Matt Gibson

>Not to change the generated alg header (The decoding should not care about the given header anyway) I just ran into this when trying to use a Ruby-generated token in...

I was using [Joken](https://github.com/joken-elixir/joken), but the underlying low level library it uses is [JOSE](https://github.com/potatosalad/erlang-jose)

>If you are in control of the token generation you could change the constant controlling this locally: JWT::Algos::Eddsa::SUPPORTED = ["Ed25519"] Yes, I found that and it worked when generating the...

Ah, that's a good point. Possibly Joken is not implementing the RFC correctly either.

Ah, OK. Sounds like a bug, then. I stepped throught that code with the debugger and `association` already contains the existing 3 records at that point, so it adds the...

I think a docs section would be a great idea. Especially with example code. It's clearly not one-size-fits-all.

@rpechayr Thanks. I removed the Sinatra one as it seemed unnecessary alongside the new implementation and adding it back would mean rewriting all the tests for it in Minitest. This...

@rpechayr Ah, I see what you mean. Yes, that does make sense. Resque 2.0 seems to be in limbo from my reading of https://github.com/resque/resque/issues/1170 so no telling when it will...

@rpechayr Here's a first go at a gem: https://github.com/mattgibson/resque-scheduler-web There's a bit of test coverage missing, but all the old tests are there, and ported over to Rspec. I'm going...

@markets Thanks. Glad it's useful :) Yes, the consensus seems to be that this should not be in Resque Scheduler until Resque 2, but in the meantime, I have added...