A shot of Absinthe: from zero to GraphQL in 40 minutes
summary
Elixir plus Absinthe runs as a production GraphQL server at Nested, powering an admin app that morphed into core business software. Schema-first GraphQL exposes query/mutation/subscription types, custom scalars, and resolvers that return {:ok, value} or {:error, reason}. Built-in tools handle N+1 with data loader and async resolution, limit abuse with complexity analysis, and emit metrics for deprecated-field usage. Playground and introspection via GraphiQL, bearer-token auth via middleware/context, WebSocket transport and caching caveats are demonstrated.
AI-written summary · about
chapters (29)
- 0:00Intro
- 0:38Absinthe
- 1:59Admin Panel
- 2:47Picking the right technologies
- 3:13What is GraphQL
- 3:45GraphQL vs REST
- 4:29GraphQL confusion
- 6:23GraphQL is not dead
- 6:46GraphQL is slow
- 7:57Recap
- 8:21Why GraphQL
- 9:56Benefits of GraphQL
- 10:11Restaurant analogy
- 12:40GraphQL like a buffet
- 13:32GraphQL Discoverability
- 15:22GraphQL Benefits
- 17:14Should I use GraphQL
- 19:26Fundamentals of GraphQL
- 19:45GraphQL Schema
- 21:57Scalar Types
- 23:32Types
- 24:42Query
- 26:47Use Absinthe
- 28:05Example schema
- 33:17Datadog
- 34:16Graphical
- 34:54Demo
- 39:22GraphQL Documentation
- 40:28GraphQL Authentication
related talks
38mNo REST for the Wicked. Building a GraphQL
35mGraphQL in the Enterprise
55mGraphQL - APIs the New Way
40mGraphQL: A Case Study
18mCrafting compelling real-time web experiences with GraphQL
32mFrom learning to thinking in GraphQL
59mGoodbye REST; Hello GraphQL
27mPower to the (Mobile) People: Clojure and GraphQL