July 2020 Summaries

The article discusses the migration of a news publishing app from Django to Hasura and Next.js due to performance issues and increasing AWS costs. It highlights the benefits of using GraphQL with Hasura for better developer experience, improved performance, and cost savings on AWS. The author shares their experience in setting up the new stack and provides detailed metrics showing significant improvements in various aspects such as database CPU performance, application CPU performance, Apache Bench tests, Lighthouse speed scores, latency, and request counts.

The latest stable release of Hasura introduces Scheduled Triggers, which enable the execution of custom business logic via webhooks at specific points in time. This feature extends Event Triggers (which are driven by table events) to a flexible API that can be used to schedule events at any time for any payload and webhook. Scheduled Triggers come with two types of scheduling: Cron Triggers, which execute periodic events according to a cron schedule, and One-off Scheduled Events, which are used for one-time execution. The feature is well suited for serverless applications as each event can be mapped to an individual function that scales infinitely with no operations needed.

Hasura has released version 1.3.0 with significant features including Data Federation, Relay support and Scheduled Triggers. The new Data Federation feature enables a unified data layer for all your data sources through Remote Joins, allowing relationships across different data models in a single query. Relay support allows combining React and GraphQL to build highly modular and performant Javascript applications. Scheduled Triggers enable triggering webhooks based on a cron schedule or timestamp, opening up new use-cases like generating daily reports or sending post-signup emails. Other improvements include accessing session variables inside computed fields and automatic dependency cascade in the console.

A series of online events called "GraphQL Fireside Chats" is being organized by This Dot Media and will focus on various front-end frameworks. The first event, scheduled for July 28th, 2020, will discuss GraphQL + Flutter with special guests Chris Sells, Tanmai Gopal, and others. Subsequent events will cover topics such as Vue, Angular, and Relay. These chats aim to bring together the GraphQL community for discussion and cross-collaboration while maintaining safe social distancing practices.

This post discusses using Hasura Actions to convert a Node.js REST API to GraphQL. It explains that setting up a GraphQL server from scratch can be complex, but Hasura reduces this complexity by providing instant GraphQL CRUD for databases and allowing users to extend the Graph through different ways such as custom business logic or existing REST APIs. The post then delves into Actions, which are a way to extend Hasura's schema with custom business logic using custom queries and mutations. It provides examples of how to implement both new and existing REST APIs, handling permissions, and connecting to databases within the Node.js handler. Additionally, it mentions that the Hasura CLI can be used to generate action boilerplate code from the command line.

The article discusses the comparison between REST and GraphQL, two different approaches to data querying and updating. It explains that while both have their merits, they are not directly comparable as they serve different purposes. REST is an architectural style for designing distributed systems, whereas GraphQL is a language specification for defining, querying, and updating data. The article delves into the details of REST, including its constraints such as client-server pattern, statelessness, cacheability, uniform interface, unique resource identification, representation manipulation, self-descriptiveness, hypermedia as the engine of application state (HATEOAS), layered system, and code on demand. It also discusses how REST is implemented in practice and highlights that while some constraints are baked into HTTP, others need to be explicitly followed. The article then introduces GraphQL, a language developed by Facebook for defining data schemas, queries, and updates. It explains the advantages of using GraphQL, such as allowing clients to fetch only the exact data they need, enabling frontend teams to execute faster with less dependency on backend teams, and generating type-safe clients. However, it also notes that implementing resolvers in GraphQL can be challenging. The article concludes by stating that "GraphQL vs REST" is not a valid comparison as both serve different purposes. It suggests that architects should understand the requirements of their systems and choose technology accordingly to meet those requirements effectively.

AuthGuardian by OneGraph is a free service that enables users to add sign-on with various services like GitHub, Twitch, Stripe, Salesforce, and more. It also allows users to easily describe authentication and authorization rules for their app, API, or service. The service works natively with Hasura authorization out of the box. Users can set a user's X-Hasura-User-Id to their Spotify userId, add an admin default role if they're part of your Stripe organization, and set a session variable for discounts based on their GitHub contributions. AuthGuardian also supports remote schemas feature in Hasura, allowing users to query into Salesforce, Stripe, GitHub, and more.