June 2022 Summaries

This article discusses how to use Database-to-Remote Schema (DB-to-RS) Joins in GraphQL API. DB-to-RS Joins allow connecting data from an existing database with an external GraphQL service by creating a relationship between a field in the database and a field in the external GraphQL service. An example is provided where Hasura, a remote schema, is connected to a fulfillment service, another GraphQL API. The connection is established without altering the external GraphQL API or writing any code.

Hasura has released an alpha version of its open-source GraphQL Data Connectors to enable users to build data connectors for any data source and obtain instant GraphQL APIs. The release includes an API Agent, an SDK for development, and a reference application showcasing the process of creating new connections to varied data sources. Hasura currently supports Postgres, SQL Server, and BigQuery but plans to add support for more databases in the future. The Hasura GraphQL Data Connector acts as an intermediary abstraction between a data source and the Hasura GraphQL Engine. An SDK is provided to build, test, and deploy Data Connectors for various databases, while a reference connector demonstrates how to create a working connector example. The overall process of building a connector involves getting the SDK, modifying the reference implementation to meet specific database requirements, and finally deploying it.

The Hasura GraphQL Joins feature enables the creation of a unified GraphQL API by joining data from different GraphQL sources. It allows federating queries and mutations across multiple GraphQL services as if they were a single GraphQL schema without requiring extra code or changes to underlying APIs. There are four types of joins: Remote Schema to Remote Schema (RS-to-RS), Database to Database (DB-to-DB), Database to Remote Schema (DB-to-RS), and Remote Schema to Database (RS-to-DB). This article focuses on RS-to-RS Joins, which combine multiple GraphQL APIs via user-defined relationships. An example demonstrates joining a store service for handling orders with a fulfillment service for shipping using Hasura's GraphQL Joins feature without modifying the original APIs.

HasuraCon'22 announced the release of an SDK to extend Hasura to any data source, making GraphQL available on any datasource. The event also featured other product announcements such as Streaming Subscriptions for capturing large and fast-moving data, MS SQL Event Triggers for building event-driven workflows on SQL Server with native CDC, GitOps for GraphQL with automated CI/CD using Github Integration in GA on Hasura Cloud, enhanced observability support for OpenTelemetry Traces in Hasura Cloud & EE, and increased cloud-vendor availability with Hasura Enterprise on AWS Marketplace and Hasura on Google Cloud Platform.

The article discusses the use of Hasura's GraphQL Joins feature to join data from different sources. It specifically focuses on Remote Schema to Database (RS-to-DB) Joins, which allow for joining data coming from a GraphQL API with existing data in a database. This is achieved by creating a relationship between the remote schema and the Hasura app. The article provides an example of how this can be applied in an e-commerce scenario where a store service containing items and orders is connected to a Hasura application that tracks users and their orders. By creating an array relationship from the remote schema to the database, user information can be retrieved alongside the orders without writing additional code or modifying external GraphQL APIs.

DigitalOcean recently released their serverless solution, Functions. This article demonstrates how to integrate DigitalOcean Functions with Hasura for various use cases such as custom business logic via Hasura Actions and event logic via Hasura Event Triggers. The integration is shown through an example of triggering DigitalOcean Functions with Hasura Events by creating a new function, setting up a namespace, and configuring Hasura Event triggers to capture insert events on the users table and invoke the serverless DigitalOcean function. Potential use cases for this integration include sending push notifications and emails based on database events, transforming and loading data into external data-stores, and deploying static sites with content backed by Postgres. The possibilities are endless, and developers can explore more examples and ideas through Hasura's documentation or Discord server.

The third annual HasuraCon is set for June 28th-30th. This event brings together the global Hasura community to share product releases, keynotes, real-world use cases, workshops, and HASURAwards. Key highlights include new data backends for Hasura, keynote speeches on GraphQL and DevOps, real-world Hasura & GraphQL at scale use cases, specialized tracks for frontend & fullstack devs, insights into how Hasura is built, hands-on workshops, HASURAwards, sponsor talks, custom avatars, and opportunities to connect with the community.

This article presents the benefits and drawbacks of using an Authentication Provider and compares existing ones that support JWT integration. The pros include abstracting complexity, not storing user information, leveraging social logins, and better security. However, there are cons such as difficulty in migrating, provider lock-in, and costs. The article then evaluates seven authentication providers: Auth0, Firebase Auth, Clerk, KeyCloak, Cognito, SuperTokens, and Nhost based on six criteria including open source, custom JWT claims, single sign-on support, machine to machine authentication, free tiers, and ease of integration. The conclusion emphasizes that choosing an Authentication Service Provider depends on one's needs and requirements.

Hasura Cloud has announced the open beta of GCP Region Support. This feature allows users to deploy Hasura Cloud projects in regions where Google Cloud Platform (GCP) is available. The support was initially in private beta since December 2021 and has now been expanded for wider availability. Users with production databases and custom business logic hosted on GCP products can leverage this feature to reduce latency, improve connectivity, and enhance performance by co-locating their Hasura project. Currently, the supported regions include us-west-2 for the free tier and additional regions like us-central-1 and europe-west-3 for the paid tier. More regions will be added as they near general availability (GA). Users are encouraged to try out this feature and provide feedback.