September 2017 Summaries

Announcing Speakers & Sessions for GraphQL Summit 2017` GraphQL Summit 2017 is just around the corner, taking place on October 25 and 26. The conference boasts an impressive lineup of speakers who will be sharing their expertise on GraphQL best practices, adoption stories, and experimental features. The speakers represent some of the largest companies using GraphQL today, including IBM, KLM, Twitter, Intuit, Capital One, and more, as well as maintainers of major services and open-source libraries. The sessions cover a range of topics, from authorization to caching, and even experimental features like live queries. Additionally, the conference is offering a GitHub Diversity Scholarship to make GraphQL Summit more accessible to underrepresented groups. Tickets are selling fast, so it's essential to reserve your seat now with a promo code or buddy pass promotion for discounts.

The Facebook team has granted full patent rights to all users of the GraphQL specification, licensing it under the Open Web Foundation Agreement (OWFa) v1.0. This change provides a clear and unambiguous grant of patent rights, allowing anyone to use GraphQL in any setting without worrying about copyright or patent issues. The GraphQL.js reference implementation will also be relicensed under the MIT license, which is a permissive and widely-used open source license. This move aims to empower as many people as possible to adopt GraphQL, making it easier for developers to build and contribute to GraphQL projects without legal concerns.

At Quri, a company that has been betting big on GraphQL since early 2016, they have adopted a development workflow called "GraphQL First" or "schema first" which involves drafting a schema proposal before starting any implementation. This approach allows front-end and back-end developers to work independently after the contract is settled. To implement this concept, Quri used a CLI tool called graphql-faker, which enables them to extend an existing schema by specifying its endpoint and writing new types in the GraphQL IDL Language. The process involves creating a pull request with proposed changes, getting feedback from both front-end and back-end teams, incorporating it, and merging it once consensus is reached. This workflow has improved their feedback loop, increased productivity, and reduced sequencing, allowing the front-end team to start implementing components without waiting for the back-end implementation to begin.

Apollo Mission Briefing is a new video series that aims to provide bite-sized, topical videos on GraphQL best practices and tips. The series is hosted by Peggy Rayzis, the newest member of the Apollo team, and covers topics such as connecting a GraphQL server to Apollo Client 2.0, customizing the cache and network interface with Apollo Link. The goal is to make it easier for developers to learn and level up their GraphQL skills in short, digestible videos that can be easily consumed during busy schedules. `</|im_start>

This tutorial covers the implementation of cursor-based pagination in a React + GraphQL application. The server's schema is updated to expose an API for loading smaller chunks of data, and the client-side code is modified to use this new API. A new resolver is added to fetch messages from the server using cursors, which allows the client to load only a page's worth of data at once. The Apollo Client cache is also updated to handle the new message feed structure. Finally, a `fetchMore` function is implemented to load older messages when the "Load Older Messages" button is clicked, demonstrating cursor-based pagination in practice.

Server-side rendering (SSR) in React with GraphQL enables faster load times by pre-rendering HTML on the server instead of waiting for client initialization. This is achieved through two phases: an asynchronous phase that fetches data and a synchronous phase to render markup. React Apollo, a package for managing GraphQL in React applications, achieves this through its tree walk algorithm, which crawls the component tree to find queries and resolves them before rendering the UI. The package also offers batching to reduce the number of HTTP calls to the API server. For incremental migration from existing SSR applications that use other data fetching styles, React Apollo can be used by abstracting data loading into higher-order components or using its `asyncAction` feature to adapt Flux data. This allows developers to leverage GraphQL without rewriting their entire application.

The latest version of Apollo Client, 2.0, brings a faster and more flexible experience while maintaining the API's familiarity. The new version introduces features such as modifiable request chains, custom data stores, and improved performance, with benchmarks showing speeds ranging from 2x to 5x faster than the previous version. The client is also smaller in size, reducing its footprint by around 40%. The updates enable users to create custom links for handling requests and caching, allowing for a wide range of new features and possibilities. With minimal changes required to upgrade, developers can take advantage of these improvements without significant rewrites.

The concept of schema stitching in GraphQL allows developers to merge multiple GraphQL schemas into one unified API, making it easier to access data from different sources. This enables a single request to fetch data from all the APIs involved, reducing the complexity and number of roundtrips required. The technology has already undergone active testing and is being refined through community feedback. It aims to provide a more seamless and efficient way to develop APIs, especially in scenarios where microservices or public APIs speak GraphQL. The schema stitching concept can help bridge the gap between different data sources and enable new use cases such as combining internal schemas with external ones, adding public APIs, or creating mashup APIs.