🧪 static-analysis-kit

A modular, extensible, and future-ready toolkit for defining, adapting, and sharing static and dynamic analysis rules across projects in the Manager monorepo.

🚀 What Is It?

@ovh-ux/manager-static-analysis-kit provides a unified framework for code quality automation across all applications, libraries, and modules — covering both static (lint, duplication, type coverage) and dynamic (tests, performance) checks.

It separates semantic rule definitions from tool-specific adapters to ensure configurability, reusability, and long-term consistency across the monorepo.

📦 Tools Exposed by the Kit

The kit currently provides:

• ESLint (flat config, adapters for React, TS, a11y, imports, naming, Storybook, Tailwind, TanStack, Vitest)
• Code Duplication Reports (manager-code-duplication)
• Performance Budgets (manager-perf-budgets)
• Tests Coverage Reports (manager-tests-coverage)
• Types Coverage Reports (manager-types-coverage)
• W3C & A11Y Unit Tests Matchers (expect(container).toBeAccessible(), await expect(html).toBeValidHtml())

Coming soon:

• Madge (dependency graph & circular analysis)
• Code Health Meter (complexity, Halstead, maintainability, SLOC)

Each tool has its own README inside the kit with full usage & configuration details.
See:

• Code Duplication
• Performance Budgets
• Tests Coverage
• Types Coverage
• HTML Validation
• Accessibility Validation

🧑‍💻 Combined Quality Checks

The kit also provides a single entrypoint for running all static & dynamic quality checks:

# Run all quality checks
yarn manager-static-dynamic-quality-checks

# Run only test-related checks
yarn manager-static-dynamic-quality-checks --tests

🔧 Launch Modes

Each CLI (e.g. manager-code-duplication, manager-types-coverage) supports flexible launch modes for targeting different scopes within the monorepo.

| Flag | Target | Path Resolution | |------|---------|----------------| | --apps / --app | Manager applications | manager/apps/* | | --packages / --package | Package-level analysis | Resolves via package.json | | --libraries / --library | Internal libraries | packages/manager/core, packages/manager/modules, packages/components |

Analyze specific apps (by folder name)

yarn manager-types-coverage --apps zimbra,container

or single app:

yarn manager-types-coverage --app zimbra

Here you provide app folders located under manager/apps.

Analyze specific packages (by package.json name)

yarn manager-types-coverage --packages @ovh-ux/manager-container-app,@ovh-ux/manager-zimbra-app

or single package:

yarn manager-types-coverage --package @ovh-ux/manager-container-app

In this mode:

• The CLI resolves the package names to their corresponding apps.
• Reports are generated per app folder.

Analyze specific libraries (by package.json name)

yarn manager-types-coverage --libraries @ovh-ux/manager-wiki,@ovh-ux/muk

or single library:

yarn manager-types-coverage --library @ovh-ux/muk

In this mode:

• The CLI resolves the library names to their corresponding folders inside configured library roots: > - packages/manager-react-components/*
  • packages/manager-wiki/*
• Reports are generated per library folder.

⚠️ Important Requirements

⚠️ Important:

These analysis tools only process React-based applications and libraries — Angular or legacy frameworks are not supported.

Each analysis type has specific requirements to run successfully:

| Analysis Type | Required Configuration | Description | |----------------|-------------------------|--------------| | Type Coverage | tsconfig.json | The tool analyzes only TypeScript projects with a valid configuration file. | | Tests Coverage | Vitest or Jest setup | Requires test files (*.test.ts[x], *.spec.ts[x]) and a working Vitest/Jest config. | | Performance Budgets | vite.config.ts or vite.config.js | Requires a Vite-based build configuration to extract bundle metrics. | | Code Duplication | React source files (.js, .ts, .tsx) | Automatically skips non-React or invalid modules. |

Modules that don't meet these conditions are skipped automatically with a warning, but the CLI will still analyze all valid React apps and libraries to ensure partial success in mixed environments.

🧱 Structure Overview

static-analysis-kit/
├── src/
│   ├── code-analysis.ts               # Central export point
│   ├── configs/                       # Tool-agnostic rule definitions
│   └── adapters/                      # Tool integrations
│       ├── eslint/                    
│       ├── code-duplication/
│       ├── perf-budgets/
│       ├── tests-coverage/
│       ├── types-coverage/
│       ├── ts-config/
│       ├── html-a11y-validation/
│       ├── html-w3c-validation/
│       └── architecture-analysis/      # Coming soon
├── bin/
│   ├── utils/                         # Shared CLI utilities
│   │   ├── args-parse-utils.js        # CLI flags parsing (apps/libs/packages)
│   │   ├── module-utils.js            # Discovery and resolution logic
│   │   ├── file-utils.js              # JSON read/write helpers
│   │   └── log-utils.js               # Colored CLI logging
│   └── cli-path-config.js             # Centralized base paths & root dirs
└── README.md

🧠 Design Principles

• Composable: Adapters are modular, rules are grouped by category.
• Decoupled: Semantic configs live separately from tool logic.
• Pluggable: Easy to add new tools (e.g., Biome, Oxlint).
• CI/CD Ready: All tools produce JSON + HTML reports for pipelines.

🧪 Validation & CLI Testing

Each CLI in the Static Analysis Kit (e.g., manager-code-duplication, manager-perf-budgets, manager-tests-coverage, manager-types-coverage) includes automated validation suites that ensure correct argument parsing, module discovery, and exit-code semantics.

These tests simulate real-world usage across apps, packages, and libraries — both valid and invalid — to guarantee stability and predictable CLI behavior.

To execute these tests, from the root run:

yarn static-dynamic-quality-check-tests

🧭 Test Matrix

| Category | Example Command | Expected Exit | |-----------|----------------|----------------| | Single app | yarn manager-code-duplication --app zimbra | ✅ 0 | | Multiple apps | yarn manager-code-duplication --apps container,zimbra | ✅ 0 | | Single package | yarn manager-code-duplication --package @ovh-ux/manager-container-app | ✅ 0 | | Multiple packages | yarn manager-code-duplication --packages @ovh-ux/manager-zimbra-app,@ovh-ux/manager-pci-workflow-app | ✅ 0 | | Single library (static) | yarn manager-code-duplication --library manager-react-components | ✅ 0 | | Single library (dynamic) | yarn manager-code-duplication --library shell-client | ✅ 0 | | Multiple libraries | yarn manager-code-duplication --libraries manager-wiki,manager-react-components,shell-client | ✅ 0 | | Mixed valid + invalid apps | yarn manager-code-duplication --apps zimbra,unknown-app,container | ✅ 0 | | All invalid apps | yarn manager-code-duplication --apps doesnotexist1,doesnotexist2 | ❌ 1 | | Turbo build simulation | yarn manager-code-duplication --app pci-ai-tools | ✅ 0 |

🧩 Each case asserts CLI correctness, exit codes, and resilience against unknown modules.

🧰 Supported CLIs

| CLI | Purpose | |-----|----------| | manager-code-duplication | Detects duplicated code across apps and libs | | manager-perf-budgets | Evaluates bundle and asset sizes | | manager-tests-coverage | Aggregates Jest/Vitest coverage reports | | manager-types-coverage | Reports TypeScript type coverage |

Each CLI supports:

• --apps / --app
• --packages / --package
• --libraries / --library

📬 Feedback & Contributions

The Static Analysis Kit aims to become the central standard for static and dynamic quality assurance across all OVHcloud Manager projects.

We welcome contributions that:

• Add new analyzers or integrations
• Improve performance or modularity
• Enhance developer experience or CI/CD compatibility

💡 Goal: A composable, maintainable, and organization-wide framework for modern code quality governance.

© 2025 OVHcloud — Manager Engineering Team