- Add comprehensive test suite covering all Claude Explain functionality:
- Basic pane opening and consent flow
- no-ai directive detection
- API interactions and error handling
- Options/customization features
- Caching behavior and persistence
- Compilation state handling
- State persistence across page loads
- Fix caching bug in explain-view.ts:
- Cache was incorrectly implemented as instance variable, losing cached explanations when panes were closed/reopened
- Made cache static to persist across all pane instances (matches consent persistence pattern)
- Fixes failing "Caching and reload" Cypress test
- Aligns implementation with documented behavior: "shared across all explain views in the session"
- Add test utilities and helpers:
- Monaco editor content manipulation using clipboard events
- Claude Explain specific helpers moved to test file
- General utilities remain in utils.ts
- Performance optimizations:
- Clear Cypress intercepts in afterEach to prevent O(n²) degradation
- Use :visible selectors to avoid GoldenLayout template elements
- Proper mock setup timing to prevent race conditions
- Add comprehensive README with lessons learned and best practices
All tests use fake test data (test_first, focus_a, etc.) to clearly distinguish from production values and prevent accidental API calls.
Add Claude Explain feature for AI-powered code explanations
This PR introduces Claude Explain, a new feature that provides AI-powered explanations of compiler output directly within Compiler Explorer.
Key features:
Claude Explain functionality:
- New explain view pane
- Explains compiler output with full context of source code and compilation output
- Configurable audience level and explanation type
- Response caching to improve performance and reduce API calls
- Usage statistics display showing requests used and token counts
User experience:
- Consent flow on first use explaining data handling and privacy
- AI disclaimer banner warning about potential LLM inaccuracies
- Respects "no-ai" directive in source code for users who don't want AI processing
Privacy and security:
- Data sent to Anthropic's Claude API as documented in privacy policy
- No data used for model training
- Clear consent required before first use
- Support for opting out via "no-ai" directive
The feature is marked as beta and can be enabled via configuration.
Co-authored-by: Claude <noreply@anthropic.com>
