Remove Calver type as the version scheme can be already represented using the Version type. Adopt ticket/milestone creation, Sagan versioning and Git tags to new scheme.
Use a dedicated class instead of a lambda expression because we cannot capture the entity type when using the latter which leads to false results when callbacks that should not be called get invoked due to their Object type signature.
closes: #146
The latest JGit upgrade runs tasks in parallel on the common fork-join pool. To avoid pool exhaustion and cyclic locks, we're offloading our work to an own pool.
We now sign commits if GPG properties are set. We pulled GPG properties from deployment.gpg to the top level (gpg).
For the deployment, we fall back to deployment.gpg if toplevel GPG properties aren't set.
Use StaticApplicationContext instead of AnnotationConfigApplicationContext for an empty application context. Use dedicated Person class instead of reusing it from an other benchmark class.
The build requires a specific version of Maven via enforcer plugin (3.5.0). If a developer does not have this specific version of Maven on their machine, they will be forced to install it, which is not too convenient. Having maven wrapper solves this problem
TODOs:
- Review interaction with the JDBC auto-configuration
- Wait for tcp protocol of H2 to be supported
- Wait for Postgres to accept credentialless connections (see [0])
[0] https://github.com/r2dbc/r2dbc-postgresql/issues/125
Add additional logging. Skip staging repository drop on Sonatype failures (e.g. when running into timeouts on repository close).
Fix Artifactory error parsing.
Unified benchmarks into one for JDBC and JPA. They're now both testing the data access with the plain infrastructure (JdbcTemplate and EntityManager) plus interactions via Spring Data repositories on top of that. Added build profiles to run the same benchmarks against an in-memory H2, a locally running H2 and a locally running Postgres. See the readme for setup instructions.
The JdbcFixture contains code to optionally disable event publication for the JDBC operation. Uncomment the line invoking disableEntityCallbacks(…).
Changed the benchmarks to use 10 warmup iterations instead of 5 to make sure we're at stable numbers once the measurement happens.
