build #13)
>>
>> Status: Failure
>>
>> Check console output at
>> https://builds.apache.org/job/Commons%20Math%20DeveloperBranch/13/ to
>> view the results.
>
> When checking the console output, one sees that this failure is not
> caused by the commits.
>
> Can Jenkins be configured to distinguish between human cause and
> infrastructure cause?
> [E.g. the failing to "deploy" should be reported with a "warning",
> indicating that the build itself had been successful.]
>
>
> 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
|

[Math] Regarding Naming Conventions for Functions Multiplied By 1/x

Rob Tompkins
In reply to this post by Gilles Sadowski
Hi,

I’m currently working on MATH-1359 <https://issues.apache.org/jira/browse/MATH-1358?jql=project%20=%20MATH%20AND%20assignee%20=%20currentUser()%20AND%20resolution%20=%20Unresolved> which is working an analog of the relationship between sin(x) and sinx(x) = sin(x)/x but instead for log(1+x) = log1p(x). The  sine function has the term "cardinal" associated with it when we actually perform
cardinal sine (x) = sinc(x) = sin(x)/x
however, I don't seem to see any clearly defined analog, in a quick glance at the literature , for the log function. Do you have any thoughts on what the name should be? Clearly first intuition would be
log1pc(x),
but that only follows from the
sinc(x)
naming convention and isn't necessarily rooted in the literature. Another thought that occurred to me was to name such a function
log1pMultInv(x)
implying that we're also multiplying by the multiplicative inverse. But, that is not rooted in the literature either. I would love to hear any suggestions that folks have for a naming convention that is intuitive here .

Many thanks and all the best,
-Rob


Reply | Threaded
Open this post in threaded view
|

Re: [Math] Regarding Naming Conventions for Functions Multiplied By 1/x

Gilles Sadowski
On Sun, 15 May 2016 08:04:49 -0400, Rob Tompkins wrote:

> Hi,
>
> I’m currently working on MATH-1359
>
> <https://issues.apache.org/jira/browse/MATH-1358?jql=project%20=%20MATH%20AND%20assignee%20=%20currentUser()%20AND%20resolution%20=%20Unresolved>
> which is working an analog of the relationship between sin(x) and
> sinx(x) = sin(x)/x but instead for log(1+x) = log1p(x). The  sine
> function has the term "cardinal" associated with it when we actually
> perform
> cardinal sine (x) = sinc(x) = sin(x)/x
> however, I don't seem to see any clearly defined analog, in a quick
> glance at the literature , for the log function. Do you have any
> thoughts on what the name should be? Clearly first intuition would be
> log1pc(x),
> but that only follows from the
> sinc(x)
> naming convention and isn't necessarily rooted in the literature.
> Another thought that occurred to me was to name such a function
> log1pMultInv(x)
> implying that we're also multiplying by the multiplicative inverse.
> But, that is not rooted in the literature either. I would love to
> hear
> any suggestions that folks have for a naming convention that is
> intuitive here .

Since there is no standard (and perhaps not much use outside of CM),
I'd suggest to spell it out in full: i.e.

   Log1pMultiplyInverse

[Classes "Inverse" and "Multiply" are defined in the package
"o.a.c.m.analysis.function".]

Regards,
Gilles

>
> Many thanks and all the best,
> -Rob


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