Martin, There are a few issues related to the generated IDs (for instance, the params fallback produces nil_class) which will be cleaned up in a future release (probably 1.1.0). We will still support the current (broken imho) behavior until 2.0.
-- Yehuda On Fri, Dec 12, 2008 at 6:42 AM, Martin Gamsjaeger <[email protected]>wrote: > > Hey guys, > > First off, thx for bringing up a somewhat interesting discussion! > However, I still hope that maybe someone also bothers to answer my > original question :-) > > Concerning your questions on the usefulness of namespaced models. > Well, modules are a powerful feature of ruby, and one possible use for > them is to keep namespaces clear and separated. That said, I just want > to do that, keep my namespaces separate and clear and prevent myself > from possible name clashes. Imagine using a merb slice that has a > model named the same like a model in the main app. Note that this > really doesn't mean that the underlying table _must_ be named > according to the conventions, or even that both tables have the same > name. As long as the models have the same name and/or the tables have > the same name, there is a problem. > > Now if all the slice models are namespaced, all I have to care is that > my other model is not in the same namespace/module (if it would need > to be, and it still has the same name, there is some kind of domain > model problem anyway). I must say that I also like the current > datamapper/extlib naming conventions, that produces a 'shop_products' > table out of a Shop::Product model. This takes the preventing of the > name clash down to the database layer in a scalable manner. I also > don't really see a point in modules mapping to repository names, how > would you handle multiple nestings then? (e.g. > Shop::Accounting::Address). > > I think my opinion on namespaced models is the following. It's > standard ruby to use modules also for namespacing reasons, and I think > that's a good practice! With that in mind, there is no reason I can > see why merb shouldn't allow me to write the ruby I like to write, > especially if the existing conventions cover (most things) nicely and > it's really not that hard! Furthermore, this very concept of > namespaced models can actively help in preventing nameclashes, and > that's a good thing imo. > > Again, anyone got an opinion on my original question concerning the > CSS3 id-selectors ? > > cheers > snusnu > > On Fri, Dec 12, 2008 at 11:08, jonuts <[email protected]> wrote: > > > > > > > > On Dec 12, 11:19 am, Roy Wright <[email protected]> wrote: > >> I wrote a rails app a while back that connected 5 legacy databases > >> (total 273 tables). There were a couple of tables with identical names > >> in different databases. I used a namespace for each database, which > >> worked well except for a couple of external tools I tried that just > >> didn't handle namespaces. If I was to do it again, I'd probably just > >> name my models by prepending the database name to the table name and > >> forgo namespaces. > >> > >> Just my 2 cents... > > > > Well, that was sort of my point. Putting a model under a module makes > > sense when dealing with multiple databases, but merb doesn't handle > > those situations as well as one (er, *I* at least) would expect. I > > just use your suggestion of prepending the db name to the model class > > name and that has worked fine. But I would like to know what "The" > > correct way of handling a situation like that is. > > > > > > > > > -- Yehuda Katz Developer | Engine Yard (ph) 718.877.1325 --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "merb" 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/merb?hl=en -~----------~----~----~----~------~----~------~--~---
