AtomPub

AtomPub, short for the Atom Publishing Protocol, is a content publishing protocol built on HTTP that enables clients to create, edit, and delete resources stored on a server. It uses the Atom Syndication Format as the encoding for the actual entries and is defined in RFC 5023 (2007). The protocol is designed to be RESTful, with standard HTTP methods and status codes governing resource state changes.

Discovery and structure: The server exposes a service document at a root or well-known URL that describes one or more workspaces and collections. Each collection corresponds to a publishable resource set (for example, a blog or a wiki) and includes an href to its REST endpoint. A collection can contain member entries, each represented as an Atom entry document.

Core operations: Clients retrieve a collection by GET to view existing entries, or create new ones by POSTing an Atom entry to the collection's endpoint. Upon success, the server typically responds with 201 Created and a Location header pointing to the new resource. Individual resources can be read with GET, updated with PUT (or PATCH, depending on server), and deleted with DELETE. Concurrency control is often achieved using ETags and If-Match headers.

Content and extensibility: The standard payload for new resources is an Atom entry document, although servers may support other encodings. AtomPub supports standard metadata via Atom elements, as well as extensions for categories and other metadata.

Usage and context: AtomPub found early adoption among weblogs and document management systems, providing a uniform way to publish and manage content over HTTP. In later years, JSON-based REST APIs and alternative publishing protocols gained popularity, and AtomPub remains a niche but interoperable option for certain platforms and services.