Bump emf version to match the content of 2024.3

[szarnekow]

No description provided.

[cdietrich]

nice, can you please check

• SimpleTestLanguageFragments.mwe2
• we still have complianceLevel 11 a lot

[cdietrich]

the swt bot change does not work for me. the versions oomph pulls for me is 4.0.0

[cdietrich]

it also looks like a transitive reexport of guice got lost somewhere. see failed build

[szarnekow]

it also looks like a transitive reexport of guice got lost somewhere. see failed build

$git diff origin/main | grep google
+import com.google.common.collect.ForwardingMap;

hmm

[szarnekow]

But javax.inject doesn't seem to be there anymore

Error:  Failed to execute goal org.apache.maven.plugins:maven-compiler-plugin:3.11.0:compile (default-compile) on project org.eclipse.xtext.web.example.statemachine: Compilation failure
Error:  /home/runner/work/xtext/xtext/org.eclipse.xtext.web.example.statemachine/src-gen/org/eclipse/xtext/web/example/statemachine/AbstractStatemachineRuntimeModule.java:[131,37] cannot access javax.inject.Provider
Error:    class file for javax.inject.Provider not found
Error:  -> [Help 1]

[github-actions]

Test Results

  6 461 files  +  6 461    6 461 suites  +6 461   3h 25m 25s ⏱️ + 3h 25m 25s
 43 239 tests + 43 239   42 655 ✅ + 42 655    584 💤 +  584  0 ❌ ±0 
170 246 runs  +170 246  167 904 ✅ +167 904  2 337 💤 +2 337  5 ❌ +5 

Results for commit 881c922. ± Comparison against base commit 4fb59cc.

♻️ This comment has been updated with latest results.

[cdietrich]

werent we using com.google.inject.Provider everywhere to circumvent the jaxax jakartax problem?

[cdietrich]

regenerating langs fails for me everywhere witjh
Caused by: java.lang.ClassNotFoundException: org.slf4j.LoggerFactory
at java.base/jdk.internal.loader.BuiltinClassLoader.loadClass(BuiltinClassLoader.java:641)
at java.base/jdk.internal.loader.ClassLoaders$AppClassLoader.loadClass(ClassLoaders.java:188)
at java.base/java.lang.ClassLoader.loadClass(ClassLoader.java:526)
... 35 more

likely due to our wild mix of p2 and maven

[cdietrich]

can reproduce the compile error but still have no clue what causes it

apache-maven-3.9.9/bin/mvn clean verify -f org.eclipse.xtext.maven.releng -B  -PuseJenkinsSnapshots -Pstrict-jdk-21

[cdietrich]

one suspicious diff in classpath is

/Users/dietrich/.m2/repository/.cache/tycho/org.eclipse.m2e.maven.runtime-3.9.900.20240824-1409.jar/jars/org.eclipse.sisu.inject-0.9.0.M3.jar:

i dont see that one on main

maybe something with the m2e changes?

[cdietrich]

but i have no idea why this is there at all

@HannesWell do you have an idea?

[cdietrich]

wait. why do have have

org.eclipse.xtext.util/build.properties:additional.bundles = org.eclipse.m2e.maven.runtime

[LorenzoBettini]

one suspicious diff in classpath is

/Users/dietrich/.m2/repository/.cache/tycho/org.eclipse.m2e.maven.runtime-3.9.900.20240824-1409.jar/jars/org.eclipse.sisu.inject-0.9.0.M3.jar:

i dont see that one on main

maybe something with the m2e changes?

I seem to understand that's the new additional bundle, maybe due to the change in bump versions that reads the dev bom?

[cdietrich]

Maybe tycho sips add. Bundles into downstream classpathes too

[cdietrich]

Removing the additional helped locally. Reverting the manifest changes not

[LorenzoBettini]

Removing the additional helped locally. Reverting the manifest changes not

I don't understand: what's solved and what's not?

[cdietrich]

Reverting the two manifest changes regarding M2e still compile error
Reverting the add bundle this compile errors gone.

[szarnekow]

I’ll move BumpVersions to its own bundle

[HannesWell]

/Users/dietrich/.m2/repository/.cache/tycho/org.eclipse.m2e.maven.runtime-3.9.900.20240824-1409.jar/jars/org.eclipse.sisu.inject-0.9.0.M3.jar:

maybe something with the m2e changes?

Recently there haven't been any change in the embedded Maven-runtime in m2e:
https://github.com/eclipse-m2e/m2e-core/commits/master/org.eclipse.m2e.maven.runtime

Or do you mean why Sisu is included in the runtime? Sorry if I don't get the question right.

[cdietrich]

yes was wondering about sisu
and why tycho puts add. bundles to transitive classpaths

[szarnekow]

I’ll move BumpVersions to its own bundle

Done.

Build is certainly making more progress now.

I had to bump log4j to 1.2.26 because 1.2.25 disappeared

[szarnekow]

regenerating langs fails for me everywhere witjh

How do you regenerate them?

[cdietrich]

regenerating langs fails for me everywhere witjh

How do you regenerate them?

go through the list and calling the workflows from eclipse.
am not sure if it is related to me having mwe imported in workspace

https://github.com/eclipse-xtext/xtext/blob/main/dev-doc/RegenerateLanguages.md

[cdietrich]

hi, what is the status on the open points?
or should we cherry pick the log4j changes to unblock main

[szarnekow]

On this branch, I can regenerate languages.
I'm working on a workflow that builds them all from xtext.relengutils

Of course I've plenty changes since almost all files get a new copyright header

[cdietrich]

Of course I've plenty changes since almost all files get a new copyright header

• which emf/tp version do you use to regen?
• isnt the header configured in the genmodels?

[szarnekow]

Ok, a single script doesn't work because the workflows are sensitive to global registry content.

[cdietrich]

so maybe a launch group similar to the one in ref projects.maybe with wait on each

[szarnekow]

Yes, I'm looking into that.

which tp

2024-03

[szarnekow]

isnt the header configured in the genmodels?

Yes, it is. Some files have not been regenerated yet in 2024

[cdietrich]

maybe we can do it in a follow up then if there is no significant other change

[szarnekow]

maybe we can do it in a follow up then if there is no significant other change

Yes, we surely can do that

[cdietrich]

otherwise what is open besides
lorenzos comments
e.g. on org.eclipse.xtext.relengutil/.classpath

[szarnekow]

otherwise what is open besides lorenzos comments

Made adjustments accordingly and I think we are good to go now.

[LorenzoBettini]

Thanks!

[cdietrich]

thx @szarnekow