@yukihiro_matz @evanphx @headius is overriding thread-local variable methods (e.g. Thread#[]) completely crazy? https://github.com/mperham/sidekiq/issues/318 …
-
-
Replying to @bascule
@bascule@yukihiro_matz@headius Please don't. Please. If you want something else, give people a new mechanism. Don't change Thread#[].1 reply 0 retweets 0 likes -
Replying to @evanphx
@evanphx@yukihiro_matz@headius I strongly agree! However the problem is legacy code which uses thread locals, e.g. ActiveRecord1 reply 0 retweets 0 likes -
Replying to @bascule
@bascule@yukihiro_matz@headius Is it because Thread#[] respects Fibers?1 reply 0 retweets 0 likes -
Replying to @evanphx
@evanphx@yukihiro_matz@headius it's one reason yes. I'd like for thread locals to be "actor locals" inside actors1 reply 0 retweets 0 likes -
Replying to @bascule
@bascule@yukihiro_matz@headius So add Actor#[]. I am sad there is no way to access the non-Fiber aware locals.1 reply 0 retweets 0 likes -
Replying to @evanphx
@evanphx@yukihiro_matz@headius how do I deal with e.g. ActiveRecord that already uses Thread#[]?1 reply 0 retweets 0 likes -
Replying to @bascule
@bascule@yukihiro_matz@headius I'd rather you monkey patch AR.2 replies 0 retweets 0 likes -
Replying to @evanphx
@evanphx@yukihiro_matz@headius well, I've at least been talking to@tenderlove about fixing AR so it doesn't need thread locals1 reply 0 retweets 0 likes -
Replying to @bascule
@bascule@headius@evanphx@yukihiro_matz not sure we can ever get completely away from thread locals. e.g. http://thierryroussel.free.fr/java/books/martinfowler/www.martinfowler.com/isa/registry.html …2 replies 0 retweets 1 like
@tenderlove @headius @evanphx @yukihiro_matz persist it for the duration of a request with other request state instead of a thread local
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.