2013/6/25 Durchholz, Joachim <[email protected]>

> > The plans being discussed here will allow for very
> > precise compile-time checking of whether a clause
> > / expression is supported by a set of dialects
> > (except when choosing API union). There is no way
> > to "try things" that will not work, unless I'm
> > misunderstanding something?
>
> Ad-hoc generation of the relevant subset of the API?
> Sure sounds cool, but also sounds scary.
> Also sounds like it's going to complicate the build process. Building is
> already a complicated thing in Java land; whatever the build activity, it
> needs to be easy to integrate into Ant, Maven, Gradle, SBT, and (hopefully)
> all the other cool build tools that will come along in the future. Oh, and
> it needs to integrate seamlessly with IDEs, and in my experience, that's
> the hardest subtask for getting the build under control - configuring
> Eclipse for an externally-run build step is clunky and not the kind of
> expertise that the average Eclipse user has.
>
> (I might be missing the point since I obviously missed that part of the
> discussion.)


I don't think we need to be scared, here.

1. The above wouldn't be necessary if adding jOOQ as a regular dependency.
jOOQ will continue to ship with the DSL API already available, which
is what most users will choose.
2. Those users who want to tweak the API / DSL will be ready for a one-shot
effort, integrating jOOQ as a source dependency into their build. I suspect
there will be Maven plugins to achieve building the required API parts,
or APT. In any case, they will probably do this once, and put the generated
sources and binary into a Maven repository, or something else.

jOOQ already has a code-generator for database objects. I don't think this
one will be any harder to integrate.

-- 
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/groups/opt_out.


Reply via email to