feat(extract-subsystems): extract Context, @In, InjectionHelper and @Share to Subsystem
#4930
Conversation
… and `@Share` to Subsystem
github-actions
bot
added
the
Type: Improvement
There was a problem hiding this comment.
CI did not run any org.terasology.engine.registry tests: https://jenkins.terasology.io/teraorg/job/Terasology/job/engine/job/PR-4930/lastCompletedBuild/testReport/
| @@ -31,7 +26,7 @@ public void setup() { | |||
| @Test | |||
| public void testContextChange() { | |||
| CoreRegistry.setContext(new ContextImplementation()); | |||
| assertNotEquals(CoreRegistry.get(Context.class), context); | |||
| Assertions.assertNotEquals(CoreRegistry.get(Context.class), context); | |||
There was a problem hiding this comment.
Was this intentional? (Adding the Assertions. qualifier back in, instead of a static import of the assertNotEquals method.)
There was a problem hiding this comment.
Just personal prefer to avoid static imports
Nice catch! Our pipeline ignores
This affects all |
|
about |
|
@DarkWeird In today's contributor meeting the question came up, whether it makes sense to invest into this when the "proper" way to do this seems to be gestalt v8...? |
see description:
this pr is trying to split injections before gestalt v8 usage |
|
@DarkWeird Hehe okay, let me rephrase my question: What's the benefit ofextracting these things into a subsystem before using gestalt v8? |
Path to extract another things (to subsystems) before then gestalt v8 will be integrated. |
|
From my pov, if this makes it easier to get rid of it after migrating to gestalt v8, I'm fine with reviewing and merging it (provided you resolve the conflicts first 😅 ) |
Contains
Part of #4304.
Provide possible to Injectable classes(
@Share,@InandContextusage) with another subsystems.Should be replaced with gestalt v8 (di)
How to test
1.Run your favorite game (hint: Terasology)
2. Check it don't fail whit null pointer exception at
@Infield.Outstanding before merging