voonhous commented on code in PR #19020:
URL: https://github.com/apache/hudi/pull/19020#discussion_r3586248409


##########
pom.xml:
##########
@@ -379,6 +379,17 @@
         <configuration>
           <source>${java.version}</source>
           <target>${java.version}</target>
+          <!-- Declare the annotation processor path explicitly instead of 
relying on classpath

Review Comment:
   nit: worth a sentence here (or in the description): since JDK 23 javac 
disables annotation processing unless explicitly configured, and the 
`-processorpath` this produces counts as explicit - so this also un-breaks 
Lombok under JDK 23+ toolchains, where it is currently skipped silently. 
Strengthens the case beyond the `-T` race.



##########
pom.xml:
##########
@@ -379,6 +379,17 @@
         <configuration>
           <source>${java.version}</source>
           <target>${java.version}</target>
+          <!-- Declare the annotation processor path explicitly instead of 
relying on classpath
+               service discovery. Lombok is the only annotation processor in 
the build, and an
+               isolated processor path avoids the flaky "Provider lombok... 
could not be
+               instantiated" ServiceLoader race seen under parallel (-T) 
reactor builds. -->
+          <annotationProcessorPaths>

Review Comment:
   nit: the PR description says this lives in the parent `<pluginManagement>`, 
but it is the root `<build><plugins>` entry (the `<pluginManagement>` compiler 
entry further down is a bare placeholder). Same outcome - `build/plugins` is 
the stronger placement - just worth fixing the description so readers look in 
the right section.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to