Robert Mosolgo

Hmmm, sorry for the trouble! I'm torn because, on the one hand, you'd want to provide the nicest error, just like you've described here. But on the other hand, I...

Hi, thanks for the detailed write-up! I agree it should work like you expect. I wrote up a replication script and it _did_ work: ```ruby require "bundler/inline" gemfile do gem...

Hey, sorry this slipped off my radar 😞 Thanks for sharing those further details. I just now created a stand-alone replication based on your code above: input_object_test.rb ```ruby # mutations/add_inputs_nested.rb...

Ok, I did some debugging and found the cause for this strange issue. The underlying issue whether `modifies ...` is called _before_ or _after_ `mutation(...)` is configured. When `mutation(...)` is...

Ok, this sounds like an autoloading issue. What file is `Mutations::Inputs::AddInputsAttributes` in? I think Zeitwerk would check `app/graphql/mutations.rb`, `app/graphql/mutations/inputs.rb`, and `app/graphql/mutations/inputs/add_inputs_attributes.rb`, but if it isn't defined in one of those...

Ok, sounds good. Let know if you run into any more trouble on the GraphQL side of things! Sorry this order dependency wasn't documented before and thanks for helping me...

Hi! Sorry for my slow response. I certainly _expect_ it to work the way you describe. Here's a minimal example for modifying the arguments of a "Relay Classic" mutation with...

Oh... thanks for sharing these details. I bet there's some bug about how `loads: ...` converts the keyword argument from `something_id: ...` to `something: ...`. I'll take another look --...

Hey, thanks for the suggestion and starting this conversation. I'm definitely open to supporting something like this. At GitHub, we used GraphQL::Client to build some REST APIs and had mixed...

Hi! Yes, I don't think it freezes for _any_ operation that uses a Dataloader source, but maybe we can narrow down the problem and find out why it's not working...