Hello, another annoying implementation/pattern question .
Lets say I have several read views for an application that can need to be updated, at least partially, by a certain event. Should I have one event handler that updates all the read tables or should have a several handlers, basically a one to one handler to view?
thanks,
r
Entirely up to you depending on how coupled you want your view generation code to be, whether you want them to be independently rebuildable, how (if) you want to enlist them in a transaction or whatever. I’d look at Projac from Yves as being a good starting
point if you want to look at this kind of thing.
James
Ah, Good point, it’s probably a pretty simple concept but not having a history with event sourcing this is just the sort of thing I would overlook. Rebuilding the view seems like the most important concern here. Just because one view is jacked you don’t want to have to rebuild everything.
This does sort make me concerned that there will be shit ton of handlers but that’s premature nit picking.
Thanks for the response.