Understood. Thanks,
-Rob
> On Jul 8, 2024, at 4:41 PM, Alex Herbert wrote:
>
> Hi Rob,
>
> This behaviour is documented in the release note.
>
> The tests for randomness use a random seed. So they fail with a given
> probability. We have a set number of reruns configured. A few failures get
>
Hi Rob,
This behaviour is documented in the release note.
The tests for randomness use a random seed. So they fail with a given
probability. We have a set number of reruns configured. A few failures get
logged as a flake. If they fail 3 times then the build will fail. With a
p-value of 0.001 this
I was running the build on java 21 and got this:
Results:
[INFO]
[WARNING] Flakes:
[WARNING]
org.apache.commons.rng.core.ProvidersCommonParametricTest.testUniformNextBytesPartialBuffer(UniformRandomProvider)[8]
[ERROR] Run 1:
ProvidersCommonParametricTest.testUniformNextBytesPartialBuffer:124
Hi Gary,
Thanks for the feedback. I'll fix the docs module tomorrow. It is only for
an aggregate javadoc report for the site. It depends on all the other jar
modules and builds a single aggregate javadoc. This requires it to be a jar
artifact in the pom. But there is no output jar. I'll disable ja
Oops, actually 'mvn clean install -DskipTests japicmp:cmp' fails with
[INFO] Reactor Summary for Apache Commons RNG 1.6:
[INFO]
[INFO] Apache Commons RNG . SUCCESS [ 21.207 s]
[INFO] Apache Commons RNG Client API .. SUCCESS [ 24.159 s]
[INFO] Apa
Hello,
+1
This email contains this link that returns a 404 for
https://dist.apache.org/repos/dist/dev/commons/rng/1.6-RC1/site/changes-report.html
I see that for some reason, the site is in a personal folder instead of
https://dist.apache.org/repos/dist/dev/commons/rng/1.6-RC1/site, so I imagi
On 08.07.2024 11:18, Julian Reschke wrote:
On 07.07.2024 17:12, Gary Gregory wrote:
We don't want to break binary compatibility within the 4.x release line.
Gary
...
Would we have to?
OK, so this is about the return values.
It's still unfortunate, because it makes it harder to convert exis
On 07.07.2024 17:12, Gary Gregory wrote:
We don't want to break binary compatibility within the 4.x release line.
Gary
...
Would we have to?
Best regards, Julian
-
To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org
F
