Message formats

Installation (Gradle)

Argo: compile group: "org.http4k", name: "http4k-format-argo", version: "2.35.1"

Gson: compile group: "org.http4k", name: "http4k-format-gson", version: "2.35.1"

Jackson: compile group: "org.http4k", name: "http4k-format-jackson", version: "2.35.1"

Xml: compile group: "org.http4k", name: "http4k-format-xml", version: "2.35.1"


These modules add the ability to use JSON/XML as a first-class citizen when reading from and to HTTP messages. Each implementation adds a set of standard methods and extension methods for converting common types into native JSON/XML objects, including custom Lens methods for each library so that JSON/XML node objects can be written and read directly from HTTP messages:

Auto-marshalling capabilities

Some of the message libraries (eg. GSON, Jackson, XML) provide the mechanism to automatically marshall data objects to/from JSON and XML using reflection.

We can use this facility in http4k to automatically marshall objects to/from HTTP message bodies using Lenses:



There is a utility to generate Kotlin code for JSON and XML data class code here . These data classes are compatible with using the Body.auto<T>() functionality.

FAQ (aka gotchas) regarding Auto-marshalling capabilities

Uppercase JSON key names

The Jackson version of the automarshalling does NOT work for objects with uppercase key names - this is a known issue with the Jackson library and we are unable to fix it. Either use the GSON alternative, or annotate your Data class fields with JsonAlias annotations to work around this.

JSON arrays

When handling raw JSON array messages, such as: [123, 456, 567], there is a slight gotcha when auto-marshalling messages from JSON.

This is demonstrated by the following, where you can see that the output of the auto-unmarshalling a naked JSON is NOT the same as a native Kotlin list of objects. This can make tests break as the unmarshalled list is NOT equal to the native list.

As shown, a workaround to this is to use Body.auto<Array<MyIntWrapper>>().toLens() instead, and then compare using Arrays.equal()