Re: SCM plugin with Multi-Branch Pipeline throwing InaccessibleObjectException

['Jesse Glick' via Jenkins Developers]

How are you running Jenkins when you see this? Should not occur when either
using `mvn hpi:run` nor `mvn test` on a `JenkinsRule`-based test so far as
I know.
https://github.com/jenkinsci/plugin-pom/blob/692c7757f468ea6b1ecf52a39edb8379020816c3/pom.xml#L65-L70
and associated setup should be passing the right JVM argument to allow this
reflection.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3fMOePYaXEa6q4uFc-%2BJMB8dN%2B80-aLN7V4GtnMV_muQ%40mail.gmail.com.

Re: GitHub repo configuration for PR and CI?

['Jesse Glick' via Jenkins Developers]

On Wed, Apr 17, 2024 at 10:22 AM Antoine Musso  wrote:

> I would like to ensure CI passes before a pull request is merged. To do
> so, I have added a rule to request the change to be updated (== rebased)


If you mean *Require branches to be up to date before merging* then I would
not recommend it. This defends against a scenario (logical merge conflict
without textual merge conflict) which in practice is extremely rare, and
tends to interfere with routine work. If you are still concerned, I would
rather suggest checking the newer option * Require** merge queue* which
offers the same safety without the downsides.


> when doing a maven release:perform, the git push is refused


Not if you have admin rights to the repo, though you may instead consider
https://www.jenkins.io/doc/developer/publishing/releasing-cd/ which would
bypass any such issues.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3ejMYnVxLuqv7oCJj9BPqX1x0Pj6TgVTFqoauyr6wZHg%40mail.gmail.com.

Re: Jenkins.getInstance().getAllItems(ParameterizedJob.class) slow in large installations

['Jesse Glick' via Jenkins Developers]

On Sat, Apr 13, 2024 at 1:13 AM Tomas Bjerre 
wrote:

> - Is there a faster way
> to Jenkins.getInstance().getAllItems(ParameterizedJob.class); ?
>

Yes,
https://javadoc.jenkins.io/hudson/model/ItemGroup.html#allItems(java.lang.Class)
is faster since it avoids the need to sort. Also if this code is running as
a particular user identity yet the particular block of logic should apply
regardless of the user’s permissions to learn of the existence of
particular jobs, you can impersonate `ACL.SYSTEM` which will be faster
since a typical `AuthorizationStrategy` implementation has a quick path for
this principal.

Without knowing what the plugin does or why it is enumerating all projects
in a performance-sensitive call site, I cannot offer further advice, but
creating a cache smells like a poor solution. Can the plugin be redesigned
to look up the required project by name rather than conducting a search?

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr14JQ-4HvKu-S04_NOo1dtQdEZV6rTwVbKiXmjDX8nVSA%40mail.gmail.com.

Re: Modernize core dependency json-lib library

['Jesse Glick' via Jenkins Developers]

On Wed, Apr 10, 2024 at 6:27 PM Basil Crow  wrote:

> Even if we did want to migrate core and plugins to […] a different JSON
> library […], we'd have to continue to support JSON-lib under the net.sf
> package namespace during the transition period at the very least.
>

I think it is worse than that, because jsonlib types are exposed in core
API signatures. For example https://issues.jenkins.io/browse/JENKINS-67344
cannot be solved merely by switching to a reasonable parser, since you
would still need to provide compatibility for
https://javadoc.jenkins.io/hudson/model/UpdateSite.html#getJSONObject() and
similar. So in the short term, being able to deliver patches to jsonlib
would be helpful.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3D%2B7ZmhxR7eBSfr5UfTsJ8F50KfZBvup7g4cUCtpsS_Q%40mail.gmail.com.

Re: Windows-arm64 native support

['Jesse Glick' via Jenkins Developers]

On Thu, Mar 28, 2024 at 8:02 AM Pierrick Bouvier <
pierrick.bouv...@linaro.org> wrote:

> for winp, I'm not sure what we are missing by not being able to load this
> DLL (less control on processes launched?)
>

I think it would be worthwhile to search for usages of winp in
@jenkinsci—probably
mostly confined to Jenkins core, such as in `ProcessTree`, though plugins
are not blocked from using it—and checking whether it could simply be
retired. Not loading native libraries is generally desirable anyway. Java 9
introduced richer process control APIs which might suffice for the
particular use cases in Jenkins. The main usage of all this code that I am
aware of is to kill off whole process subtrees during builds, typically on
an agent. Say you had a Pipeline such as

node('something-remote') {
  sh '''
export DISPLAY=:999
xvfb &
make gui-tests
  '''
}

with no join on the background task. By the close of the `node` block, it
is expected that all processes started during the user script have been
killed off. This is currently done by passing a special random “cookie” as
an environment variable to the main shell script, which should be inherited
by descendant processes, and finally searching for any remaining processes
with that environment set and sending them a termination signal. Maybe this
could be replaced with a `ProcessHandle.descendants` call? (`durable-task`
runs the user process in the background, however, and this can survive
parent JVM sessions.)

Even if there is not a simple solution using the Java Platform, I am
guessing winp specifically could be removed in favor of executing `wmic`
commands or something like that, or that could be a fallback in case of an
unsupported binary platform.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3HJnh8XuzRebGexFWmWdn5TFT2JiUzgHwmKSxPzP%2B%2B_Q%40mail.gmail.com.

Re: Stop publishing fat plugin aws-java-sdk

['Jesse Glick' via Jenkins Developers]

On Thu, Mar 21, 2024 at 9:20 AM Vincent Latombe 
wrote:

> prompt any remaining plugin that may depend on the fat plugin to do the
> right thing, and implement option 2.
>

Did you already file PRs for _all_ `@jenkinsci` plugins using the fat dep?
Otherwise it might not be apparent that an update is required, unless they
were consuming the fat dep via `bom` and see that proposed `bom` updates
fail (“no version specified for…”).

At some point we would need to “tombstone” the fat plugin, deleting all the
bundled JARs and marking it deprecated so that admins with it currently
installed would see that it could be uninstalled, or at least if they
blindly update, will not have all the JARs on disk.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2cUXzmTZhi7rAueM4qfq2pyQ1TD-HqhUQyQ_n%3DeQDs3A%40mail.gmail.com.

Re: Exclusive JEP-229 CD mode announcement

['Jesse Glick' via Jenkins Developers]

I guess this should be mentioned in
https://www.jenkins.io/doc/developer/publishing/releasing-cd/#enable-cd-permissions
and/or
https://www.jenkins.io/doc/developer/publishing/releasing-cd/#fallback ?

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0N-%3DsDbOHy2V_BeOvq8FnXr3t5-JG9UTq7c%2BUahz2ugw%40mail.gmail.com.

Re: Banned dependencies error compiling SAML plugin PR

['Jesse Glick' via Jenkins Developers]

On Wed, Dec 13, 2023 at 5:56 AM Ullrich Hafner 
wrote:

> I think our tool chain will accept Java 17 dependencies if you explicitly
> declare it as supported by setting the java.version in you pom.
>

If you meant java.level, no, this is gone as of
https://github.com/jenkinsci/plugin-pom/releases/tag/plugin-4.40

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0H-T7OQ8bUcw7KUFkA7bi8YZMXeY-n%2B4O_xkP%2BR5-%3DvA%40mail.gmail.com.

Re: License for library plugins

['Jesse Glick' via Jenkins Developers]

On Tue, Dec 12, 2023 at 7:44 AM Mark Waite 
wrote:

> The changes made to wrap the library into the plugin are usually not very
> large


Well. In
https://github.com/jenkinsci/apache-httpcomponents-client-4-api-plugin/blob/d43026ee871292547b224f11d6a2016303720165/src/main/java/io/jenkins/plugins/httpclient/RobustHTTPClient.java#L2-L4
for example the code does not consist of “changes” to the wrapped library,
it is fresh Jenkins-specific development; and while not large it is
certainly not trivial (well above the threshold considered significant for
copyright). I think it is appropriate for the *plugin* to remain under the
MIT license like most of the rest of Jenkins code. Obviously the JAR it
bundles is under a different license, but so what? The license refers to
the source files in the repository.

As to plugins which bundle a third-party library and have absolutely
nothing in `src/` other than `src/main/resources/index.jelly`, I guess it
is less confusing to just mark the plugin with the same license, though I
am not sure it makes any practical difference since the only conceivably
significant “source” would be in `pom.xml` and this is most often pure
boilerplate plus predictable ``. (Or `cd.yaml`, which is
almost always just a copy from template.)

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0LfBgYE%2BzHNmMinLjEWN9mcOjiZP4kHUXV4VSQ_ca08g%40mail.gmail.com.

Re: bitbucket-plugin - why didn't my PR got released

['Jesse Glick' via Jenkins Developers]

On Sun, Dec 10, 2023 at 12:09 PM Mark Waite 
wrote:

> Apply the label to the existing pull request and then run the cd action
> from the "Actions" menu of the repository.
>
> That's worked for me in the past.
>

Please update
https://www.jenkins.io/doc/developer/publishing/releasing-cd/#releasing if
anything seems to be missing or unclear.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0G-%2BZOtN_1eKBhsNRxKwJZwnE0oEV-QNQbRd3pmy_C_A%40mail.gmail.com.

Re: GetStatic Jelly Tag

['Jesse Glick' via Jenkins Developers]

On Thu, Dec 7, 2023 at 11:44 AM Bryan Stopp  wrote:

> I'm making the assumption here that the classloader functions the same
> when running stand-alone vs mvn hpi:run
>

Probably an invalid assumption.

I would advise never using the
https://commons.apache.org/proper/commons-jelly/tags.html#core:getStatic or
https://commons.apache.org/proper/commons-jelly/tags.html#core:invokeStatic
tags at all. Do as much as you possibly can in a Java class associated with
the view, and have the Jelly XML limit itself to calling methods on that
from JEXL, like ${it.someProperty} or ${descriptor.someMethod(someObject)}.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0_kN%3DaTgRVBgTprMG_tBB9x38j60%2Bg_N_d_4JJJrnAWQ%40mail.gmail.com.

Re: repeatable with selector and items to be hidden or visible on selection

['Jesse Glick' via Jenkins Developers]

On Wed, Nov 29, 2023 at 7:32 AM Cinemo IT  wrote:

> Anyway, basically i want to have a repeatable with add button that has a
> drop down menu for example with set of selections, once a selection is made
> a few other fields should either show up or get invisible, and on save and
> reload for config page it should have the items I've selected earlier to be
> preloaded.
>

Nothing of this kind is supported by the Jenkins configuration form
framework. Try a simpler and more standard UX.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2g00o2cXNPg21wAb%2Bj08LRRCraLfnrK7ww0me7j4L%3D2A%40mail.gmail.com.

Re: How can I test my plugin in pipeline?

['Jesse Glick' via Jenkins Developers]

In other words, write some basic tests using JenkinsRule first. If you then
want to test interactively to check UI and so on, use

mvn hpi:run

Since the relevant Pipeline plugins (minimally `workflow-job` +
`workflow-cps`, usually also `workflow-basic-steps` +
`workflow-durable-task-step`, occasionally others) will already be present
as `test`-scoped dependencies, appropriate compatible versions of these
will already be loaded in the local test server.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr04SKY6_AqWZSv2ogMY4uJ%2BU7B8P%3Dd6WV_K8xFCffe46A%40mail.gmail.com.

Re: How can I test my plugin in pipeline?

['Jesse Glick' via Jenkins Developers]

On Thu, Nov 16, 2023 at 9:58 AM tzach solomon 
wrote:

> I want to test [my new plugin] in pipeline
>

https://github.com/jenkinsci/archetypes/blob/282d79c14f8453bdffe1c74d5010dfdcf776ed9f/hello-world/src/main/resources/archetype-resources/src/test/java/HelloWorldBuilderTest.java#L64-L74

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1ixSdEB8fZS7W0cbYutvDs%3DD%3DPxGoiUmAb5jnznSn-JA%40mail.gmail.com.

Re: Config list of simple strings?

['Jesse Glick' via Jenkins Developers]

On Tue, Oct 24, 2023 at 5:55 PM Bryan Stopp  wrote:

> I can't seem to find any examples of repeatableHeteroProperty that don't
> use a some custom class, where as i just want to store a String.
>

Because it is not supported: https://issues.jenkins.io/browse/JENKINS-27901

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2-O8%2BkcDX-kJGA_BHE7YuvT_i610O52ybGZ--mZr%2BA2w%40mail.gmail.com.

Re: Backporting for LTS 2.414.3 has started

['Jesse Glick' via Jenkins Developers]

On Fri, Oct 13, 2023 at 11:42 AM 'Alexander Brandes' via Jenkins Developers
 wrote:

> a remoting PR has been labeled as “backporting-candidate”
>

Reminder (since I am not sure if this is mentioned in the backporting
guide): if and when this is selected for backporting, the technique is to
create a branch in `remoting` from the release included in the core LTS
branch, using for example
https://gist.github.com/jglick/86a30894446ed38f918050c1180483e2, and you
can then trigger the GitHub Actions workflow on the branch to produce a
release you can propose to include in core POM. Analogous example from
Stapler:

https://github.com/jenkinsci/stapler/compare/1711.v5b_1b_03f0fcf2...1711.x
https://github.com/jenkinsci/stapler/releases/tag/1711.1713.vc400cfb_5597a_
https://github.com/jenkinsci/jenkins/pull/7034

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1kWtewW_QXX0juBuOOkosNVn_hOSFF9SG8QZGs%2B7DZ_g%40mail.gmail.com.

Re: Splitting a plugin into a legacy and supported part

['Jesse Glick' via Jenkins Developers]

On Fri, Oct 13, 2023 at 2:38 AM Ullrich Hafner 
wrote:

> there might be still some users that use the old and deprecated steps
>

Like, Pipeline steps? So you cannot confidently delete this code without
potentially breaking user workflows?


> Move the old code to a new plugin with new plugin id. Make that new plugin
> a dependency of the code-coverage-api plugin.
>

I think this is backwards—then you are permanently locking the deprecated
code into the dependency chain.

`workflow-cps-global-lib-plugin` originally contained an in-Jenkins Git
server that you could push Pipeline Groovy libraries to. Later we added an
option to pull libraries from the SCM of your choice, and deprecated the
original code, but it stuck around and was a maintenance burden. So finally
we introduced `pipeline-groovy-lib-plugin` containing just the newer code
(in the same package structure and everything), made
`workflow-cps-global-lib` depend on `pipeline-groovy-lib`, and deprecated
the whole `workflow-cps-global-lib` plugin. Thus existing servers with
`workflow-cps-global-lib` installed would upgrade plugins, which would
automatically install `pipeline-groovy-lib` as a dependency, but see no
behavioral change. Admins could then at their leisure uninstall
`workflow-cps-global-lib` assuming they were not in fact using the old Git
server. New servers need never install `workflow-cps-global-lib` to begin
with (no longer a dep of `workflow-aggregator` or otherwise offered in the
setup wizard, etc.).

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1%2B2s9Ry%2BTU8y8h5ANST%2BZLyA9SpZuGNwJAt5D8PDdHFA%40mail.gmail.com.

Re: Backporting for LTS 2.414.3 has started

['Jesse Glick' via Jenkins Developers]

On Mon, Oct 2, 2023 at 1:49 AM Kris Stern  wrote:

> Fixed:
> https://issues.jenkins.io/browse/JENKINS-71479?jql=labels%20%3D%202.414.3-fixed
>

I think you meant to link to
https://issues.jenkins.io/issues/?jql=labels%20%3D%202.414.3-fixed ?

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1Jdikfn8nTNjYaVu-zih0m9Nzv3JTLiuei-WFy_AcssA%40mail.gmail.com.

Re: GitHub Packages access

['Jesse Glick' via Jenkins Developers]

On Thursday, September 28, 2023 at 9:43:59 AM UTC-4 Vladimir Belousov wrote:

We also thought about publishing *exhort-java-api* to Jenkins Artifactory.


Best to discuss in a new 
ticket: https://github.com/jenkins-infra/helpdesk/issues/new/choose 

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/9501a044-e718-4198-adeb-edbbf8eb12c4n%40googlegroups.com.

Re: GitHub Packages access

['Jesse Glick' via Jenkins Developers]

On Thu, Sep 28, 2023 at 4:37 AM Vladimir Belousov 
wrote:

> We use dependencies that are hosted on GitHub Packages in our plugin.


I guess you mean
https://github.com/jenkinsci/redhat-dependency-analytics-plugin/blob/f4b606b8b509795917edc2f2915c6a3322a85e4d/pom.xml#L212-L215
to access https://github.com/RHEcosystemAppEng/exhort-java-api

This is not standard practice and is likely to cause issues. Normally any
dependencies you need should be published either to Jenkins Artifactory, if
they are specific to Jenkins, or Maven Central if not.

I am well aware that
https://github.com/RHEcosystemAppEng/exhort-java-api/blob/ed0cb76f5ccd1d0d74bdbc6d36a4c04b2900d51c/.github/workflows/release.yml#L56-L61
is vastly simpler to manage than deploying to OSSRH. At some point
https://sigstore.github.io/sigstore-maven-plugin/ should make it possible
to deploy to Central using GHA OIDC tokens, but it is not ready yet and
AFAIK there is no published timeline.

If you really want to access GH Packages, you can probably do so with
`GITHUB_TOKEN` in GHA without needing a PAT. This would work for the CD
action, probably with custom modifications, but would not work for
ci.jenkins.io so Jenkinsfile would be useless; you would need to set up
your own CI.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3cZ-hh_%3D7zJ8t91kdf3aEi08c%2Br1-A9eQrgo9NaLa_yg%40mail.gmail.com.

Re: [Information] Release block "Beta" program

['Jesse Glick' via Jenkins Developers]

On Thu, Sep 28, 2023 at 4:11 AM 'wfoll...@cloudbees.com' via Jenkins
Developers  wrote:

> I would like to get more opinions on the topic before investing time on it
> as the current version seems to be satifsactory for the beta testers.
>

FWIW the current system has been very reassuring for me. There is a
constant flow of PRs on various plugins I have write access to which either
fix some minor bug or unblock PCT etc., but even in cases where I have
personally seen a SECURITY ticket a couple of weeks ago it is very hard to
keep up with the current list of plugins with pending advisories. Seeing
that I suddenly lack write permission is enough to remind me of the hazard,
and prevent me from accidentally merging a PR, triggering a release, etc.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1OYqZ5MQ5RXvqudvPx-mF_duhJ9Xxke0G%3DfztKnEQNqA%40mail.gmail.com.

Re: remove j-interop and jcifs from core?

['Jesse Glick' via Jenkins Developers]

On Mon, Sep 18, 2023 at 7:22 AM 'jn...@cloudbees.com' via Jenkins
Developers  wrote:

> The ability to install Jenkins as a service[1] when it is running (by java
> -jar) on a windows machine
>
> I forgot to mention - which potentially valid - I would argue that the
> user should just download the MSI package for this instead, afterall we do
> not attempt to install a systemd file when running on Linux.
>

We *did* used to do just that (and analogously on various platforms), but
only when running from Java Web Start. If this dependency was only needed
for windows-slave-installer-module then it is obsolete as of
https://github.com/jenkinsci/jenkins/pull/6543

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1Rvq%2BQoYpbzrHYfqvMy6nOVjpB4SswqA8D-6bSrWyong%40mail.gmail.com.

Re: bitbucket plugin - JENKINS-65697 - How to inject environment variable into WorkflowMultiBranchProject (buildEnvironment isn't called...)

['Jesse Glick' via Jenkins Developers]

On Tue, Jul 25, 2023 at 2:24 AM tzach solomon 
wrote:

> get build from the Queue Item
>

I am not sure what that means. You are going to need to be more specific,
preferably with code references, ideally a JenkinsRule-based test failing
in the “right” way.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1%2B6wVKuRED8CmKXwp6aoj-UZnwiyohoSUce%3Dcr0ymJMg%40mail.gmail.com.

Re: Can't see plugin name in build steps plugin list.

['Jesse Glick' via Jenkins Developers]

Maybe you are trying to run the plugin in a nonstandard way that is
skipping the required annotation processor. From a shell:

mvn clean hpi:run

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2bFc4-itfnQoMqPE-5_GxH5bA71sdzXSCR3zpGgQk1PA%40mail.gmail.com.

Re: bitbucket plugin - JENKINS-65697 - How to inject environment variable into WorkflowMultiBranchProject (buildEnvironment isn't called...)

['Jesse Glick' via Jenkins Developers]

On Mon, Jul 24, 2023 at 1:58 AM tzach solomon 
wrote:

> I've checked the code and it looks like the triggered job is of class
> *WorkflowMultiBranchProject*. The future job I'm getting once scheduled
> OK is *jenkins.branch.MultiBranchProject$BranchIndexing*.
>
> I've tried to inject environment variables into it by using the already
> existing class *BitBucketPayload* which looks like this *BitBucketPayload
> extends InvisibleAction implements EnvironmentContributingAction*
>
> My problem is that the *buildEnvironment* method isn't called.
>

Nor should it be. That is an extension point for builds. You are discussing
branch indexing. Maybe you are looking in the wrong place.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1d4PqcO4LGmFrOhw%2B73DypfCvJE65QeMt9OMr4rKQ2ow%40mail.gmail.com.

Re: How to get listener from the Run object

['Jesse Glick' via Jenkins Developers]

On Wed, Jul 19, 2023 at 10:25 AM Michael Carter 
wrote:

> I'm in an area where all I've got is the Run object.
>
> Where do I need to go to get the TaskListener so I can write to the logs?
>

Well, for Pipeline you could use
https://javadoc.jenkins.io/plugin/workflow-api/org/jenkinsci/plugins/workflow/flow/FlowExecutionOwner.html#getListener()

For other project types, there is no supported mechanism for writing to the
build log from outside. I guess you could do some tricks with RunListener
and a WeakHashMap.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2k76CzwB78PrbK_-ZZa69-1vCvo4CPVCeC45RaYnXAKA%40mail.gmail.com.

Re: GitSCM is marked as Serializable but can't be serialized because of GitSCM.extensions

['Jesse Glick' via Jenkins Developers]

I guess the question is prompted by
https://github.com/jenkinsci/git-push-plugin/blob/5c9525da6da7bed8fd70e8a45d62eafd1daad3f1/src/main/java/io/jenkins/plugins/git_push/GitPushStep.java#L109
which is odd because while `StepExecution` in general is `Serializable`, a
`SynchronousNonBlockingStepExecution` does not need to be meaningfully
serializable since it will just be aborted across a controller restart. In
other words the `readObject` and `writeObject` methods are useless and
could be deleted.

`GitPushStep` as a whole seems dubious anyway, since you can just write e.g.

withGit(…) {
  sh 'git push …'
}

(More realistically, the `git push` would be embedded in some larger
script.)

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1uVX_uL6aJZ-jPwFAPbW89dWV2_Yaw5Wo%3DCr%3DFMeH-rw%40mail.gmail.com.

Re: Adopting the github-oauth-plugin plugin

['Jesse Glick' via Jenkins Developers]

On Thu, Jul 13, 2023 at 9:33 AM Kris Stern  wrote:

> In case there is anything else I have missed
>

https://groups.google.com/g/jenkinsci-dev/c/iZz5-GqKw-8/m/RKQr3qXTBQAJ :-)

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr09rKgjxB_WfOKumJOit2%2BB5Xrts7BWeXk9JaO6XoYkJg%40mail.gmail.com.

Re: TaskListenerDecorator/Factory on agents

['Jesse Glick' via Jenkins Developers]

On Fri, Jun 23, 2023 at 11:45 AM Alexis Tual  wrote:

> we're trying to parse any pipeline logs with TaskListenerDecorator to
> detect some pattern and report with a custom action attached to the run.
> It's working fine on agents by default but not when
>
> org.jenkinsci.plugins.workflow.steps.durable_task.DurableTaskStep.USE_WATCHING
>  is
> true. In that case it only works on the builtin node.
> The issue is that the TaskListenerDecorator eventually needs the Run to
> set the action on, and this run is null when executed on agent. How can I
> pass this Run instance so it's available on agents ?
>

You cannot; that is a model object which can only be accessed from inside
the controller.

What you are asking is tricky. The agent must tell the controller to do
something, but normally the controller will not trust anything the agent
says, so the communication has to be designed with care.
https://github.com/jenkinsci/workflow-basic-steps-plugin/blob/6e8791a4448c099cdf6e35bb0ec2601fb88a953a/src/main/java/org/jenkinsci/plugins/workflow/steps/TimeoutStepExecution.java#L298-L311
does something similar; here the `id` is basically an opaque token
generated on the controller and given to the agent, and the only thing the
agent can do is inform the controller that log output from this step in
this build is still being produced. In your case you would need some
similar token but also send back ome information (about a pattern match?)
which should be considered untrusted and sanitized accordingly.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2avo44QofyVg%2Btm0dwrrowXPRRs4HN03J%3DS63VdbPdqQ%40mail.gmail.com.

Re: Jenkins 2.387.3 LTS RC testing started

['Jesse Glick' via Jenkins Developers]

I discovered what seems to be a regression in the RC (and weeklies):
https://issues.jenkins.io/browse/JENKINS-71139

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr38itNNjkenYs%3DFJ%3DBbb3ri9%2BDnROGEqLj_m3bB9fry%2Bg%40mail.gmail.com.

Re: Help needed with simple AutoComplete -- Here's my code, what's wrong?

['Jesse Glick' via Jenkins Developers]

On Wed, Apr 19, 2023 at 6:32 PM Tamás Mucs  wrote:

> jellys of ParameterDefinitions don't normally have forms around their
> controls as I see.
>

Sure they do—the configuration form for a project (named `config`).

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2Z%2BjbtLCsuQc-USZfj%3Dk%3DvdKSR-5KxQGD1mfEmbCrqAQ%40mail.gmail.com.

Re: Help needed with simple AutoComplete -- Here's my code, what's wrong?

['Jesse Glick' via Jenkins Developers]

On Wed, Apr 19, 2023 at 7:29 AM Tamás Mucs  wrote:

> is there something off with my code
>

It is pretty odd. You are using form controls not inside a form, to start
with.

Perhaps you meant to use `GlobalConfiguration`? (There is a Maven archetype
to help.)

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2JnHTZbxNi7CtzexbxnLKinEVdJ-Erj_s0wGifKy93Xw%40mail.gmail.com.

Re: How to retrieve sensitive data in Jenkins agent node

['Jesse Glick' via Jenkins Developers]

On Thu, Apr 13, 2023 at 12:55 PM d...@kinneartech.com 
wrote:

> How can we pass those sensitive args to Jenkins agent machine without
> getting disclosed. We already tried adding them to envVars but there seems
> to be an issue that the envVars are disclosed in a parallel job in a
> different workspace when using some commands to print them.
>

This is pretty vague. Disclosed how, to whom? Which commands?

https://javadoc.jenkins.io/plugin/credentials-binding/org/jenkinsci/plugins/credentialsbinding/masking/SecretPatterns.html
can be used to mask secrets which may be bound in environment variables in
scope (for example via
https://javadoc.jenkins.io/plugin/workflow-step-api/org/jenkinsci/plugins/workflow/steps/EnvironmentExpander.html)
though it is better to have users run the standard `withCredentials` step
instead.

Anyway

a parallel job in a different workspace
>

would one way or another likely have access to any secrets actively in use
by an executor on the same agent machine used by another build: it is not
straightforward to create a strict separation between two processes running
on the same machine, under the same user account. (Any secrets written to a
temporary file would trivially be visible to any other builds, secrets
bound to environment variables in other processes can be seen in
`/proc/*/environ` on Linux, and even secrets solely present inside a JVM
can by default be accessed using the Java Attach API.) If there is any
possibility that different projects might be owned by different teams who
do not trust one another, then you must reserve an agent for the exclusive
use of a single build, preferably using “one-shot” provisioning in
containers or VMs.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1%3D%3Drno-OoACrA6xcqhYSr2MqTpg%2BaJsAUcSG48kSY3kA%40mail.gmail.com.

Re: [Regression] Non-unique IDs

['Jesse Glick' via Jenkins Developers]

https://github.com/jenkinsci/jenkins/pull/7823 is a straightforward fix
addressing at least one reported case of regression, with no need to revert
anything or modify plugins. There could of course be unrelated issues.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0BS5YzUoMOXP4jz0EkyG_duC4bgZpwYC0OTMArN8w1Bg%40mail.gmail.com.

Re: script to configure repository labels

['Jesse Glick' via Jenkins Developers]

On Fri, Mar 31, 2023 at 10:01 AM Tim Jacomb  wrote:

> It can be setup at the org level as a repository default, Gavin also had
> an app where you could click a button and it would set it all up for your
> repository
>

If there is something that works to at least copy labels from
`jenkinsci/.github` on a one-off basis, that would be very helpful and
ought to be documented at
https://www.jenkins.io/doc/developer/publishing/releasing-cd/ (or perhaps
elsewhere also, since it is useful even for MRP-based plugins).

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1%3D66gWBrPQB_uFj8%2BnM%2BJg5Zpm7znt0tTYNN%3DKhCu1og%40mail.gmail.com.

Re: Form validation in Action? doCheck of Descriptor is not called before form submission.

['Jesse Glick' via Jenkins Developers]

On Thu, Mar 23, 2023 at 12:25 PM Tamás Mucs  wrote:

> doSubmit() is called before any validation would happen. I tried to add
> the following method to my *descriptor *but it is *not *called
>

Form validation is advisory; it can display a warning message but does not
block save. https://issues.jenkins.io/browse/JENKINS-19584

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3OSX_30wAW%2Bhwi-FB4EL9E8uEcQ1ZQ9vVbqZzH_06irw%40mail.gmail.com.

Re: one-shot-executor-plugin

['Jesse Glick' via Jenkins Developers]

That experiment was indeed abandoned. Use
https://javadoc.jenkins.io/plugin/durable-task/org/jenkinsci/plugins/durabletask/executors/OnceRetentionStrategy.html

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3_c7pc6FNLyZ0xzRDObCQsn%2BULSemmZMJtWeKsOV77zg%40mail.gmail.com.

Re: *-api-plugin (tests downstream plugins)?

['Jesse Glick' via Jenkins Developers]

On Sun, Feb 19, 2023 at 4:50 PM Olivier Lamy  wrote:

> I'm not quite sure how to implement this technically because
> downstream plugins may use the api-plugin as direct dependency with
> version or using the bom so this means we would have to inject an
> incremental bom or change version.
>

https://issues.jenkins.io/browse/JENKINS-45047 gives the basic idea. The
basic machinery is already used from `jenkinsci/bom`. See the second bullet
point in
https://github.com/jenkinsci/maven-hpi-plugin/pull/66#issuecomment-1110203758
for discussion.

how to detect all downstream plugins?


As Mark mentions, that would rarely be practical. We could test some
*representative* downstream plugins.

A different approach would be to make the UC generator continue to pluck
the latest version from Artifactory for “long-tail” plugins, but for
plugins managed in `bom`, pick up their versions from the last stable
`master` build or whatever. That would ensure that regressions among a
large set of commonplace plugins do not accidentally slip onto the UC
before they are corrected. This would not have helped in this case, though,
since `kubernetes` has a very specialized test suite that is not covered by
`bom`. Similarly for `artifact-manager-s3` vs. AWS SDK, etc.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1sk28k%2B-gVMuiHHAp2EuynnEXw8cP%3DMVFpt1D-PAD8ZQ%40mail.gmail.com.

Re: How does /monitoring skip starting splash page?

['Jesse Glick' via Jenkins Developers]

On Wed, Feb 8, 2023 at 8:55 AM 'Daniel Beck' via Jenkins Developers <
jenkinsci-dev@googlegroups.com> wrote:

> PluginImpl#start adds…
>

Yes, as I wrote, but please bear in mind that `Plugin.start` is
deprecated—use `@Initializer` instead.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1qHoGpipb_i3d%3DYC1uK8qL2r5e4BdMugNC6O8NLcOEHw%40mail.gmail.com.

Re: How does /monitoring skip starting splash page?

['Jesse Glick' via Jenkins Developers]

It may be possible by calling `PluginServletFilter` in an `@Initializer`,
as filters will be consulted before the request is even routed to Stapler
(where the `HudsonIsLoading` object is installed until startup completes
and it is replaced with the `Jenkins` singleton). Remember that, among
other things, you are bypassing authentication checks (requests may be
anonymous), so this is only appropriate for carefully vetted code and
non-sensitive data.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3aQ9uSTO-AH%2Bn6E2Lb%3DP7rALL8pL9E9_D%3DA_V4d8Yn2g%40mail.gmail.com.

Re: Plugin POM inheritance

['Jesse Glick' via Jenkins Developers]

On Thu, Jan 19, 2023 at 6:19 PM Basil Crow  wrote:

> I think such a change, while not trivial, is likely feasible.
>

+1, thanks for reviewing & classifying the bits we would not want to
inherit (or which are obsolete).

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1jQ%2BU8NCzY%3DbmmAwiyMoWESYzK7z2i8YxBsE84rVzmXA%40mail.gmail.com.

Re: Towards new kubernetes-client 6.x

['Jesse Glick' via Jenkins Developers]

On Thu, Jan 5, 2023 at 2:35 PM Vincent Latombe 
wrote:

> the package on fabric8 side is not updated to a new one, so both plugins
> wouldn't be able to co-exist in the classpath.
>

You can simultaneously run multiple wrapper plugins with the same package,
so long as any given caller’s transitive dependency closure only uses one
of them. I think that means there would have to be a
`kubernetes-credentials-6` plugin, since its `kubernetes-client-api-6`
dependency makes it implicitly incompatible. Seems like a mess.

Setting that aside, it would be awkward to create a new
`kubernetes-client-api-6` plugin because then people would be left with the
old `kubernetes-client-api` installed and enabled but unused, and need to
manually clean up.

Pending JENKINS-49651, I agree that just releasing the batch of updates on
the same day and asking users to please refrain from partial updates is the
least disruptive option.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr23xAA7OuPqi_V1-FAGz1XE%3DudW7LXSpCFo%3D45BNpvpKw%40mail.gmail.com.

Re: Migration to Java11 and modules

['Jesse Glick' via Jenkins Developers]

On Tue, Jan 3, 2023 at 8:08 AM Ullrich Hafner 
wrote:

> Do we already have some guidelines on how to migrate a plugin to Java 11?


Yes; normally there is very little to do.


> Do we need to create module-info.java files for plugins?


No.


> How do we access libraries that use modules?
>

Jenkins does not use the Java module system, so there should be no change
from Java 8 usage modes. If there is some incompatibility in Java 11 in
this regard, then you may be the first to encounter it so you will need to
debug accordingly. Is this failure during the build itself, or in Surefire
test runs? Without any details it is hard to even guess what is wrong.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr15XHaVQ%3Dz0wVe_n%2B-bOZcA3rZ9d7muGiy3mZ%3D7JZTq_g%40mail.gmail.com.

Re: User.CanonicalIdResolver extension point "context"

['Jesse Glick' via Jenkins Developers]

On Fri, Dec 30, 2022 at 8:17 AM Cristian Morales Vega 
wrote:

> I am looking at https://issues.jenkins.io/browse/JENKINS-70342, but
> I'm not sure how the "context" in CanonicalIdResolver is supposed to
> be used. The only thing I have seen in the docs is "Can be used (for
> example) to distinguish ambiguous committer ID using the SCM URL".
>
> DefaultUserCanonicalIdResolver, User.FullNameIdResolver,
> User.UserIDCanonicalIdResolver ignore it; and I have not seen it used
> anywhere else, neither giving nor receiving a context. But I'm new
> here, there is no easy way to see all the implementations of
> CanonicalIdResolver in all the plugins, there is?
>

https://github.com/jenkinsci/additional-identities-plugin/blob/f98f43fc8a2b1831779e48cb5355bf1295ce347f/src/main/java/com/cloudbees/jenkins/plugins/AdditionalIdentityResolver.java#L51-L62
seems to look for it. But it could only come from
https://github.com/jenkinsci/jenkins/blob/9d961dcce5f9e6094d2b9bf38ae4a112a27df4d8/core/src/main/java/hudson/model/User.java#L509-L510
and I do not recall any code ever passing this argument to `User.get`. I
suspect this aspect of the API was never really developed.


> The use of "idOrFullName" in resolveCanonicalId() also feels a bit
> overspecific? While you could call an email an "id", you could argue
> the same about a full name.
>

*Full name* would be something like “First M. Last”. It is possible to pass
this to `User.get` to look up a user by either id or full name. In practice
we try to always use `getById`. The id might be something like `flast` or
like `fl...@github.com` or something else, depending on the `SecurityRealm`.

The entire system of associating “ids”, full names, and email addresses
with both `SecurityRealm` authentications and SCM commit “authors” in
Jenkins is a complicated mess I am afraid.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3NXt_BjJaumyhoxcW%3DKtor_pT0vODb1f2ZwWOGNnPmyQ%40mail.gmail.com.

Re: Renaming a plugin

['Jesse Glick' via Jenkins Developers]

On Sun, Dec 18, 2022 at 7:47 AM Alexander Brandes 
wrote:

> take a look at the RPU documentation
> 
>  how
> to do that.
>

I think this doc should be made to emphasize that you *really really*
should not ever do this unless you are pretty sure your plugin has no users
under its existing name.

Changing the `name` (display name) is fine at any time. Changing the
`groupId` will break any Maven dependencies from other plugins, which may
or may not be an issue. But changing the `artifactId` means changing the
identifier by which the Jenkins plugin manager differentiates this plugin
from others, and will cause mayhem for users who have already installed it
under the old name.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2eiZ%3DsKfcBBzdjpgmhQ3LMw0_N6o4y1Xum39J%3DCe890w%40mail.gmail.com.

Re: Solving “failing Parent Pom upgrade from 4.51 to 4.52 build” cheatsheet

['Jesse Glick' via Jenkins Developers]

On Wed, Dec 14, 2022 at 4:10 PM Ullrich Hafner 
wrote:

> increment the MAJOR version when you make incompatible API changes
>

It is a nice theory, but in practice nearly any change *could* be
incompatible when the “API” is defined with less than mathematical
precision. 4.52 obviously *was* incompatible for many repositories, but
4.49 also had an incompatible change affecting some repositories, and 4.50
had three. Best to just treat the version as opaque, and if your Dependabot
CI build breaks (or even if it does not), read the release notes.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2A_Mk3MzXmPft5dKGrCNY-0RLJ3L02q2KX%2BMqBgQkhXg%40mail.gmail.com.

Re: Next baseline shift

['Jesse Glick' via Jenkins Developers]

On Wed, Nov 30, 2022 at 2:41 PM James Nord  wrote:

> this would affect plugins that have less active development more, which
> means users are probably less familiar with our tooling to begin with
>

Right. Maybe safer to leave the default as is, but formally deprecate
running `buildPlugin` without `configurations`, and then we have a tool we
could use to propose  `Jenkinsfile` changes *if* we first fix
https://issues.jenkins.io/browse/JENKINS-46795 and push that to
ci.jenkins.io.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2U0kPHgD8wvjPS06ZzyJc2z_Nv7vjHgbBCpyvuOTS8Xg%40mail.gmail.com.

Re: Servlet container support (was: Nov 28, 2022 Governance Board Agenda)

['Jesse Glick' via Jenkins Developers]

On Tue, Nov 29, 2022 at 6:29 PM James Nord  wrote:

> if you can not use tomcat or $container then Jenkins must not be a war,
> which has a large overlap but is subtly different
>

Interesting…so if the basic Jenkins download were to physically switch
format somehow, then it would no longer be subject to the Tomcat rule? Or
would some practical benefit be lost by not using Tomcat?

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1euH3OE1oaFxtnB30FAVUtuU-ng5hWzdSAzvWYHrTAQA%40mail.gmail.com.

Re: Next baseline shift

['Jesse Glick' via Jenkins Developers]

On Tue, Nov 29, 2022 at 6:19 PM James Nord  wrote:

> This bytecode difference trips up spotbugs (causing several new false
> positives)
>

Mm, OK, that sounds like a real risk that would preclude changing the
default any time soon.

Ideally a plugin repository would pin a version of the library, and
Dependabot or updatecli could offer updates. Historically we have not done
that. I am not even sure it would be practical, since `buildPlugin.groovy`
contains a mixture of actual build instructions that ought to be
synchronized with source code changes in general (such as the choice of
JDK), as well as interactions with ci.jenkins.io infrastructure that are
better synchronized with the site configuration.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr368GK1QrdyQKrBm%3DsSYuj6wArcVimYOoH2j%3DtaUZjioQ%40mail.gmail.com.

Re: Next baseline shift

['Jesse Glick' via Jenkins Developers]

On Mon, Nov 28, 2022 at 2:53 PM Basil Crow  wrote:

> > Independently of that, we can consider switching the default JDK for
> ci.jenkins.io `buildPlugin` from 8 to 11.
>
> I am not sure to whom you are referring when you write "we". Is this
> something you are planning to implement, something you are asking me to
> implement, or something that you are looking for a third party to
> implement?
>

Not really any of those—it is basically a one-line change, if “we” want to
do it, so I was soliciting opinions on whether it sounded desirable and
safe enough to try (as opposed to leaving the default as 8 indefinitely).

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr17AhVJ0R2JawKnsmuGd5u2j3FXSjZWTzpoR1MGneNX%2Bg%40mail.gmail.com.

Re: Next baseline shift

['Jesse Glick' via Jenkins Developers]

Your proposal sounds great to me, and thanks for volunteering to implement
the changes. I think a late-2022 timeline (sometime after 2.375.1 has been
released) makes sense—if there is some advantage in waiting longer, I do
not get what. The only likely drawback for plugin maintainers who
deliberately wish to stay on a pre-2.361.x baseline is that Dependabot (if
configured) will offer `plugin-pom` updates that fail CI builds, which is
not so bad since the release notes will explain why, and the PR and its
replacements can simply be ignored indefinitely. (And of course you will
not get access to new stuff in the toolchain, but that seems fair.)

Independently of that, we can consider switching the default JDK for
ci.jenkins.io `buildPlugin` from 8 to 11. Even if your plugin supports
older baselines and Java 8, running the build on JDK 11 is typically fine,
since `-release 8` will ensure the right bytecode and even prevent the
infamous `ByteBuffer` problem. The likely issues: you will lose Java 8 test
coverage (while gaining 11 test coverage, which is more important!); and
occasionally some tool like `javadoc` will be stricter (in a way you ought
to fix anyway though you might have preferred to take your time).

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1mdb5zZx7%3D%2B2M%2BgtUrCMXtfCQcJ406WhvQojhx5NTbXw%40mail.gmail.com.

Servlet container support (was: Nov 28, 2022 Governance Board Agenda)

['Jesse Glick' via Jenkins Developers]

On Mon, Nov 28, 2022 at 9:29 AM Mark Waite 
wrote:

> Mark Waite to propose a pull request documenting the web application
> server support policy […]
>
>-
>   -
>  -
>
>  Many people are saying that “should work” but we don’t test them
>  […]
> -
>
> All web containers moved to tier 2 except bundled web
> container
>
>
Not sure if there was a thread somewhere in which this topic was discussed
in writing, but my personal recommendation would be to formally drop
support for running in a general servlet container, so that you could only
use the bundled Winstone/Jetty via `java -jar jenkins.war` (we could even
drop aspects of the WAR packaging not needed by `executable.Main`).

The obvious thing that does not work in other containers is WebSocket
support:
https://github.com/jenkinsci/jep/blob/master/jep/222/README.adoc#non-jetty-containers

I do not agree that the situation is analogous to the web browser, OS, or
Java support policy. We do not bundle a pretested web browser, JVM, or
operating system with Jenkins (except insofar as the latter two are
supplied by the Docker image).

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr15_NQxx4k_dOawaPx%2B5MnhL4BaUJOPvB%2BptQ06NN%2BSwA%40mail.gmail.com.

Re: BOM compiles plugins with target 11

['Jesse Glick' via Jenkins Developers]

On Thu, Nov 10, 2022 at 2:03 AM Ullrich Hafner 
wrote:

> This causes a failure of:
>
> [ERROR] Failed to execute goal
> org.apache.maven.plugins:maven-surefire-plugin:3.0.0-M7:test (default-cli)
> on project checks-api: Execution default-cli of goal
> org.apache.maven.plugins:maven-surefire-plugin:3.0.0-M7:test failed:
> org.junit.platform.commons.JUnitException: TestEngine with ID 'archunit'
> failed to discover tests: ExceptionInInitializerError: class
> com.tngtech.archunit.lang.ArchRule$Factory$SimpleArchRule cannot be cast to
> class com.tngtech.archunit.lang.syntax.elements.ClassesShouldConjunction
> (com.tngtech.archunit.lang.ArchRule$Factory$SimpleArchRule and
> com.tngtech.archunit.lang.syntax.elements.ClassesShouldConjunction are in
> unnamed module of loader
> org.apache.maven.surefire.booter.IsolatedClassLoader @5cf39df6) -> [Help 1]
>

I just tried `mvn -f checks-api-plugin clean verify` locally using JDK 11
and it passed.

Is there a toggle to define in the BOM that the plugin build is not ready
> for JDK11?
>

No, I think all plugins are expected to be buildable on JDK 11 even if
their runtime baseline remains Java 8.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1mpAeos5ZC2%3D%3DiFpCXta6z-Dc9-PT7pY%3D50cb-wA8fAQ%40mail.gmail.com.

Re: Adoption request for junit-attachments-plugin

['Jesse Glick' via Jenkins Developers]

Also, is there a particular reason you are proposing to adopt the plugin?
https://github.com/jenkinsci/junit-attachments-plugin/pulls shows no open
PRs with user-visible changes. Adoption is generally used to move forward
development when there has been no maintainer able to review proposed
changes.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr30h-jibEqWAdj-fqJaUDqx_ULCv-VcKH1Lrzyj4fk2ew%40mail.gmail.com.

Re: Adoption request for junit-attachments-plugin

['Jesse Glick' via Jenkins Developers]

The links here do not match the text; please resend with corrections.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2qvrm4_-VUUT5GqDYSC1oOyn8pSp1ebqiUkeYOZQOp9w%40mail.gmail.com.

Re: Graceful job termination / cleanup before abort

['Jesse Glick' via Jenkins Developers]

While you could probably intercept this somehow on the Jenkins plugin side,
Jenkins should be sending SIGTERM (on Unix; not sure about Windows) so it
would seem most appropriate to have the native command handle that
gracefully. You would need to ask in another forum how best to do that in
C#.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1SuN_f5BouSuhznEy2nOSXM1DwtpwrkhLzF9ga%2BF1Neg%40mail.gmail.com.

Re: Folder Credential by credentialId not found using credentials API

['Jesse Glick' via Jenkins Developers]

I suspect you wanted
https://javadoc.jenkins.io/plugin/credentials/com/cloudbees/plugins/credentials/CredentialsProvider.html#findCredentialById(java.lang.String,java.lang.Class,hudson.model.Run,com.cloudbees.plugins.credentials.domains.DomainRequirement...)
which should take into account the folder context.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0CYLXLqWkztBdqJVWuQd%3D056cMT%3DbFchnvxGMGrVGWnQ%40mail.gmail.com.

Re: Help upgrading acegi-security to spring-security

['Jesse Glick' via Jenkins Developers]

At least start by reading
https://github.com/jenkinsci/reverse-proxy-auth-plugin/pull/40 and
https://github.com/jenkinsci/ldap-plugin/pull/49.

https://github.com/jenkinsci/reverse-proxy-auth-plugin/pull/37 would
perhaps help. The root problem is the lack of a general way in Jenkins to
compose security realms—often to get “fallback” functionality (e.g. to log
in while the primary system is down), but more pertinently here, to
delegate calculation of granted authorities (“groups”) to another system
such as LDAP while controlling the user identifier (login).
`reverse-proxy-auth` therefore includes its own LDAP call layer because it
cannot delegate that to the `ldap` plugin, so to migrate to Spring Security
you have a lot of code to rewrite.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2%2B-hEi_a0Ppbgty6WQgoRUPofFuR3ARqBQOyR9NcCwgw%40mail.gmail.com.

Re: XStream Serialization: SortedSet vs. NavigableSet vs. TreeSet

['Jesse Glick' via Jenkins Developers]

On Mon, Oct 17, 2022 at 5:43 AM Ullrich Hafner 
wrote:

> Shouldn’t XStream always pick the correct dynamic type (and not the static
> type) when serializing things?
>

As far as I know it does, but then again the representation for exotic
collection types is complex.

You should declare the persisted field to be of some simpler type,
preferably `ArrayList`. If you need to enforce uniqueness, sorting, etc. in
memory during operations, do so using any other mechanism, perhaps a
`transient` field.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2w0LfDEQV3uybALYmJG7houQumg3oNrNQAVekf3-RfSw%40mail.gmail.com.

Re: Trying to automate release of Plugin

['Jesse Glick' via Jenkins Developers]

On Thu, Sep 15, 2022 at 7:52 AM Karl Henselin  wrote:

> I didn't know that github had a jenkins action, which apparently it does,
> or something causes it to find a jenkinsfile and know that means to run
> some github actions.
>

No, these things are decoupled. ci.jenkins.io notices the `Jenkinsfile` and
runs it. Whenever that build completes (actually after every stage and so
on), ci.jenkins.io publishes a check to GitHub, and if on the master branch
that event triggers the GitHub Action to look whether there is a successful
Jenkins build and if so whether the running changelog merits a release.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1JJh8nm_sHAEqXaK_Qv1484kescp%2Bu4rN0xzS9xPed4A%40mail.gmail.com.

Re: Trying to automate release of Plugin

['Jesse Glick' via Jenkins Developers]

On Fri, Sep 9, 2022 at 1:13 PM Karl Henselin  wrote:

> I followed the CD automation guide for Jenkins plugins, and I think that I
> was successful with all the steps.
> However, when I make a pull request, it never releases a new version.
> I tried running the CD job with debug logging turned on, but I can't
> understand the logs to know what might be wrong.
>

`verify-ci-status` fails, because there is no CI status—you have no
`Jenkinsfile`.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0dJb684HVsg1ak%3DAmEkpQxnbvs%2B%3DTtkuqdqWapz1Jxxw%40mail.gmail.com.

Re: Add a GitHub App to allow marking pull request dependant on another

['Jesse Glick' via Jenkins Developers]

On Wed, Sep 7, 2022 at 6:42 AM 'Herve Le Meur' via Jenkins Developers <
jenkinsci-dev@googlegroups.com> wrote:

> I think it could be useful on the jenkinsci and jenkins-infra GitHub
> organizations, WDYT?
>

Sure, assuming it passes some sort of security review. As someone who
frequently creates sets of interrelated PRs in @jenkinsci, I would at least
try it. (Really I think it should be something built into GitHub with
first-class presentation.)

Note that its model is a bit simplistic—as soon as the upstream PR is
merged, the downstream is unblocked. That suffices for PRs within a
repository, but when using Maven dependencies across repositories what we
actually want is for the upstream PR to be *released* and for the
downstream PR to encode that release version in a dependency. Typically I
have encoded this by leaving the downstream PRs in draft status and keeping
a TODO comment visible in the diff noting the upstream PR, which works but
is clumsy.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3-6tup%2BzTtO-5w4a6AL71ZSt%2BrOy5KqdB9%3DMR%2B9ZqYpA%40mail.gmail.com.

Re: Merging multiple build steps into one (with nested Describables)

['Jesse Glick' via Jenkins Developers]

On Sun, Aug 14, 2022 at 12:14 PM Tim Van Holder 
wrote:

> there seems to be no checking that such a symbol is valid/usable, neither
> at build time nor at run time. […] It would certainly be useful to get a
> diagnostic for that at maintenance/build time

Might be possible at runtime.

I'm using  for the nested describables which
> mostly works nicely. However, this does not surface any help files for the
> selected describable in the UI. Is that a bug

Try updating to latest core.

So my only option would seem to be to essentially create a new set of Steps
> just so they could have descriptors marking themselves as advanced
>
Yes.

that would involve the symbols being associated with those new classes;
> would that break anything?
>
No, symbols must only be unique per domain.

The old class would remain a Builder
>
Just make it not be a `SimpleBuildStep`.

is all this just a consequence of the steps in question getting run via a
> meta-step?
>
Yes.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1uZ1sOE3242hwdvnt534mpXCO6P48NAtaVBC4LUpka3Q%40mail.gmail.com.

Re: Advanced button issue

['Jesse Glick' via Jenkins Developers]

On Fri, Aug 12, 2022 at 6:52 AM Nikhil Bhoski 
wrote:

> I am keeping my configure-advanced.jelly in the same folder as of my
> action class next to config.jelly
>

https://github.com/mathworks/jenkins-matlab-plugin/tree/genscript_advance_button/src/main/resources/com/mathworks/ci/RunMatlabTestsBuilder
has no such file.

I would not suggest trying to `st:include`. This is for advanced use cases
and you probably do not need it. Just put some controls inside `f:advanced`.

You can probably ignore anything you find on wiki.jenkins.io. As this
specific page points out,

much of this document is old and suggests code patterns which are *not
> recommended*
>

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2zY2fwAiBg7_G6%3D%2Bf1L-x7R9jRyajYha-3uLH0iz6aVA%40mail.gmail.com.

Re: Storing the results for git cache maintenance tasks

['Jesse Glick' via Jenkins Developers]

On Thu, Aug 11, 2022 at 11:49 PM Mark Waite 
wrote:

> effective and reasonable ways to store the most recent results of a task
> and present the most recent results of tasks to the administrator
>

There is no standard technique for this. Branch indexing copies a bunch of
code from `Run` and implements its own log storage (no history). Would make
sense to define a convenience API for build-like logs, especially as that
could be a single place to implement https://jenkins.io/jep/207 and so on.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3T4uohKUCBc%3DYw5bv3p%3DnkFe0eEtCFOFwaBTvWY5xL2A%40mail.gmail.com.

Re: Script-security

['Jesse Glick' via Jenkins Developers]

https://github.com/jenkinsci/bmc-change-manager-imstm-plugin/blob/929fd09178bb30bce1787fe4844a75fcf6751444/src/main/java/com/bmc/ims/BmcDlpBuilder.java#L445-L447
is not how this API is designed to be used at all. From
https://javadoc.jenkins.io/plugin/script-security/org/jenkinsci/plugins/scriptsecurity/sandbox/groovy/SecureGroovyScript.html

May be kept as the value of a field and passed in a DataBoundConstructor
> parameter
>

I have no idea what you are attempting to do, but it will probably not
work, or if it does, not be secure. I would advise not attempting to use
in-process Groovy scripting if you can possibly avoid it.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3A5UtxUV%2B%3Dcw2VAjjmjPkQ1Sx5qfDrBpLHtqspvrCv%3DQ%40mail.gmail.com.

Re: Action disappears after restart

['Jesse Glick' via Jenkins Developers]

On Mon, Jul 25, 2022 at 5:38 PM Nozim Islamov <
mukhammadnozim.isla...@gmail.com> wrote:

> I use transient Run
>

If by that you mean you have a `transient` field of type `Run` in your
`RunAction2`, that is normal and correct, not the source of your issues.

In general, for this sort of question it is much more helpful to include a
GitHub URL with a small (minimal), self-contained, reproducible example of
the problem you face.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr03%3DT1wphLCfJkRs9JGTXXcPo0mHeUUeNTOS6-hBx_mgA%40mail.gmail.com.

Re: Parent pom 4.43.1 message from validate?

['Jesse Glick' via Jenkins Developers]

I did not see any such message in `priority-sorter-plugin`. Is there a way
to reproduce?

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1BsphGaDZUiRWhA0cvz-HP1bko2B1VE4Ci0fKKJ2HBcQ%40mail.gmail.com.

Re: Let renovate handle dependency updates in jenkinsci/jenkins

['Jesse Glick' via Jenkins Developers]

Also +1 on Renovate for JS deps (at least unless and until Dependabot
improves in this area) so long as Dependabot is retained for other (mainly
Java) deps.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0tMBxVk7aLoBtW4Svxcg6q-jFY_TrUDcBto0yAmYWznQ%40mail.gmail.com.

Re: Releasing a new plugin fails due to 'Check interesting categories' failure

['Jesse Glick' via Jenkins Developers]

https://github.com/jenkinsci/bmc-change-manager-imstm-plugin/releases seems
to be working now?

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3uPw1NViWH873gup_B5Pqb781GAF4ETo_MCXyaRENywQ%40mail.gmail.com.

Re: Parent pom 4.43.1 message from validate?

['Jesse Glick' via Jenkins Developers]

On Sat, Jul 16, 2022 at 8:09 AM Mark Waite 
wrote:

> for artifact:
>   org.jenkins-ci.main:jenkins-test-harness:jar:1802.v9de0d87365d2
>
> from the specified remote repositories:
>   central (https://repo.maven.apache.org/maven2, releases=true,
> snapshots=false)
>

It is of course not in Central, it is in the Jenkins Artifactory. Which the
POM for at least `priority-sorter` seems to correctly declare. Maybe your
`~/.m2/settings.xml` declares a mirror which is not functioning correctly?
Try

*,!repo.jenkins-ci.org,!incrementals

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1C9RPgFiaG%2B8QuLw%2BqkWLX2MH5k24OGnJewx_zg%3DQupA%40mail.gmail.com.

Re: Changing build status after it finished

['Jesse Glick' via Jenkins Developers]

On Fri, Jul 15, 2022 at 2:03 PM Nozim Islamov <
mukhammadnozim.isla...@gmail.com> wrote:

> I wanted to change the color of build status after the build is completed
>

You cannot change the status of a completed build.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2VW9c4RHkvt2Hjc_Yh3Xt6kdgpnsVrmY7mKzeMxW4FuA%40mail.gmail.com.

Re: Periodic rebuilding of core PRs on ci.jenkins.io through branch indexing

['Jesse Glick' via Jenkins Developers]

By the way, we could try running a GitHub merge queue for the
`jenkins@master` branch. I experimented with it recently. It does prevent
the possibility of an undetected logical conflict, without requiring a
person to be standing by manually updating each PR and waiting for CI to be
green before merge. It does not increase throughput by batching tests—it
still verifies that each proposed trunk commit passes CI individually—and
would increase load on the server compared to the simple policy of merging
an approved PR once its head commit is green. It also seemed buggy at the
time, though of course GitHub is likely ironing out problems without
necessarily announcing that.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0gK0iK24we3S%3DFc034m8wC7BoK-V9a_q_V9EUwhQq%3DOw%40mail.gmail.com.

Re: Periodic rebuilding of core PRs on ci.jenkins.io through branch indexing

['Jesse Glick' via Jenkins Developers]

On Sun, Jul 3, 2022 at 5:13 AM Alexander Brandes 
wrote:

> Maybe I'm missing something, but I can see downsides only for this
> [PR-merge] strategy
>

The upside is that the commit status will reflect whether the PR would
allow trunk to continue passing if it were merged *now*. With the simpler
PR-head strategy, a PR might pass in its original form, and have no textual
merge conflicts (edits within ~3 lines) yet actually be broken relative to
the current trunk. Thus by merging it you might inadvertently break trunk,
which would be disruptive.

In practice this is not a very common problem, and the downsides are very
real, so I recommend switching to PR-head:
https://github.com/jenkins-infra/helpdesk/issues/1355

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2RgzN0-QhOsDiaVyvqzYmH5t9S3f_Du9gpqW6%3DcpvgEg%40mail.gmail.com.

Re: pipeline-syntax/generateSnippet: java.lang.IllegalArgumentException: Unable to convert to class java.lang.Object

['Jesse Glick' via Jenkins Developers]

On Sat, Jul 2, 2022 at 9:11 AM Shihaaz Buhary  wrote:

> We have been using a boolean variable in our plugin (Pipeline type). Now
> we have decided to include more options and decided to go ahead with
> something like 0, 1, 2 String/Integer as well. Meanwhile we want to
> preserve the backward compatibility as well. We do not want to effect any
> customers who are already using true/false. We will internally map false to
> 0 and true to 1. Is there an approach which will serve this purpose?
>

Just introduce new fields with better types, and deprecate the old one
while leaving it untouched.

(`Integer` and `String` are not what you want to represent choices. You can
use an `enum` for simple cases. Where choices have their own structure, use
`Describable` types with `@Symbol`s.
https://www.jenkins.io/doc/developer/plugin-development/pipeline-integration/#defining-symbols
has details.)

We tried Object type
>

Do not do that.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2-5mj%3D_94sOoXkB644G3a5uEo86dZrhPcrPA8jihUHmA%40mail.gmail.com.

Re: Is there interest for GitHub issues in core components?

['Jesse Glick' via Jenkins Developers]

On Fri, Jul 1, 2022 at 3:22 AM 'Daniel Beck' via Jenkins Developers <
jenkinsci-dev@googlegroups.com> wrote:

> we have 130k users in Jira. While many are probably not legitimate users
> of Jira […or] haven't been logged in in years
>

If we expect and even encourage that most plugins switch to GitHub for
issue tracking, so that we expect Jira (Cloud) to be used mainly for core
and associated components and tools (plus the `SECURITY` project), does
that change the calculus? There would still be plenty of people creating an
account just to report some sort of core bug (or a bug which they cannot
easily map to a particular plugin), but probably not tens of thousands of
them.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3w9F1JTRh%3D21NvLSWkUHTtMVmaYo%2BedBT_gdy%3D4q0kVQ%40mail.gmail.com.

Re: Jenkins Test Harness with existing Jenkins instance?

['Jesse Glick' via Jenkins Developers]

On Tue, Jun 28, 2022 at 12:51 PM steven...@gmail.com <
steventerr...@gmail.com> wrote:

> testing scenarios where your controller is running on kubernetes with
> agents.
>

https://github.com/jenkinsci/kubernetes-plugin/blob/908e7db10d06671abf69e592d8b64e8b46cdc58e/kind.sh#L18-L31
https://github.com/jenkinsci/kubernetes-plugin/blob/908e7db10d06671abf69e592d8b64e8b46cdc58e/src/test/java/org/csanchez/jenkins/plugins/kubernetes/pipeline/KubernetesPipelineTest.java#L134-L135
etc.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2FR5eHY8dd8pi%2BtJVRzpX_km%2Be0qckMAJ7w8UvhGYrnQ%40mail.gmail.com.

Re: Jenkins Test Harness with existing Jenkins instance?

['Jesse Glick' via Jenkins Developers]

On Mon, Jun 27, 2022 at 4:37 PM steven...@gmail.com 
wrote:

> Does anyone know if it's possible to configure the Jenkins Test Harness to
> forego standing up an ephemeral Jenkins instance and instead use a
> controller at a given URL with authentication?
>

No, it is not designed for that.

I'm interested in writing integration tests for pipeline libraries.
>

You can do that with `JenkinsRule` if you set up the library definition and
any associated context & infrastructure.
https://github.com/jenkinsci/pipeline-groovy-lib-plugin/blob/aafc25d520e96733e3cfa4b46d1bfdcd015a350e/src/test/java/org/jenkinsci/plugins/workflow/libs/LibraryAdderTest.java#L327-L343
for example.

nothing beats the real thing when testing
>

If by *real thing *you mean an existing server whose configuration you have
somehow managed and presume is stable enough, because you cannot readily
mock out everything you interact with, then there is another way:
https://github.com/jenkinsci/pipeline-groovy-lib-plugin/blob/8fb8c55868cc91791f7e5bc39c40594597329128/src/test/java/org/jenkinsci/plugins/workflow/libs/SCMRetrieverTest.java#L58-L98
The key bit is to create a `Jenkinsfile` in the library which loads itself
via the near-magical `retriever: legacySCM(scm)` idiom and then runs some
smoke tests by way of either passing or failing the build. If you create a
multibranch project on your real controller based on this repo, you can
file a PR refactoring the library and check whether it works in the real
context.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3Dq0kmpT5Eg8TjL8KJiaAvx_G-CNsx5Kno5APj7MCCBQ%40mail.gmail.com.

Re: Pipeline RunAction2 class integration to side-panel

['Jesse Glick' via Jenkins Developers]

https://www.jenkins.io/doc/developer/plugin-development/pipeline-integration/
generally. There is no Pipeline equivalent to
https://javadoc.jenkins.io/hudson/model/JobProperty.html#perform(hudson.model.AbstractBuild,hudson.Launcher,hudson.model.BuildListener)
specifically; you can offer a `Step`.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0Gv6y-x9prJ5_f9D6jw14_Qun6RqFFfJUOFzqLbX4jzQ%40mail.gmail.com.

Re: Question about SSHUserPrivateKey snapshots and git checkouts

['Jesse Glick' via Jenkins Developers]

On Mon, Jun 13, 2022 at 12:58 PM James Robson 
wrote:

> I had created set of vms on my laptop just for this work
>

I would strongly recommended writing an integration test using
Testcontainers to do things like retrieve an SSH private key from Vault
running in a Docker container (using a specific pinned version of all
relevant software—let Dependabot keep your `Dockerfile` fresh) and then
passing it to an agent running in another container (again with a specific
version of Git, etc.) so as to check out from an authenticated Git server
running in yet another container. This would allow you or anyone else
working on the plugin to debug and examine the complete flow in a realistic
environment, without concern that a test environment might have become
corrupted or subtly altered; and ci.jenkins.io can run such a test as well
if you supply the `docker` label in `Jenkinsfile`.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3W0e-44E79GtJp_2pgEp5o9UfgX088_6_A%3DNmYhMrfgw%40mail.gmail.com.

Re: BOM build errors due to conflict of baseline 2.319.3 and 2.319.1

['Jesse Glick' via Jenkins Developers]

On Mon, Jun 13, 2022 at 12:43 PM Ullrich Hafner 
wrote:

> Shouldn’t the BOM also follow our guidelines and support only
> releases 2.303.3 and 2.319.3 and newer?
>

Perhaps. The current guideline for the plugin BOM is stated here:
https://github.com/jenkinsci/bom/#lts-lines

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1oiRmF3D13DVSjET%3D-drKMo10HoeBTCVKqwvFUiSvEZg%40mail.gmail.com.

Re: Is there interest for GitHub issues in core components?

['Jesse Glick' via Jenkins Developers]

On Mon, Jun 13, 2022 at 10:02 AM 'Daniel Beck' via Jenkins Developers <
jenkinsci-dev@googlegroups.com> wrote:

> On Mon, Jun 13, 2022 at 3:55 PM Tim Jacomb  wrote:
>
>> The barrier to entry on Jira is a lot higher than on GitHub, many people
>> struggle to report issues.
>>
>
> Is people not reporting issues they experience really a problem we have?
>

FWIW I filed JENKINS-68727 
just a few days ago after receiving private email from someone saying

I would like to report a bug, but the bugtracker on jira does not allow me
> to create a new account.
>

Not the first time I have gotten such a message. I do not know if this
person had a GitHub account.

It is true that there are more bugs filed than people qualified and willing
to triage them. Perhaps making it awkward to report (or comment on) bugs
helps by filtering out people who would have filed vague and unreproducible
reports anyway; or perhaps it just discourages everyone uniformly.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0gxeAhhkP78pBzF5Vppd%3DZmR7u9XAL%3Dcj%3DpAfMcECodQ%40mail.gmail.com.

Re: Plugin Updates

['Jesse Glick' via Jenkins Developers]

On Wed, May 25, 2022 at 1:37 PM priya jagyasi 
wrote:

> Can anyone please let me know how much time it takes to have the plugin
> under 'Updates' in the Jenkins update center for any instance after the
> GitHub workflow release is completed successfully? Is it immediate or any
> approximate time?
>

Generally it should appear within the hour.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2UPhGZXFZ6xuoOkadv%2BkAVa_sH3V5w55aSbg%3D2Av0h4w%40mail.gmail.com.

Re: 409 errr on plugin release

['Jesse Glick' via Jenkins Developers]

https://github.com/jenkinsci/lightstep-incident-response-plugin/pull/6

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0P_S0hO_R%3DoF6%2BEPAKNWOCS0fLvER1sAKy0fqUs9aCAA%40mail.gmail.com.

Re: 409 errr on plugin release

['Jesse Glick' via Jenkins Developers]

On Tue, May 24, 2022 at 1:09 PM priya jagyasi 
wrote:

> could you please let me know what needs to be done to enable CD on my
> plugin.
>

Please read the instructions.
https://www.jenkins.io/doc/developer/publishing/releasing-cd/#update-maven-pom-and-config

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0e1S5kZ1ofgGB3FwQgzv8irkbNMmXJ%2BiP5Na2hFEu6Aw%40mail.gmail.com.

Re: 409 errr on plugin release

['Jesse Glick' via Jenkins Developers]

On Sat, May 21, 2022 at 4:21 PM 'Daniel Beck' via Jenkins Developers <
jenkinsci-dev@googlegroups.com> wrote:

> You are trying to release a plugin using JEP-229 CD without having it
> configured correctly
>

As mentioned in
https://github.com/jenkinsci/lightstep-incident-response-plugin/pull/1#issuecomment-1132113366
your POM remains incorrect for a CD-enabled plugin:
https://github.com/jenkinsci/lightstep-incident-response-plugin/blob/82fa43c17e6c1c8ab097cf0502c153edb032fad2/pom.xml#L16-L23
is the metadata for an incrementalified but not CD-enabled plugin.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0mvhwbTtj6WjGnDomdrvKLeYv1gu1dduwOvtz-2dBzxQ%40mail.gmail.com.

Re: Configure release drafter

['Jesse Glick' via Jenkins Developers]

>
> On Thursday, May 19, 2022 at 5:26:51 AM UTC-6 Priya wrote:
>
>> The step here
>> https://www.jenkins.io/doc/developer/publishing/releasing-cd/#configure-release-drafter
>>
>
Are you intending to configure your plugin for continuous delivery?

says to include *_extends: .github* on
>> https://github.com/jenkinsci/.github/blob/master/.github/release-drafter.yml
>> file. Could you help how to do it? I see this file already has some
>> content, what is expected here to do?
>>
>
 As the page says:

Make sure to remove any tag-template override that may have been configured
> previously.


and:

Optionally you can also remove .github/release-drafter.yml as the default
> is to assume _extends: .github
>

If you are *not* intending to configure your plugin for continuous
delivery—you expect to release it yourself, with traditional-style version
numbers like 1.3 etc.—then you should not be on this page.

There are other problems here.
https://github.com/jenkinsci/lightstep-incident-response-plugin/pull/1#issuecomment-1132113366

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1qvdghD1QdqGSPfs1UXFAUasU8kEdcGrZU5_M%2B6pHk1A%40mail.gmail.com.

Re: Guidance on whether/where to file issues for build warnings

['Jesse Glick' via Jenkins Developers]

On Fri, May 13, 2022 at 5:50 PM Tim Van Holder 
wrote:

> I don't have any direct references to either of these. Is this a bug in
> the BOM?
>

Or the parent POM, or something. If you manage to track down the root
cause, please file a PR to fix it.

And these warnings went away when using Maven 3.8.5 instead of 3.8.3.
>

Do not use 3.8.5; it is known to be broken. Use 3.8.4.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2-K%2BudRWruddTjtTXZVJS1apFuaRgp8%2BNNo1%2B2gPvHSA%40mail.gmail.com.

Re: Moving ATH tests closer to the code under test

['Jesse Glick' via Jenkins Developers]

On Tue, Apr 26, 2022 at 12:28 PM Basil Crow  wrote:

> I wonder if we could design a system such that for the common case only a
> single Maven project is needed


Sounds tricky. `acceptance-test-harness` has a bunch of dependencies some
of which clash with those in Jenkins core or some plugins, so you would
need to either shade them all, or otherwise somehow ensure the ATH
dependency trails are given minimum priority.

I wonder how the ATH implementation compares to RealJenkinsRule


You could in principle use `RealJenkinsRule` with some GUI testing library
other than HtmlUnit that better supports exotic JavaScript. If you wanted
to reuse the rich Selenium page object library in ATH you would need to
factor that out into a standalone library.

There are other significant differences:

   - Although ATH uses RESTish endpoints for a few purposes, for the most
   part the test *setup* (not just the actual assertions) uses the browser.
   In all cases the whole test run is “black-box”. This has its appeal
   (stronger coverage) but GUI setup makes tests much slower and either REST
   or GUI setup can also make it a lot more awkward to write tests than when
   using `[Real]JenkinsRule`, which are “white-box” and let you quickly set up
   an environment and run some assertions at the Java level. JTH also lets you
   write test-only extensions, which ATH does not.
   - ATH installs plugins, and selects plugin versions, using user-mode
   tools. `[Real]JenkinsRule` follow the Maven test classpath. Possibly ATH
   could be given an option to define a mock UC based on a test classpath.
   - ATH uses Guice for configuration. JTH does not.
   - ATH uses `docker-fixtures` for containerized tests. JTH can use that
   also, but Testcontainers is recommended instead.



> the benefit of this idea depends on _how_ fragile.


Also whether the fragility is in the actual assertions, or test setup
(point #1 above).

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2HJN1KKt0D5c%2BYsPAhMMvw5ThwGTbqxBLGjEaEckRUwA%40mail.gmail.com.

Re: get all agents used by a build

['Jesse Glick' via Jenkins Developers]

On Fri, Apr 22, 2022 at 11:18 AM Mathieu DELROCQ 
wrote:

> get the duration of the stage wich use the agent
>

Should be possible to calculate given
https://javadoc.jenkins.io/plugin/workflow-api/org/jenkinsci/plugins/workflow/actions/TimingAction.html

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0V0%3Dtdg%3D0bQSV46t31wYE1ASTxREhp_pkxW9otH_1-gA%40mail.gmail.com.

Re: get all agents used by a build

['Jesse Glick' via Jenkins Developers]

On Wed, Apr 20, 2022 at 4:03 AM Mathieu DELROCQ 
wrote:

> I have a Run object and I would like to get all agents used in this run.
>

For `AbstractBuild` (traditional job types), there is a getter. For
Pipeline (`FlowExecutionOwner.Executable`) you would need to iterate the
`FlowNode`s looking for `WorkspaceAction` (this would be added to the block
entry nodes for the `node` step).

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0hhKzUjUXg1qYxyUR-OcNkiKpbViV3yjtnJvLE8SL7JA%40mail.gmail.com.

Re: Checking whether a step from a plugin is in actual use in an instance

['Jesse Glick' via Jenkins Developers]

On Mon, Apr 18, 2022 at 3:16 PM 'Daniel Beck' via Jenkins Developers <
jenkinsci-dev@googlegroups.com> wrote:

> Potentially doable via readResolve as well.
>
>
Unfortunately you cannot do this sort of thing via `readResolve` so far as
I know.

Ideally there would be some API which you could call from `readResolve` or
otherwise mark a class which, if loaded in the context of some top-level
model object (in this case `Project`), would trigger a callback after the
top-level load completes, which you could use to immediately rewrite the
reference, add to `OldDataMonitor`, etc.

In the absence of that, your only option is to add a global initialization
hook which scans all build steps in all projects. Obviously this adds
startup overhead for everyone who installs your plugin, even if it is
unused. You could use a `GlobalConfiguration` (or simple `FileBoolean`) to
track whether the translation has been run before.

Another common use case is migrating `Secret` to `credentialsId`.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2v5kykDcTky_Y_9eT93n%2BDJ_sfK7LO6P0rph13ismRvQ%40mail.gmail.com.

Re: Checking whether a step from a plugin is in actual use in an instance

['Jesse Glick' via Jenkins Developers]

On Mon, Apr 18, 2022 at 4:38 AM Tim Van Holder 
wrote:

> Can a plugin include code that will update a freestyle project that uses
> the DotNetFoo builder to use the DotNet builder with a Foo argument instead?
>
It is possible with an `@Initializer`.

I'd have to keep them around for pipeline use anyway
>
Correct.

is there anything in Jenkins that will report warnings for use of
> deprecated steps in a pipeline job?
>
No, but you could print a message to a `TaskListener` in case the user is
paying attention. If you want to be more aggressive, there is always
`AdministrativeMonitor`.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1hYyguDRxfxdWGyEKp_Yv%3D-XxjnGVbUhAnLGFMQQsKEQ%40mail.gmail.com.

Re: Jenkins build failure: "...project cli: Error creating shaded jar: duplicate entry: META-INF/services/io.jenkins.cli.shaded.jakarta.websocket.ContainerProvider"

['Jesse Glick' via Jenkins Developers]

On Fri, Apr 15, 2022 at 12:08 PM 'Cyrille Le Clerc' via Jenkins Developers <
jenkinsci-dev@googlegroups.com> wrote:

> Is it a known build problem?
>

Other than Maven 3.8.5 being broken as Mark mentioned, there are not in
general known build problems. Someone would fix the problem right away, if
it even somehow got merged to trunk despite CI checks.

I would suggest starting with `mvn clean`. If that still fails, try a build
of a fresh checkout inside a Docker container.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3DGfFqG2yykcusa4WtS%2BU0RQhj1kTaEv8ygu59TCPhJw%40mail.gmail.com.

Re: Capturing the remoteIpAddr on login failure?

['Jesse Glick' via Jenkins Developers]

On Wed, Apr 13, 2022 at 5:21 AM 'Cyrille Le Clerc' via Jenkins Developers <
jenkinsci-dev@googlegroups.com> wrote:

> did we consider making the Jenkins SecurityListener
>  an
> adapter of the Spring ApplicationEventPublisher
> 
>

I do not believe that was ever considered or proposed. The focus was on
moving from Acegi Security idioms to the nearest Spring Security equivalent
with the minimum effort required. Expanding the coupling between Jenkins
authentication and Spring Security specifics feels like the wrong
direction—the migration to Spring Security would have been much easier had
the vast surface area of the Acegi Security API not been directly exposed
to plugins.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0GWYXMNSu%2BMOwpm%3DG5pOakqhgXq-%3DsjLhN9H0ZMP%3DQJw%40mail.gmail.com.

Re: Checking whether a step from a plugin is in actual use in an instance

['Jesse Glick' via Jenkins Developers]

On Tue, Apr 12, 2022 at 1:40 PM Tim Van Holder 
wrote:

> a relatively simple way to have a build step for freestyle that would
> essentially then have a dropdown for the 11 "real" steps which then shows
> their configuration when selected
>

This is straightforward. (`ui-samples-plugin` may be instructive.) Just an
abstract `Describable` supertype for the real things, and the `config.xml`
for the `Builder` can have a `f:dropdownDescriptorSelector` and maybe
nothing else. Each real thing has its own `config.jelly`. If you also add
`@Symbol`’s, then for Pipeline you would get syntax like

dotnet tool: makeAssembly(assemblySpecificOption: true), genericOption:
'whatever'

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr3eG-v5vNV4ScKYLPHRSCgCheLjGJsQ9H7_JVCbmTx6Xw%40mail.gmail.com.

Re: Java 17 support added to build toolchain

['Jesse Glick' via Jenkins Developers]

Thanks Basil! Mark I added an analysis in your PR.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0B28N_v6UYAt5U4AEq98K6%3DenTcBfO3omfkukJcn%3DZJg%40mail.gmail.com.

Re: New maven message from hpi plugin 3.26?

['Jesse Glick' via Jenkins Developers]

Filing as https://github.com/jenkinsci/maven-hpi-plugin/issues/325.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1_DVg-RaAvcCYx01eoNCCTEchqJjZx8LCKroX9JNhZoA%40mail.gmail.com.

Re: Governance meeting Feb 23, 2022

['Jesse Glick' via Jenkins Developers]

On Wed, Feb 23, 2022 at 6:46 PM Slide  wrote:

> Windows installers have issues with versions like this


Interesting. In general native packagings may have to process the version
number to follow specific conventions or restrictions.

The product version is already mangled to support LTS releases, it might
> get confusing as to what version is installed if we have to mangle more.


Indeed a native package version ought to at least visually include the true
(Maven) version.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr1ivR9LXjmrJdGfkDpPGSueXLX4TimBKmF9%3DNXme-UtKg%40mail.gmail.com.

Re: Java 11 as minimum? (Jetty 9.4.x EOL)

['Jesse Glick' via Jenkins Developers]

On Thu, Feb 24, 2022 at 8:42 AM Mark Waite 
wrote:

> more exploring of potential issues related to the transition.
>>
>
My main concern is about bugs, like JENKINS-59910 which no one seems to be
evaluating. If stuff works when run on 8 and does not work when run on 11
then users will not accept the upgrade.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to jenkinsci-dev+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr2_vAA6m3BiW1MzvQdx5dq-rHi0gxeiJcEV2Gd6grRdWQ%40mail.gmail.com.