Soeren Roerden
Soeren Roerden
Well, the code should not deadlock anymore, but I don't think it is really reasonable to expect the API to save people from themselves in this case. The send will...
Well, hopefully the deadlock is actually fixed by my patch, so that concern might be addressed. If you are fine with messages not being sent on the sled side if...
Yeah, you are right that it is too easy to deadlock hat api as is. I didn't really look at all the places this was being called and on what...
I'm not sure if that seems reasonable to me. It compromises the semantics of the database for downstream consumers of events. What if you still want to keep adding values...
This is also an issue building folly, specifically during the boost discovery step
Unfortunately I don't really have time to minimize this and I've instead fixed it by updating to a newer boost version without this file
That null check actually seems to be necessary - should probably have been spelled out in their docs that they simply return a null instead of an empty collection in...
This performs several allocations for no real benefit. The var should be a val though, that was an oversight :)