Adding new drivers about dungbeetle

[knadh]

Hi @josejuanmontiel. This is definitely interesting and the more database SQL jobber can support, the more useful it is. Do you have any ideas for an architectural change? I guess there's no way but to hardcode support for multiple DBs into the core. If only Go had a usable plugin system!

[josejuanmontiel]

Hi... i think we could talk the architectural here because all options incluide change and organice the code in some way to differenciate for specific cases and reuse where it't make sense... i'll try to open a PR in draft in the next days to see this changes in my case (oracle and csvq)... and talk about those...

[josejuanmontiel]

Hi! Sorry for the delay... the idea could be something like this... #21 still need some refactor... but this could be the idea to reuse the base "sql code" in another kind of tricky sql database.. what do you think?

[knadh]

Hi @josejuanmontiel. A little confused by the PR. This seems to have a new CSV backend too. Could you describe the PR in some detail?

[josejuanmontiel]

I commented... hope that understand... but, in resumen i decided implement new CSV backend... better than ifs ... to later import dinamically this code as plugin... and better readability i think :)