also misses change in org.eclipse.xtext.dev-bom/pom.xml

/org.eclipse.xtext.xtext.wizard/src/org/eclipse/xtext/xtext/wizard/TargetPlatformProject.xtend
also has the 0.0.0

so where is the bad one coming from?
or is the problem that 2025-12+I-builds has a mixture of both

My workspace is dirty including that file you mention.

I only added the one *.target file change to see if that fixes the jenkins build. Depending on that we can try to move forward.

I think the failing test has already passed in the Jenkins build:

I think you are saying that those ranges also need to be here:

Of course the other *.target files need this change too:

For #3530 (comment) I think you are saying these version should match exactly the current version is the current latest Orbit aggregation?

Please correct if I got any of that wrong.

so where is the bad one coming from?
or is the problem that 2025-12+I-builds has a mixture of both

I answered that already here in #3528 (comment)

https://download.eclipse.org/tools/orbit/simrel/orbit-aggregation/table.html

Test Results

  6 445 files  + 15    6 445 suites  +15   3h 22m 43s ⏱️ - 1m 22s
 43 212 tests  -  23   42 629 ✅  -  18    581 💤  -  3  2 ❌  - 2 
169 931 runs  +104  167 598 ✅ +111  2 325 💤  - 11  8 ❌ +4 

For more details on these failures, see this check.

Results for commit 0808593. ± Comparison against base commit 1d41d02.

org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf00100
org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf00200
org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf00400
org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf00800
org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf01600
org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf03200
org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf06400
org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf12800
org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf25600
org.eclipse.xtend.core.tests.building.XtendIncrementalBuilderPerformanceTest ‑ testPerf51200
…

♻️ This comment has been updated with latest results.

My workspace is dirty including that file you mention.

@merks unfortunately some of the dirty files are related to the Xtend compiler and are not to be considered.
We still haven't found a way to avoid that.

The GHA failures look unrelated to asm and junit changes. Perhaps a change in jdt or the platform? The Jenkins build was successful.

dont know whats broken/changed in platform / jdt that could break AutoEditTest
somehow an enter does no longer auto insert a star

So is this a blocker to merging this PR? It seems to me orthogonal and something that can be dealt with later, time/resource permitting.

i dont have time or infrastructure to investigate ..

@LorenzoBettini @szarnekow wdyt?

have started https://ci.eclipse.org/xtext/job/xtext/job/PR-3530/4/ do see proper test results.
these logs from github are unreadanle and i cannot find the link to the rest report for gh actions anymore

it looks like its broken at least since the 9th
https://ci.eclipse.org/xtext/job/xtext/job/main/2450/

update: 8th
https://ci.eclipse.org/xtext/job/xtext/job/cd_tycho50/935/

This is the original bug related to those failing tests: https://bugs.eclipse.org/bugs/show_bug.cgi?id=453205

Can we ask someone from platform?

looking at platform.ui log did not spot anything obvious

FYI, test results are here:

https://github.com/eclipse-xtext/xtext/pull/3530/checks?check_run_id=52627697324

It seems relatively readable if somewhat hard to find.

maybe eclipse-platform/eclipse.platform.ui@1848058

created #3531

thx @iloveeclipse @merks

@iloveeclipse do you know about the platform change around projection viewer?

Given that large parts of the Platform itself are now at Java 21, I'm doubtful that Java 17 launching is workable when using a latest target platform.

Of course one can always added a 17 and set it as the default, for for "latest" I think it needs to be this way.

no its not
therefor in the old tps/workspaces i use 17 and in the new one 21

Given that large parts of the Platform itself are now at Java 21, I'm doubtful that Java 17 launching is workable when using a latest target platform.

Of course one can always added a 17 and set it as the default, for for "latest" I think it needs to be this way.

I agree.