Files
scrabble-game/backend/internal/pushgrpc/server.go
T
Ilia Denisov bf7dca0a09
CI / changes (pull_request) Successful in 1s
CI / unit (pull_request) Successful in 9s
CI / integration (pull_request) Successful in 11s
CI / ui (pull_request) Has been skipped
CI / gate (pull_request) Successful in 0s
CI / deploy (pull_request) Successful in 1m6s
Stage 17: fix the robot-nudge frequency + per-game push language
Two owner-reported defects from a live contour game.

A. Frequency: the robot's proactive nudge fired hourly for 12h+ (a 12h idle threshold
   then the 1h cooldown, uncapped). Replaced with a lengthening, randomized schedule
   (proactiveNudgeGap): the first nudge ~60-90 min into the human's turn, each later gap
   growing toward 1-6h (uniform sample in [60min, ceil], ceil ramping 90min->6h over 12h
   of idle, measured from the previous nudge), so a long wait gets a handful of
   increasingly-spaced reminders instead of a stream.

B. Language: out-of-app push routed by the recipient's GLOBAL service_language
   (last-login-wins), so after re-logging via the RU bot an English game's nudges came
   from the RU bot. Now a game push (your_turn, game_over, nudge, match_found) carries
   the game's own language (engine.Variant.Language) on push.Event, and the gateway
   routes by it (falling back to service_language for non-game pushes). The New-Game
   variant-gating guarantees the game's bot is one the player has started, so delivery is
   never blocked.

Tests: proactiveNudgeGap unit + retimed TestRobotProactiveNudge; TestVariantLanguage;
emit your_turn/game_over language; TestNudgeRoutedByGameLanguage integration. Docs:
ARCHITECTURE (§7 nudge, §10/§13 routing), FUNCTIONAL (+ _ru), PLAN tracker.
2026-06-09 08:06:58 +02:00

110 lines
3.1 KiB
Go

// Package pushgrpc serves the backend -> gateway live-event stream: a gRPC
// server exposing the scrabble.push.v1 Push service (docs/ARCHITECTURE.md §2).
// It bridges the in-process notify.Hub to the wire — each Subscribe stream
// drains a hub subscription and forwards every Intent as a push Event. The
// gateway opens one long-lived Subscribe at startup and fans the events out to
// its clients.
package pushgrpc
import (
"context"
"fmt"
"net"
"go.opentelemetry.io/contrib/instrumentation/google.golang.org/grpc/otelgrpc"
"go.uber.org/zap"
"google.golang.org/grpc"
"scrabble/backend/internal/notify"
pushv1 "scrabble/pkg/proto/push/v1"
)
// Service implements pushv1.PushServer over a notify.Hub.
type Service struct {
pushv1.UnimplementedPushServer
hub *notify.Hub
log *zap.Logger
}
// NewService constructs a Service that streams the hub's intents.
func NewService(hub *notify.Hub, log *zap.Logger) *Service {
if log == nil {
log = zap.NewNop()
}
return &Service{hub: hub, log: log}
}
// Subscribe opens a hub subscription and forwards every intent to the gateway
// until the stream's context ends (the gateway disconnected or the server is
// shutting down). It returns nil on a clean disconnect.
func (s *Service) Subscribe(req *pushv1.SubscribeRequest, stream grpc.ServerStreamingServer[pushv1.Event]) error {
ch, cancel := s.hub.Subscribe()
defer cancel()
s.log.Info("gateway push subscription opened", zap.String("gateway_id", req.GetGatewayId()))
defer s.log.Info("gateway push subscription closed", zap.String("gateway_id", req.GetGatewayId()))
ctx := stream.Context()
for {
select {
case <-ctx.Done():
return nil
case in, ok := <-ch:
if !ok {
return nil
}
ev := &pushv1.Event{
UserId: in.UserID.String(),
Kind: in.Kind,
Payload: in.Payload,
EventId: in.EventID,
Language: in.Language,
}
if err := stream.Send(ev); err != nil {
return err
}
}
}
}
// Server wraps the gRPC listener serving the Push service. Its Run mirrors the
// HTTP server's: serve until the context is cancelled, then stop gracefully.
type Server struct {
grpc *grpc.Server
addr string
log *zap.Logger
}
// NewServer builds a gRPC server bound to addr that streams hub events.
func NewServer(addr string, hub *notify.Hub, log *zap.Logger) *Server {
if log == nil {
log = zap.NewNop()
}
gs := grpc.NewServer(grpc.StatsHandler(otelgrpc.NewServerHandler()))
pushv1.RegisterPushServer(gs, NewService(hub, log))
return &Server{grpc: gs, addr: addr, log: log}
}
// Run starts the listener and blocks until ctx is cancelled, then stops the
// server gracefully. It returns the first error that is not a clean shutdown.
func (s *Server) Run(ctx context.Context) error {
lis, err := net.Listen("tcp", s.addr)
if err != nil {
return fmt.Errorf("pushgrpc: listen %s: %w", s.addr, err)
}
errc := make(chan error, 1)
go func() {
s.log.Info("push grpc listener starting", zap.String("addr", s.addr))
errc <- s.grpc.Serve(lis)
}()
select {
case err := <-errc:
return err
case <-ctx.Done():
s.log.Info("push grpc listener stopping")
s.grpc.GracefulStop()
return nil
}
}