Parsing with AMF

AMF can parse AML dialects and the following API specifications:

• RAML 0.8
• RAML 1.0
• OAS 2.0
• OAS 3.0
• AsyncAPI 2.0

The following section explains the basics of parsing, and the example shows how to parse each specification in Java or JavaScript.

Parsing Basics

Parsing is the process of analyzing a sequence of tokens and then building a data structure from them.

AMF uses SYAML to read a JSON or YAML file and generates an abstract syntax tree (AST) from the file, which is later used to generate a graph that represents the model. The model can be either a Web API model (representing an API, independent of its specification), or an AML document model (a dialect, vocabulary, or dialect instance).

Parsing returns a BaseUnit object, which is a graph of the model. Depending on the content in the parsed file, the BaseUnit object can be a Fragment, a Module, a Document or another type.

For more information about the AMF model, see the AMF model documentation resource.

The following figure shows the parsing process:

• SYAML is used to generate an AST from a YAML/JSON file
• AMF uses the AST to generate a graph of the model, called BaseUnit
  • This model will be used in the following stages of AMF (resolution, validation, emission)

Parsing code examples

The file paths you give the parser must have the following structure:

Windows:

• Absolute path: file:///C:/testing/api.raml
• Relative path: file://api.raml

MacOS and Unix:

• Absolute path: file:///Users/aml/testing/api.raml
• Relative path: file://api.raml

Also, before parsing an AML dialect instance, you must register the dialect using the AMF.registerDialect() function.

Java

Code extracted from the examples GitHub repository.

JavaScript

Code extracted from the examples GitHub repository.