Hi Sergey, In my experience, Spring Repositories configuration stays minimalistic, and one of two approaches is followed to create a schema in an underlying storage engine.
The first approach is when we can execute schema initialization scripts that will pre-create all SQL tables, indexes, etc. This approach works for a relational database as well as Ignite. Once the schema is set, we can start our Spring application with Repositories and other entities in place. An Ignite caches/tables can also be pre-established via CacheConfiguration with all required expiration policies and other parameters. The second method is to let Spring Data initialize the schema dynamically based on the configuration of your @Entity classes. That's probably what you're suggesting to do by adding custom configuration parameters to @Repositories. If we want to support this second dynamic approach, then we should do this via @Entity classes and not through @Repositories. It should be possible to introspect @Entity specific annotations and produce DDL commands or CacheConfigurations that will create Ignite caches/tables as soon as your Spring Data application starts. Let me know if I miss anything. - Denis On Wed, Jun 17, 2020 at 6:17 AM Sergey Moldachev <sergeymoldac...@gmail.com> wrote: > Hello, Igniters! > > I'd like to discuss with you changes for *ignite-spring-data*. You can read > about motivation and the idea in [1]. > > [1] https://issues.apache.org/jira/browse/IGNITE-13152 > <https://issues.apache.org/jira/browse/IGNITE-13152#> > > -- > Regards, > Sergey >
