On Wed, Dec 19, 2018 at 10:29 AM Daniele Antonini <
[email protected]> wrote:

> Micro container approach with production data is a way, and I already use
> this technique (but microservices space have a lot of fog, so we should
> specify clearly our intent). Could be useful for reproduce bug and
> debugging, but if you want always reproducible non regression test I don't
> know if is the right way.
>

Sure, I know what you mean. It is difficult to get this right, but not
impossible. And I agree that once the meta model is known in a way that can
be easily reflected upon (like jOOQ's meta model), then it might be easier
to generate only a partial model and fill it with relevant data for the
code under test. Some sort of DSL might be helpful here. Just like the
MockFileDatabase, which can parse a specific query/result set
representation:

select 'A', 'B' union all 'C', 'D';
> A B
> - -
> A B
> C D
@ rows: 2

https://www.jooq.org/doc/latest/manual/sql-execution/mock-file-database/

That result set representation could be used to fill the relevant data in
the relevant tables. That would be the relational approach. Maybe a more
interesting approach would be to use something like GraphQL to set up test
data in a hierarchical way. I've been thinking about how GraphQL and jOOQ
could interact in the past.

-- 
You received this message because you are subscribed to the Google Groups "jOOQ 
User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to