-
Notifications
You must be signed in to change notification settings - Fork 24
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
update to MicroProfile Context Propagation 1.3 #343
Conversation
This is very much work in progress, there's a ton of test failures that I don't understand (Weld screams about unsatisfied dependencies) and I don't understand anything about |
Hmm, some of those error are really weird.
For instance this test doesn't even use the Either way, the issue seems to be with injected TC and ME instances. E.g. functionality provided by this CDI extension |
Ouch, I did it again! Or forgot to do it, actually. I was struggling for quite a while with MicroProfile Fault Tolerance TCK in SmallRye Fault Tolerance, only to figure out that I've changed everything except the name of the |
Fixed. Now, I really don't feel like dealing with |
Ah OK, that's RestAssured still depending on Jakarta EE 8 stuff. |
Replaced the usage of RestAssured with Apache HTTP Client and the |
@Ladicek the CI failure is related to formatting BTW.
Hm, don't know about that. In this codebase I only found https://github.com/smallrye/smallrye-context-propagation/blob/main/tests/src/test/java/io/smallrye/context/storage/RESTEasyContext.java |
OK, so I found that RESTEasy moved a lot of MicroProfile stuff to a separate repo https://github.com/resteasy/resteasy-microprofile, including Context Propagation, and that also entails changed Maven coordinates. What used to be |
OK the formatter is bloody idiotic, I changed the commented out piece of code to appease its tasteless taste. |
FTR, there are some strange error messages that are caused by a bug in RESTEasy (https://issues.redhat.com/browse/RESTEASY-3073). I fixed that in resteasy/resteasy#3004 |
OK, that test failure was a silly mistake I made when transforming the RestAssured code to use Apache HTTP Client. All should be passing now. |
LOL
That's some dedication. Thanks for looking deep into this :-) |
OK, all tests are passing, so I'm undrafting this. The 1st commit seems fine to me, though the 2nd one certainly deserves some discussion. |
e5e8e9f
to
764477b
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Both commits look good to me. Thanks @Ladicek!
@@ -93,6 +96,23 @@ | |||
<artifactId>rxjava</artifactId> | |||
<version>${version.rxjava1}</version> | |||
</dependency> | |||
|
|||
<!-- SmallRye Parent 34 manages Weld to an older version that doesn't have JBoss Class File Writer 1.2.5.Final --> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Shouldn't there be a parent update anyway before all other SR projects move to jakarta namespace?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Parent is actually updated in this commit, and smallrye-jakarta-parent
34 should be fine. Except it has old Weld version, which leads to the infamous issue of jbossas/jboss-classfilewriter#24 :-)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Oh, I see. It has Weld version 4 but for some reason not the latest release :-/
Just FTR, it should be enough to just override the classfilewriter version here but having latest Weld version is likely even better.
Ah good point. I'm not sure why MP ConProp went with a minor bump, all other MP specs (that I know of) have bumped a major. I'd suggest bumping a major here, too. Let me add a commit. |
Ah and one other thing -- I believe what SmallRye projects do right now is having a dedicated |
This includes the move from Jakarta EE 8 (javax.*) to 9 (jakarta.*).
This is to workaround RestAssured's dependency on Jakarta EE 8. See rest-assured/rest-assured#1510.
Bumped to |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
PR is LGTM, thanks for bumping the version as well.
Let's wait for @radcortez regarding the branch question before actually merging this.
I'd assume that main
moves to jakarta
and we create another branch for javax
but that's apparently not what's planned :)
Let me help here. So jakarta changes should go into new |
+1 The reason for this is because Quarkus is still on javax and most of our development is targeting Quarkus, so I believe it would be harder for us to have to backport all the work to javax. |
I've created the |
This includes the move from Jakarta EE 8 (javax.*) to 9 (jakarta.*).