This commit introduces a comprehensive set of foundational improvements to make the API more robust, secure, and observable.
The following features have been implemented:
- **Observability Stack:** A new `internal/observability` package has been added, providing structured logging with `zerolog`, Prometheus metrics, and OpenTelemetry tracing. This stack is fully integrated into the application's request pipeline.
- **Centralized Authorization:** A new `internal/app/authz` service has been created to centralize authorization logic. This service is now used by the `user`, `work`, and `comment` services to protect all Create, Update, and Delete operations.
- **Standardized Input Validation:** The previous ad-hoc validation has been replaced with a more robust, struct-tag-based system using the `go-playground/validator` library. This has been applied to all GraphQL input models.
- **Structured Error Handling:** A new set of custom error types has been introduced in the `internal/domain` package. A custom `gqlgen` error presenter has been implemented to map these domain errors to structured GraphQL error responses with specific error codes.
- **`updateUser` Endpoint:** The `updateUser` mutation has been fully implemented as a proof of concept for the new patterns, including support for partial updates and comprehensive authorization checks.
- **Test Refactoring:** The test suite has been significantly improved by decoupling mock repositories from the shared `testutil` package, resolving circular dependency issues and making the tests more maintainable.
This commit introduces a comprehensive enhancement of the application's analytics features, addressing performance, data modeling, and feature set.
The key changes include:
- **Performance Improvement:** The analytics repository now uses a database "UPSERT" operation to increment counters, reducing two separate database calls (read and write) into a single, more efficient operation.
- **New Metrics:** The `WorkStats` and `TranslationStats` models have been enriched with new, calculated metrics:
- `ReadingTime`: An estimation of the time required to read the work or translation.
- `Complexity`: A score representing the linguistic complexity of the text.
- `Sentiment`: A score indicating the emotional tone of the text.
- **Service Refactoring:** The analytics service has been refactored to support the new metrics. It now includes methods to calculate and update these scores, leveraging the existing linguistics package for text analysis.
- **GraphQL API Expansion:** The new analytics fields (`readingTime`, `complexity`, `sentiment`) have been exposed through the GraphQL API by updating the `WorkStats` and `TranslationStats` types in the schema.
- **Validation and Testing:**
- GraphQL input validation has been centralized and improved by moving from ad-hoc checks to a consistent validation pattern in the GraphQL layer.
- The test suite has been significantly improved with the addition of new tests for the analytics service and the data access layer, ensuring the correctness and robustness of the new features. This includes fixing several bugs that were discovered during the development process.
