forked from aegis/pyserveX
- Added functionality to mark responses as cache hits to prevent incorrect X-Cache headers. - Introduced setCacheHitFlag function to traverse response writer wrappers and set cache hit flag. - Updated cachingResponseWriter to manage cache hit state and adjust X-Cache header accordingly. - Enhanced ProcessRequest and ProcessResponse methods to utilize new caching logic. feat(extension): Introduce ResponseWriterWrapper and ResponseFinalizer interfaces - Added ResponseWriterWrapper interface for extensions to wrap response writers. - Introduced ResponseFinalizer interface for finalizing responses after processing. refactor(manager): Improve response writer wrapping and finalization - Updated Manager.Handler to wrap response writers through all enabled extensions. - Implemented finalization of response writers after processing requests. test(caching): Add comprehensive integration tests for caching behavior - Created caching_test.go with tests for cache hit/miss, TTL expiration, pattern-based caching, and more. - Ensured that caching logic works correctly for various scenarios including query strings and error responses. test(routing): Add integration tests for routing behavior - Created routing_test.go with tests for route priority, case sensitivity, default routes, and return directives. - Verified that routing behaves as expected with multiple regex routes and named groups.
Konduktor (Go)
High-performance HTTP web server with extensible routing and process orchestration. (Previously known as PyServe in Python)
Project Structure
go/
├── cmd/
│ ├── konduktor/ # Main server binary
│ └── konduktorctl/ # CLI management tool
├── internal/
│ ├── config/ # Configuration management
│ ├── logging/ # Structured logging
│ ├── middleware/ # HTTP middleware
│ ├── routing/ # HTTP routing
│ ├── extensions/ # Extension system (TODO)
│ └── process/ # Process management (TODO)
├── pkg/ # Public packages (TODO)
├── go.mod
├── go.sum
└── Makefile
Building
cd go
# Download dependencies
make deps
# Build all binaries
make build
# Or build individually
make build-konduktor
make build-konduktorctl
Running
# Run with default config
./bin/konduktor
# Run with custom config
./bin/konduktor -c ../config.yaml
# Run with flags
./bin/konduktor --host 127.0.0.1 --port 3000 --debug
CLI Commands (konduktorctl)
# Start services
konduktorctl up
# Stop services
konduktorctl down
# View status
konduktorctl status
# View logs
konduktorctl logs -f
# Health check
konduktorctl health
# Scale services
konduktorctl scale api=3
# Configuration management
konduktorctl config show
konduktorctl config validate
# Initialize new project
konduktorctl init
Configuration
Uses the same YAML configuration format as the Python version:
server:
host: 0.0.0.0
port: 8080
http:
static_dir: ./static
templates_dir: ./templates
ssl:
enabled: false
cert_file: ./ssl/cert.pem
key_file: ./ssl/key.pem
logging:
level: INFO
console_output: true
extensions:
- type: routing
config:
regex_locations:
"=/health":
return: "200 OK"
Development
# Format code
make fmt
# Run linter
make lint
# Run tests
make test
# Run with coverage
make test-coverage
Migration from Python
This is a gradual rewrite of PyServe to Go. The project is now called Konduktor.
Completed
- Basic project structure
- Configuration loading
- HTTP server with graceful shutdown
- Basic routing
- Middleware (access log, recovery, server header)
- CLI structure (konduktor, konduktorctl)
TODO
- Extension system
- Regex routing
- Reverse proxy
- Process orchestration
- ASGI/WSGI adapter support
- WebSocket support
- Hot reload
- Metrics and monitoring