dimas-b commented on code in PR #3281:
URL: https://github.com/apache/polaris/pull/3281#discussion_r2624592096
##########
persistence/relational-jdbc/src/main/java/org/apache/polaris/persistence/relational/jdbc/RelationalJdbcConfiguration.java:
##########
@@ -18,8 +18,10 @@
*/
package org.apache.polaris.persistence.relational.jdbc;
+import io.smallrye.config.ConfigMapping;
import java.util.Optional;
+@ConfigMapping(prefix = "polaris.persistence.relational.jdbc")
Review Comment:
I do not oppose this change, but I'd like to highlight that this will impose
the config naming convention defined in this annotation on downstream projects.
It may be possible to provide a different config object for JDBC downstream or
it may not be... I cannot say for sure.
Old code delegated runtime configuration to the server runtime, so this may
be a behaviour change in downstream projects.
Given that NoSQL code follows a similar approach (not splitting interfaces
and `@ConfigMapping`) this change looks reasonable to me.
##########
persistence/relational-jdbc/build.gradle.kts:
##########
@@ -33,8 +33,11 @@ dependencies {
compileOnly(libs.jakarta.enterprise.cdi.api)
compileOnly(libs.jakarta.inject.api)
+ compileOnly(libs.smallrye.config.core) // @ConfigMapping for Quarkus
integration
implementation(libs.smallrye.common.annotation) // @Identifier
implementation(libs.postgresql)
+ implementation(platform(libs.quarkus.amazon.services.bom))
Review Comment:
I do believe that runtime library dependencies should be concentrated in the
`runtime/server` module.
So, I do _not_ think bringing RDS as an `implementation` dependency into
JDBC persistence is justified. JDBC persistence code should not depend on
specific drivers, only on JDBC interfaces.
--
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]
