scrabble-game/gateway/internal/config/config.go

// Package config loads and validates the gateway's runtime configuration from
// the process environment. Every variable is prefixed GATEWAY_.
package config

import (
	"fmt"
	"os"
	"strconv"
	"time"

	pkgtel "scrabble/pkg/telemetry"
)

// Config holds the gateway's runtime configuration.
type Config struct {
	// HTTPAddr is the public Connect/h2c listener address (host:port). It also
	// serves the admin console at /_gm when admin credentials are configured.
	HTTPAddr string
	// LogLevel is the zap log level: "debug", "info", "warn" or "error".
	LogLevel string
	// BackendHTTPURL is the base URL of the backend REST API (gateway -> backend).
	BackendHTTPURL string
	// BackendGRPCAddr is the backend push gRPC address the gateway subscribes to.
	BackendGRPCAddr string
	// BackendTimeout bounds a single backend REST call.
	BackendTimeout time.Duration
	// AdminUser and AdminPassword are the Basic-Auth credentials the gateway
	// checks before proxying admin traffic to the backend. Empty disables admin.
	AdminUser     string
	AdminPassword string
	// ValidatorAddr is the gRPC address of the Telegram validator side-service (home,
	// plaintext, internal). The gateway calls it to validate Mini App initData and
	// Login Widget data. Empty disables the telegram auth path.
	ValidatorAddr string
	// BotLink configures the reverse mTLS channel to the remote Telegram bot. An
	// empty BotLink.Addr disables the bot channel (out-of-app push and admin relay).
	BotLink BotLinkConfig
	// SessionTTL bounds how long a resolved session stays cached; SessionCacheMax
	// caps the number of cached sessions.
	SessionTTL      time.Duration
	SessionCacheMax int
	// PushHeartbeatInterval is the idle keep-alive cadence on a client live stream.
	PushHeartbeatInterval time.Duration
	// MaxBodyBytes caps one inbound request body on the public listener and one
	// Connect message read; oversized requests are refused without buffering.
	MaxBodyBytes int
	// RateLimit configures the in-memory anti-abuse limiter.
	RateLimit RateLimitConfig
	// Abuse configures the temporary IP ban and the honeytoken (prod-only).
	Abuse AbuseConfig
	// Telemetry configures the OpenTelemetry providers (shared bootstrap).
	Telemetry pkgtel.Config
}

// BotLinkConfig configures the gateway's reverse bot-link: the mTLS listener the
// remote Telegram bot dials, the plaintext listener the backend admin relay calls,
// and the mTLS material. The main host is already public, so exposing the bot-link
// listener on a dedicated port adds no static IP; the channel is guarded solely by
// mTLS (the bot has no fixed address to allow-list).
type BotLinkConfig struct {
	// Addr is the mTLS gRPC listener the bot dials (e.g. ":9443"). Empty disables
	// the whole bot channel.
	Addr string
	// RelayAddr is the plaintext internal gRPC listener that serves the backend
	// admin SendToUser/SendToGameChannel relay (e.g. ":9092"). Empty disables it.
	RelayAddr string
	// CertFile, KeyFile and CAFile are the gateway server certificate, its key and
	// the CA bundle that signs the accepted bot client certificates. Required when
	// Addr is set.
	CertFile string
	KeyFile  string
	CAFile   string
	// SendTimeout bounds the admin relay's wait for the bot Ack before reporting
	// the send as not delivered.
	SendTimeout time.Duration
}

// RateLimitConfig holds the token-bucket limits per class. Public and admin are
// keyed per client IP; the authenticated class is keyed per user id; the email
// sub-limit guards the costly email-code path per IP.
type RateLimitConfig struct {
	PublicPerMinute int
	PublicBurst     int
	UserPerMinute   int
	UserBurst       int
	AdminPerMinute  int
	AdminBurst      int
	EmailPer10Min   int
	EmailBurst      int
}

// AbuseConfig configures the gateway's temporary IP ban (fail2ban-style) and the
// honeytoken trap. BanEnabled gates the ban action and is off by default: it is
// only safe where the real client IP is visible (i.e. in prod, not behind the
// shared-NAT test contour). Detection of honeypot/honeytoken hits is logged
// regardless of BanEnabled — only the ban action is gated.
type AbuseConfig struct {
	// BanEnabled turns the IP ban on. Off by default (prod-only).
	BanEnabled bool
	// BanThreshold is the rate-limiter rejection count within BanWindow that bans
	// a client IP.
	BanThreshold int
	// BanWindow is the rolling window the rejection strikes accumulate over.
	BanWindow time.Duration
	// BanDuration is the length of a rejection-earned ban (tripwire and honeytoken
	// bans use their own, longer, fixed durations).
	BanDuration time.Duration
	// Honeytoken, when non-empty, is a planted bearer value: presenting it bans the
	// caller and raises a high-severity alarm. Empty disables the trap.
	Honeytoken string
}

// Defaults applied when the corresponding environment variable is unset.
const (
	defaultAbuseBanThreshold     = 100
	defaultAbuseBanWindow        = 2 * time.Minute
	defaultAbuseBanDuration      = 15 * time.Minute
	defaultHTTPAddr              = ":8081"
	defaultLogLevel              = "info"
	defaultBackendHTTPURL        = "http://localhost:8080"
	defaultBackendGRPCAddr       = "localhost:9090"
	defaultBackendTimeout        = 5 * time.Second
	defaultSessionTTL            = 10 * time.Minute
	defaultSessionCacheMax       = 50000
	defaultPushHeartbeatInterval = 10 * time.Second // under the ~15 s edge idle timeout
	defaultServiceName           = "scrabble-gateway"
	defaultBotLinkSendTimeout    = 5 * time.Second
)

// DefaultMaxBodyBytes is the default request-body cap (GATEWAY_MAX_BODY_BYTES):
// 1 MiB — far above any legitimate edge payload (drafts and chat are a few KB)
// yet small enough to stop a cheap memory-amplification upload.
const DefaultMaxBodyBytes = 1 << 20

// DefaultRateLimit returns the built-in anti-abuse limits.
func DefaultRateLimit() RateLimitConfig {
	return RateLimitConfig{
		PublicPerMinute: 30, PublicBurst: 10,
		// Per-user (not per-IP): one user may run several devices, each holding a
		// Subscribe stream and reloading state on every live event, so the authenticated
		// budget is generous (a per-user cap cannot DoS the service). It is raised
		// because multi-device play tripped the old 120/40.
		UserPerMinute: 300, UserBurst: 80,
		AdminPerMinute: 60, AdminBurst: 20,
		EmailPer10Min: 5, EmailBurst: 2,
	}
}

// DefaultAbuse returns the built-in anti-abuse settings: the ban disabled
// (prod-only) with the agreed thresholds, and no honeytoken.
func DefaultAbuse() AbuseConfig {
	return AbuseConfig{
		BanEnabled:   false,
		BanThreshold: defaultAbuseBanThreshold,
		BanWindow:    defaultAbuseBanWindow,
		BanDuration:  defaultAbuseBanDuration,
	}
}

// Load reads the configuration from the environment, applies defaults, and
// validates the result.
func Load() (Config, error) {
	var err error
	c := Config{
		HTTPAddr:        envOr("GATEWAY_HTTP_ADDR", defaultHTTPAddr),
		LogLevel:        envOr("GATEWAY_LOG_LEVEL", defaultLogLevel),
		BackendHTTPURL:  envOr("GATEWAY_BACKEND_HTTP_URL", defaultBackendHTTPURL),
		BackendGRPCAddr: envOr("GATEWAY_BACKEND_GRPC_ADDR", defaultBackendGRPCAddr),
		AdminUser:       os.Getenv("GATEWAY_ADMIN_USER"),
		AdminPassword:   os.Getenv("GATEWAY_ADMIN_PASSWORD"),
		ValidatorAddr:   os.Getenv("GATEWAY_VALIDATOR_ADDR"),
		SessionCacheMax: defaultSessionCacheMax,
		RateLimit:       DefaultRateLimit(),
		Abuse:           DefaultAbuse(),
		BotLink: BotLinkConfig{
			Addr:      os.Getenv("GATEWAY_BOTLINK_ADDR"),
			RelayAddr: os.Getenv("GATEWAY_BOTLINK_RELAY_ADDR"),
			CertFile:  os.Getenv("GATEWAY_BOTLINK_TLS_CERT"),
			KeyFile:   os.Getenv("GATEWAY_BOTLINK_TLS_KEY"),
			CAFile:    os.Getenv("GATEWAY_BOTLINK_TLS_CA"),
		},
	}
	tel := pkgtel.DefaultConfig(defaultServiceName)
	tel.ServiceName = envOr("GATEWAY_SERVICE_NAME", tel.ServiceName)
	tel.TracesExporter = envOr("GATEWAY_OTEL_TRACES_EXPORTER", tel.TracesExporter)
	tel.MetricsExporter = envOr("GATEWAY_OTEL_METRICS_EXPORTER", tel.MetricsExporter)
	c.Telemetry = tel
	if c.BackendTimeout, err = envDuration("GATEWAY_BACKEND_TIMEOUT", defaultBackendTimeout); err != nil {
		return Config{}, err
	}
	if c.SessionTTL, err = envDuration("GATEWAY_SESSION_TTL", defaultSessionTTL); err != nil {
		return Config{}, err
	}
	if c.SessionCacheMax, err = envInt("GATEWAY_SESSION_CACHE_MAX", defaultSessionCacheMax); err != nil {
		return Config{}, err
	}
	if c.PushHeartbeatInterval, err = envDuration("GATEWAY_PUSH_HEARTBEAT_INTERVAL", defaultPushHeartbeatInterval); err != nil {
		return Config{}, err
	}
	if c.MaxBodyBytes, err = envInt("GATEWAY_MAX_BODY_BYTES", DefaultMaxBodyBytes); err != nil {
		return Config{}, err
	}
	c.Abuse.Honeytoken = os.Getenv("GATEWAY_HONEYTOKEN")
	if c.Abuse.BanEnabled, err = envBool("GATEWAY_ABUSE_BAN_ENABLED", c.Abuse.BanEnabled); err != nil {
		return Config{}, err
	}
	if c.Abuse.BanThreshold, err = envInt("GATEWAY_ABUSE_BAN_THRESHOLD", c.Abuse.BanThreshold); err != nil {
		return Config{}, err
	}
	if c.Abuse.BanWindow, err = envDuration("GATEWAY_ABUSE_BAN_WINDOW", c.Abuse.BanWindow); err != nil {
		return Config{}, err
	}
	if c.Abuse.BanDuration, err = envDuration("GATEWAY_ABUSE_BAN_DURATION", c.Abuse.BanDuration); err != nil {
		return Config{}, err
	}
	if c.BotLink.SendTimeout, err = envDuration("GATEWAY_BOTLINK_SEND_TIMEOUT", defaultBotLinkSendTimeout); err != nil {
		return Config{}, err
	}
	if err := c.validate(); err != nil {
		return Config{}, err
	}
	return c, nil
}

// BotLinkEnabled reports whether the reverse bot-link channel is configured.
func (c Config) BotLinkEnabled() bool { return c.BotLink.Addr != "" }

// AdminEnabled reports whether the admin console proxy should be mounted (both
// Basic-Auth credentials are configured).
func (c Config) AdminEnabled() bool {
	return c.AdminUser != "" && c.AdminPassword != ""
}

// validate reports whether the configuration values are acceptable.
func (c Config) validate() error {
	switch c.LogLevel {
	case "debug", "info", "warn", "error":
	default:
		return fmt.Errorf("config: invalid GATEWAY_LOG_LEVEL %q", c.LogLevel)
	}
	if c.HTTPAddr == "" {
		return fmt.Errorf("config: GATEWAY_HTTP_ADDR must not be empty")
	}
	if c.BackendHTTPURL == "" {
		return fmt.Errorf("config: GATEWAY_BACKEND_HTTP_URL must not be empty")
	}
	if c.BackendGRPCAddr == "" {
		return fmt.Errorf("config: GATEWAY_BACKEND_GRPC_ADDR must not be empty")
	}
	if c.MaxBodyBytes <= 0 {
		return fmt.Errorf("config: GATEWAY_MAX_BODY_BYTES must be positive")
	}
	if c.Abuse.BanEnabled && (c.Abuse.BanThreshold <= 0 || c.Abuse.BanWindow <= 0 || c.Abuse.BanDuration <= 0) {
		return fmt.Errorf("config: GATEWAY_ABUSE_BAN_THRESHOLD/_WINDOW/_DURATION must be positive when GATEWAY_ABUSE_BAN_ENABLED")
	}
	if c.BotLink.Addr != "" {
		if c.BotLink.CertFile == "" || c.BotLink.KeyFile == "" || c.BotLink.CAFile == "" {
			return fmt.Errorf("config: GATEWAY_BOTLINK_ADDR requires GATEWAY_BOTLINK_TLS_CERT, _KEY and _CA")
		}
	}
	if err := c.Telemetry.Validate(); err != nil {
		return fmt.Errorf("config: %w", err)
	}
	return nil
}

// envOr returns the value of the environment variable named key, or fallback
// when the variable is unset or empty.
func envOr(key, fallback string) string {
	if v := os.Getenv(key); v != "" {
		return v
	}
	return fallback
}

// envBool parses the environment variable named key as a bool, returning fallback
// when it is unset and an error when it is set but malformed.
func envBool(key string, fallback bool) (bool, error) {
	v := os.Getenv(key)
	if v == "" {
		return fallback, nil
	}
	b, err := strconv.ParseBool(v)
	if err != nil {
		return false, fmt.Errorf("config: %s: %w", key, err)
	}
	return b, nil
}

// envInt parses the environment variable named key as an int, returning fallback
// when it is unset and an error when it is set but malformed.
func envInt(key string, fallback int) (int, error) {
	v := os.Getenv(key)
	if v == "" {
		return fallback, nil
	}
	n, err := strconv.Atoi(v)
	if err != nil {
		return 0, fmt.Errorf("config: %s: %w", key, err)
	}
	return n, nil
}

// envDuration parses the environment variable named key as a Go duration,
// returning fallback when it is unset and an error when it is set but malformed.
func envDuration(key string, fallback time.Duration) (time.Duration, error) {
	v := os.Getenv(key)
	if v == "" {
		return fallback, nil
	}
	d, err := time.ParseDuration(v)
	if err != nil {
		return 0, fmt.Errorf("config: %s: %w", key, err)
	}
	return d, nil
}