> Well, I never asked you for query reconstruction abilities for shard-making > with jooq ;) I asked for listeners and filters. The examples of possible use > I provided are picked up randomly. There are plenty more.
But what I don't understand: Why do you need a listener of the type onExecute(), that will be executed, well, pretty much after you call execute() anyway? Can you please list a couple of interesting use cases, which you mentioned? > I also share your concern about automatic query reconstruction, but > shard-query seems to be a more or less mature project, which was tested by > Percona: > > http://www.mysqlperformanceblog.com/2011/05/06/scale-out-mysql/ > http://www.mysqlperformanceblog.com/2011/05/14/distributed-set-processing-with-shard-query/ > > And I really respect those guys and I don't think they would waste time to > test rubbish. I don't doubt that. However, I'm a bit spoiled with my Oracle 10g/11g cost based optimiser that does these things automatically for me. As soon as I partition a table, Oracle starts parallelising queries automatically (by transforming filters into unions, etc). I think that's the way it should be done.
