Gaetan de Menten wrote: > It's not a question of whether it's the way I want it done or not. > It's just that this particular code will never find more targets than > the code above it, so it's just useless. I thought this was clear in > the log message...
Fair enough, it was a short log message, thanks for the clarification :) > Besides, as it is currently, the associable thing is not very useful > anyway since all it does is let you have one less table in case you > have more than two many to many relationships to a particular entity, > at the expense of slower relationships. That's a tradeoff I don't > understand. What's the problem with having a few more tables? So for > me, users currently using the associable thing, could just as well use > standard many to many relationships. Sure, I agree. But, users had code that they were perfectly happy with that broke. I just want to make sure we protect those people. Down the road we might consider deprecating this extension, or offering something else in its place, but for now, I think we should try to keep things working. The breakage wouldn't have happened if we had a test for the extension at the time, so thats really the lesson we learn, I think :) > Anyway, even with that said, I'll try to correct things as soon as > possible. Thanks, I saw you committed a fix, and our users will certainly appreciate it! > I'm not entirely happy with what I've done though. One problem is with > the extra methods attached to the entity. As you've seen from the > syntax I've proposed, I've now moved polymorphic relationships right > into the core. But I don't like to have some methods just appear in > that case. When using a specific extension, it's feels ok, but here, I > don't like it. Hmm, I see your point. Have you had any creative ideas on how we might get around this? Maybe providing the user the ability to state which methods to attach to their entities in the statement itself, to be more explicit? > Other than that I'm not entirely sure it's a good idea to > move all this stuff to the core. For: - it provides a nicer > syntax. Overloading the standard relationship feel natural. In an > extension, we'd have to use different statements than the "normal" > ones. Against: - it clutters the core. All good points here. I think what you have done is really quite good, actually, it just needs a little tweaking to make it work well. I'll take another look (and I would encourage other people to do so as well), and see if I can come up with anything to clean it up a bit. Thanks, Gaetan, for fixing this up again! -- Jonathan LaCour http://cleverdevil.org --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "SQLElixir" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sqlelixir?hl=en -~----------~----~----~----~------~----~------~--~---
