osdir.com


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Vote] Release Apache Commons RNG v1.2 (RC1)


On Thu, 6 Dec 2018 11:37:25 +0000, sebb wrote:
On Thu, 6 Dec 2018 at 10:58, Gilles <gilles@xxxxxxxxxxxxxxxxxxxxx> wrote:

On Wed, 5 Dec 2018 21:29:26 -0500, Rob Tompkins wrote:
> With:
>
> $ mvn -version
> Apache Maven 3.5.4 (1edded0938998edf8bf061f1ceb3cfdeccf443fe;
> 2018-06-17T14:33:14-04:00)
> Maven home: /usr/local/Cellar/maven/3.5.4/libexec
> Java version: 9.0.4, vendor: Oracle Corporation, runtime:
> /Library/Java/JavaVirtualMachines/jdk-9.0.4.jdk/Contents/Home
> Default locale: en_US, platform encoding: UTF-8
> OS name: "mac os x", version: "10.14.1", arch: "x86_64", family:
> "mac"
>
> I get this:
>
> [ERROR] Failed to execute goal
> org.apache.maven.plugins:maven-site-plugin:3.7.1:site (default-site)
> on project commons-rng-sampling: Error generating
> maven-javadoc-plugin:3.0.1:javadoc report:
> [ERROR] Exit code: 1 -
>
> /Users/tompkicr/Desktop/commons-rng/source/commons-rng-1.2-src/commons-rng-sampling/src/main/java/org/apache/commons/rng/sampling/distribution/SmallMeanPoissonSampler.java:35:
> error: reference not found
> [ERROR]  * For large means, {@link LargePoissonSampler} should be
> used instead.

This is a type in the Javadoc (should have been
"LargeMeanPoissonSampler", fixed now).
In my environment, it only triggered a warning.
Does the policy consider this as a blocker?

I don't think Javadoc errors are blockers per se, however a failing
build seems to me like a blocker unless there are extenutating
circumstances.
For example a very rarely used JVM/OS combination.

Whilst it may be tedious to fix this, if we don't, there are likely to be ongoing complaints about the issue that will have to be dealt with,
possibly causing more work overall.

OK.  I'll make a new RC.

But the thing is, although we expect support from automated/reproducible build (through common maven config), every time we advocate/work towards
a single way to manage Commons projects, the discussion dies off with
TMTOWTDI.[1]
In my experience, this is what makes it painful to handle releases (for
manager and reviewers alike).[2]
The extenuating circumstance is that we don't want to enforce convergence
of the release process, and then complain that it is tedious... :-}

Regards,
Gilles

[1] https://en.wiktionary.org/wiki/TMTOWTDI#English
[2] And also the Jenkins configuration (that now seem to break with
    every update of the CI system and/or POM upgrade).


[...]


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@xxxxxxxxxxxxxxxxxx
For additional commands, e-mail: dev-help@xxxxxxxxxxxxxxxxxx