1.0 blockers

[tomhoule]

This is a placeholder/WIP issue to gather what we need before a 1.0 release. The library has settled down a lot in the recent months and I think we can commit to stability in the GraphqlQuery trait and the generated code soon, with a bit more work. Feedback and ideas on things we have to do before 1.0 is very much welcome.

Features

• [ ] Cut down unnecessary prefixing
• [ ] Figure out whether the CLI should be considered 1.0 at the same time. I would be interested in feedback from people using it
• [ ] Migrate the web client to futures 0.3. We may want to switch it to an external library at the same time. My preference would be surf. Issue: https://github.com/graphql-rust/graphql-client/issues/272
• [ ] Improve the validation of the derive attributes
• [ ] Evaluate whether new features in the GraphQL spec (input unions...) would require breaking changes

Bugs

• [x] https://github.com/graphql-rust/graphql-client/issues/286
• [ ] https://github.com/graphql-rust/graphql-client/issues/250

Breaking changes

• Remove graphql-client-web (it is a feature of graphql-client now)

[mathstuf]

I'd recommend a milestone to help track things a bit more concretely :) .

[tomhoule]

@mathstuf and @h-michael, since you've been contributing a lot to the crate, if you are interested I can ask the owners of the graphql-rust organization to add you as maintainers — I can't do it on my own atm — to give you more flexibility (like editing the top message here and adding issues to the 1.0 milestone).

I hope I can spend a lot of time to polish graphql-client myself, but my track record hasn't been great^^