> > Using sets with a technology that borrows from set theory kind of > makes sense, doesn't it :-)
One might think so ;-) Then again, they're bags, not really sets... > That said, I'm experienced enough not to pretend that I actually > understand the type system. Heh. Trust me. More experience won't necessary lead to more understanding in the case of SQL and its implementations. But it's eventually easier to figure out workarounds ;-) For example, I have a function that takes > VARCHAR(20)[] > as an input parameter. > I wonder if I could use SETOF VARCHAR(20) instead? You can't have SETOF arguments. In fact, Oracle seems to be the only database that supports table arguments to functions, although if you really want to nest them, the SQL gets really hairy. It would be extremely powerful for functions to act as set pipelines, but I guess we're not quite there yet. 2015-05-05 11:05 GMT+02:00 Ben Hood <[email protected]>: > On Tue, May 5, 2015 at 7:40 AM, Lukas Eder <[email protected]> wrote: > > If you have time to verify the fix already before the release, that > would be > > greatly appreciated of course. The change didn't incur any regressions in > > our integration tests, but there may still be edge-cases or function > return > > types that we haven't foreseen yet... > > What's the easiest way to grab this version and reference it from my > project? Do I clone the JOOQ upstream, check out that revision and > install it into my local .m2? Or are there some Maven nightly's > kicking around somewhere? This should be sufficient: $ git clone https://github.com/jOOQ/jOOQ.git $ mvn install Possibly with -Dmaven.test.skip=true in case the Open Source Edition's tests are currently failing... -- 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.
