Under the hood

Processing flow

The Builder runs one of three pipelines, depending on the mode.

In spec mode, the Assembler reads the spec attributes into a Specification. The Resolver adds the classes it references, the augmenters enrich it, and a compiler for the target OpenAPI version writes the document. Spec pipeline architecture describes each step.

Hybrid mode scans with the classic Generator and its analyser, then hands the result to the same augmenters and compilers (Hybrid).

The classic pipeline

• The Generator iterates over the given sources (Symfony Finder, file/directory list, etc.)
• The configured analyser (AnalyserInterface) reads the files and builds an Analysis object. The default is the ReflectionAnalyser.
• The Analysis object and its annotation instances are then processed by the configured processors.
• If enabled, the annotation instances are validated.
• The root OpenApi instance then contains all annotation instances and is serialized into YAML/JSON.

Context

Each annotation instance is associated with a unique Context instance. This contains details, collected by the parser/analyser, about the PHP context where the attribute or annotation was found.

Typically, there will be a processor that uses the data to augment the annotation instance.

Examples of the data collected:

• class/interface/trait/enum names
• property names
• doctype or native type hints
• file name and line number

Analysis

Contains all detected annotation instances and other relevant metadata.

It uses a SplObjectStorage instance to store the annotation instances.