AgentSkillsCN

supabase-realtime

提供全面的 Supabase 实时功能实施指南,涵盖最佳实践、可扩展模式以及迁移策略。当您在 Supabase 应用中构建实时功能,例如消息传递、通知、在线状态、实时更新、协作功能,或从 postgres_changes 迁移至 broadcast 时,可选用此指南。内容包括客户端设置、结合 realtime.broadcast_changes 的数据库触发器、RLS 授权机制、命名规范,以及性能优化技巧。

SKILL.md
--- frontmatter
name: supabase-realtime
description: Comprehensive guide for implementing Supabase Realtime features with best practices, scalable patterns, and migration strategies. Use when building realtime features in Supabase applications including messaging, notifications, presence, live updates, collaborative features, or migrating from postgres_changes to broadcast. Covers client setup, database triggers with realtime.broadcast_changes, RLS authorization, naming conventions, and performance optimization.

Supabase Realtime

Expert implementation guide for Supabase Realtime features focusing on scalable patterns and best practices.

Core Principles

Always use broadcast over postgres_changes - postgres_changes is single-threaded and doesn't scale. Use broadcast with database triggers for all database change notifications.

Use dedicated topics - Never broadcast to global topics. Use granular patterns like room:123:messages, user:456:notifications.

Private channels by default - Set private: true for all database-triggered channels. Enable private-only mode in production.

Quick Reference

Naming Conventions

  • Topics: scope:entity:id (e.g., room:123:messages)
  • Events: entity_action in snake_case (e.g., message_created)

Client Setup Pattern

javascript
const channel = supabase.channel('room:123:messages', {
  config: {
    broadcast: { self: true, ack: true },
    private: true  // Required for RLS
  }
})

// Set auth before subscribing
await supabase.realtime.setAuth()

channel
  .on('broadcast', { event: 'message_created' }, handler)
  .subscribe((status, err) => {
    if (status === 'SUBSCRIBED') console.log('Connected')
  })

// Cleanup
supabase.removeChannel(channel)

Database Trigger Pattern

sql
-- Use realtime.broadcast_changes for database events
CREATE OR REPLACE FUNCTION notify_table_changes()
RETURNS TRIGGER AS $$
BEGIN
  PERFORM realtime.broadcast_changes(
    TG_TABLE_NAME || ':' || COALESCE(NEW.id, OLD.id)::text,
    TG_OP,
    TG_OP,
    TG_TABLE_NAME,
    TG_TABLE_SCHEMA,
    NEW,
    OLD
  );
  RETURN COALESCE(NEW, OLD);
END;
$$ LANGUAGE plpgsql SECURITY DEFINER;

CREATE TRIGGER messages_broadcast_trigger
  AFTER INSERT OR UPDATE OR DELETE ON messages
  FOR EACH ROW EXECUTE FUNCTION notify_table_changes();

RLS Authorization

sql
-- Required for private channels
CREATE POLICY "room_members_can_read" ON realtime.messages
FOR SELECT TO authenticated
USING (
  topic LIKE 'room:%' AND
  EXISTS (
    SELECT 1 FROM room_members
    WHERE user_id = auth.uid()
    AND room_id = SPLIT_PART(topic, ':', 2)::uuid
  )
);

-- Required index for performance
CREATE INDEX idx_room_members_user_room ON room_members(user_id, room_id);

Implementation Checklist

  • Use broadcast not postgres_changes
  • Dedicated topics per room/user/entity
  • Set private: true for database triggers
  • Create indexes for all RLS policy columns
  • Include cleanup/unsubscribe logic
  • Check channel state before subscribing
  • Use consistent naming conventions

Use Cases for cv-optimizer

User Notifications

javascript
// Notify user when CV optimization is complete
const channel = supabase.channel(`user:${userId}:notifications`, {
  config: { private: true }
})

channel
  .on('broadcast', { event: 'optimization_complete' }, (payload) => {
    // Show notification to user
    toast.success('Your CV has been optimized!')
  })
  .subscribe()

Live Progress Updates

javascript
// Real-time progress during CV optimization
const channel = supabase.channel(`job:${jobId}:progress`, {
  config: { private: true }
})

channel
  .on('broadcast', { event: 'progress_update' }, (payload) => {
    setProgress(payload.percentage)
  })
  .subscribe()

Best Practices

  1. Always clean up channels when component unmounts
  2. Use private channels for user-specific data
  3. Index RLS policy columns for performance
  4. Prefer broadcast over postgres_changes for scalability
  5. Use granular topic naming for fine-grained subscriptions