一首诗 schrieb: > That sounds reasonable. > > But, for your facebook example, there is another problem. If we > don't want to get these 4000 thousand friends, > we have 2 choices： > > 1. don't use relationship of SA to use the default encode mechanism of > turbojson > 2. write a customized "__json__" > > I think, the 1st choice is not acceptable in most cases, but the 2nd > one is also annoying when we have more than 40 tables. > > So, how about this : > > Find a way to mark fields of an object that should be skimmed when > encode it to JSON.
But the __json__-method is exactly such a spceification. And both the __json__-method as well as any declarative approach you suggest (which would be *very* hard to implement for us, because its's SQLAlchemy/SQLObject that we'd need to change) are IMHO to limited anyway. Because there might be occasions where I different json-representations for the same classes. So instead, I suggest you familiarize yourself with the simplejson API. There you can declare custom serializers, and if you want, write one that knows how to fully traverse an SA-object and it's relations. This could be very well a recipe in the docs - but I don't think it's justified to be included in the core, as the need for customization is very high so everybody will have to write json-encoders anyway. Diez --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "TurboGears" 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/turbogears?hl=en -~----------~----~----~----~------~----~------~--~---
