gemmellr commented on code in PR #4540:
URL: https://github.com/apache/activemq-artemis/pull/4540#discussion_r1255503083
##########
artemis-hawtio/artemis-console/pom.xml:
##########
@@ -55,18 +55,6 @@
<version>${hawtio.version}</version>
<scope>provided</scope>
</dependency>
- <dependency>
- <groupId>com.google.guava</groupId>
- <artifactId>guava</artifactId>
- <!-- License: Apache 2.0 -->
- <!-- see https://github.com/google/guava/issues/3006:
checker-compat-qual could be excluded -->
- <exclusions>
- <exclusion>
- <groupId>org.checkerframework</groupId>
- <artifactId>checker-compat-qual</artifactId>
- </exclusion>
- </exclusions>
- </dependency>
Review Comment:
This was here to override the version of guava in the repackaged hawtio war
for the console. See the exclusion further down that is still present. Think we
will need to check what happens to the management console here, its possible
this may have broken it, but at least removed the version override.
##########
pom.xml:
##########
@@ -1097,6 +1094,13 @@
<arg>-XDcompilePolicy=simple</arg>
<arg>-Xplugin:ErrorProne
-Xep:ThreadLocalUsage:ERROR -Xep:MissingOverride:ERROR
-Xep:NonAtomicVolatileUpdate:ERROR -Xep:SynchronizeOnNonFinalField:ERROR
-Xep:StaticQualifiedUsingExpression:ERROR -Xep:WaitNotInLoop:ERROR
-Xep:BanJNDI:OFF -XepExcludedPaths:.*/generated-sources/.*</arg>
</compilerArgs>
+ <annotationProcessorPaths>
+ <path>
+ <groupId>com.google.errorprone</groupId>
+ <artifactId>error_prone_core</artifactId>
+ <version>${errorprone.version}</version>
+ </path>
+ </annotationProcessorPaths>
Review Comment:
I know that Emmanuel deliberately avoided doing this when overhauling the
ErrorProne usage a while back for JDK > 9, as it restricts use of annotation
processors from the classpath and requires every processor be specified here in
the XML. We do have another annotation processor being used via the classpath
(artemis-log-annotation-processor).
I would guess that is what caused the 'full test run' CI failures
@clebertsuconic noted since I know that build enables ErrorProne during
compile, meaning it probably then failed to run the other annotation processor
as it isnt also specified here (or rather outwith these profiles), and so
failed to generate the logger classes and then tripped over their absence.
Since ErrorProne is already isolated to a profile I think I would leave this
as it was. At least I dont think it really belongs in the same commit as a swap
of caching to use Caffeine.
##########
artemis-protocols/artemis-mqtt-protocol/src/main/java/org/apache/activemq/artemis/core/protocol/mqtt/MQTTUtil.java:
##########
@@ -380,7 +380,7 @@ private static String formatByte(byte bite) {
}
private static String formatCase(String string) {
- return CaseFormat.UPPER_UNDERSCORE.to(CaseFormat.LOWER_CAMEL, string);
+ return CaseUtils.toCamelCase(string, false, ' ');
Review Comment:
This seems off? If its going from _UPPER_UNDERSCORE_ its unlikely to have
space delimiters..seems more like it will end up with _lower_underscore_ after
this? Or does the original also not do what it seems like its trying?
--
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]
