Testing

A catalogue of the automated test suite, grouped by platform area. Each page covers one area, with a table per test class listing every test and what it verifies. The catalogue is a statement of behaviour as much as of coverage: a test named here is a guarantee the platform makes, so the tables double as a fine-grained functional specification of the areas they cover.

Each class entry states the fully qualified test class, the number of tests, and where applicable the test category and the xunit collection. A Kind of Fact is a single scenario; Theory is data-driven and runs once per supplied case, so the real execution count is higher than the number of rows.

The categories partition the suite by the infrastructure a test needs. Unit tests run against no external services. Service tests run against a real Postgres database and, where relevant, a real Redis cache. PenTest tests exercise the running HTTP surface adversarially. AskJooby tests cover the agent layer, and Volume tests exercise bulk data paths.

The catalogue is kept exact by the suite itself: TestDocumentationTests fails the build when a test exists without a row in these tables, when a documented test no longer exists, or when a class is documented on more than one page.

In this section

• Ask Jooby Agent
• Advanced Analytics Statistics
• Advanced Analytics Slices
• Advanced Analytics Engine and Services
• Exhaustive Machine Learning
• Analytics Foundations
• Charting and Model Diagrams
• Dictionary and Fuzzy Matching
• Dictionary Rule Extensions
• Real-Time Engine
• Rule Parsing and Compilation
• Cache and Redis
• Service Repositories
• Service Queries
• Service Model Configuration
• Service Exhaustive and Scaffolding
• Service Authentication and Users
• Service Monitoring and Observability
• Other Services
• Security Access Isolation
• Security Surface and Penetration Tests
• Data Access, Migrations and Archive
• Application and Infrastructure