Yehuda Katz wrote: > > However, it feels like the BEST option would be to add a standard database > > trigger and let the database handle it without the potential for problems. > > Perhaps this is a case for supporting arbitrary options and letting the > > adapters generate the appropriate SQL. So: > > > has n, :posts, :counter_cache => true
So... has anyone tried this yet? If not, I'll see what I can do with it soon. On Apr 9, 12:39 pm, myobie <myo...@gmail.com> wrote: > I love the SQL trigger idea, that would be the best. Let the db > maintain integrity. > > I have yet to have any out of sync issues with after_... calls doing > counter cache like functionality, but I suppose it could happen. > > Nathan > > On Apr 6, 7:34 am, Yehuda Katz <wyc...@gmail.com> wrote: > > > It might be interesting to use Merb's run_later facilities to asynchronously > > update the counter cache on a periodic basis (or only if something was added > > to a queue). > > > However, it feels like the BEST option would be to add a standard database > > trigger and let the database handle it without the potential for problems. > > Perhaps this is a case for supporting arbitrary options and letting the > > adapters generate the appropriate SQL. So: > > > has n, :posts, :counter_cache => true > > > and automigrate would generate the appropriate SQL for the trigger for each > > supported DB. What do you think? > > > -- Yehuda > > > 2009/4/5 Jacques Crocker <merbj...@gmail.com> > > > > There's no reason to ever change a counter_cache after_save. You need > > > a combination of several hooks (after_create, after_delete, etc). > > > Having it abstracted seems greatly preferred. Even with these hooks, > > > the count would often go out of sync so I had an update_cache method > > > that would synchronize everything, running once in a while. And all > > > this logic duplicated for each has n relationship. > > > > Even with this plugin, I still run into the issue that the delete > > > method on an association isnt implemented in Datamapper 0.9.11. So > > > running "user.posts.delete" returns the message "NotImplementedError: > > > destroy *with* validations has not be written yet, try destroy!". So > > > its likely I'll still need these synchronization methods that keep the > > > counters in sync after certain operations. > > > > On Apr 5, 3:58 am, Yehuda Katz <wyc...@gmail.com> wrote: > > > > Why is a counter cache harder than a simple after_save hook. Having > > > > never > > > > actually attempted to roll my own, I'm curious about what specifically > > > > is > > > > difficult. > > > > > -- Yehuda > > > > > 2009/4/5 Jacques Crocker <merbj...@gmail.com> > > > > > > If you're in need of counter caches in Datamapper, I found a pretty > > > > > nice plugin that is working great for me so far. > > > > >http://github.com/saimonmoore/dm-counter-cache > > > > > > Maybe I'm crazy, but can't imagine anyone writing a moderate sized web > > > > > app and not needing counter caches. Any summary list page is going to > > > > > cause an N+1 craziness of count(*) queries. Rolling your own just a > > > > > complete mess (thats what I did before I found this plugin). > > > > > > We need to get this into dm-more. The plugin is well spec'ed. Maybe we > > > > > should tweak the interface, I'm not sure. Dan, can you take a look and > > > > > see if its to your liking? > > > > > > This is a followup to the thread: > > > > > >http://groups.google.com/group/datamapper/browse_thread/thread/95d21c. > > > .. > > > > > > Thanks > > > > > -- > > > > Yehuda Katz > > > > Developer | Engine Yard > > > > (ph) 718.877.1325 > > > -- > > Yehuda Katz > > Developer | Engine Yard > > (ph) 718.877.1325 --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "DataMapper" group. To post to this group, send email to datamapper@googlegroups.com To unsubscribe from this group, send email to datamapper+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/datamapper?hl=en -~----------~----~----~----~------~----~------~--~---
