application processors loading multiple times

[spemmons]

the ActiveMessaging::load_processors method has a "first" parameter that suppresses logging of the loading of application processor files, but loads them anyway

i noticed this because i have a constant defined in my processor class that i'm warned is being redefined

the problem seems to occur because the Railtie implementation calls "reload_activemesaging" which, in turn, reloads the processors -- later in my poller, we even explicitly load the processors a 3rd time

perhaps this is being done to overcome some other problem, but it seems like a simple change might fix eliminate the multiple processor loading

thoughts?

cheers!

[kookster]

I think some of the redundancy is vestigial to handling various ways Rails has re-done plugin/gem loading since 0.13 when some of that code was written.

I would absolutely agree - time to clean it up (probably not even a big change) and make it load once on Rails 3 for certain, and at least behave itself on Rails 2.3.x.

I'll take a look at that this afternoon - I am also writing a new threaded poller (inspired by sidekiq) which I'll be releasing soon, that I believe does do the loading singularly.

[iachievedit]

Any update on this particular issue? I think it is causing 'superclass mismatch for class EventProcessor' when testing with Webrick.

[kookster]

I am definitely not seeing that issue with loading classes, but no, we had
a few things going on, so I haven't had a chance to take this on.

Andrew Kuklewicz

On Fri, Jul 6, 2012 at 12:18 PM, iachievedit <
reply@reply.github.com

wrote:

Any update on this particular issue? I think it is causing 'superclass
mismatch for class EventProcessor' when testing with Webrick.

---

Reply to this email directly or view it on GitHub:
#14 (comment)