Cucumber Open Architecture

Each Cucumber Open implementation has its own GitHub repository:

All implementations are built on top of several components that live in the cucumber/cucumber GitHub repository (this repository).

Cucumber Open Components

In order of importance / relevance to most contributors (hopefully):

gherkin - parser for Gherkin Documents (.feature files)
cucumber-expressions - parser for Cucumber's Regular Expressions alternative
messages - data structures for Cucumber Open's internal data model
datatable - library for manipulating data in a data table. Used in Cucumber step definitions.
- The Ruby and JavaScript implementations currently live in cucumber-ruby and cucumber-js respectively. They will stay there until the implementations have the same structure.
html-formatter - HTML report generator used by multiple Cucumber Open implementations
react - React components for rendering Gherkin Document and Cucumber results (represented as messages). Used by cucumber-html and https://reports.cucumber.io/
gherkin-utils - utilities for traversing, querying and pretty printing Gherkin document abstract syntax trees (ASTs).
query - utilities for linking results from Cucumber Open to the original source from Gherkin. This is useful because the decoupling of the Gherkin AST from Cucumber Results. Used by advanced formatter that output both the input document (Gherkin, Markdown) as well as the results.

fake-cucumber - a light Cucumber implementation written in TypeScript - used primarily for development and testing.