@vercel/queue 0.0.0-alpha.2 → 0.0.0-alpha.20

package/README.md

@@ -1,15 +1,16 @@
 # Vercel Queues
 
-A TypeScript client library for interacting with the Vercel Queue Service API with customizable serialization/deserialization (transport) support, including **streaming support** for memory-efficient processing of large payloads.
+A TypeScript client library for interacting with the Vercel Queue Service API, designed for seamless integration with Vercel deployments.
 
 ## Features
 
+- **Automatic Queue Triggering**: Vercel automatically triggers your API routes when messages are ready
+- **Next.js Integration**: Built-in support for Next.js API routes and Server Actions
 - **Generic Payload Support**: Send and receive any type of data with type safety
-- **Customizable Serialization**: Use built-in transports (JSON, Buffer, Stream) or create your own
-- **Streaming Support**: Handle large payloads without loading them entirely into memory
 - **Pub/Sub Pattern**: Topic-based messaging with consumer groups
 - **Type Safety**: Full TypeScript support with generic types
-- **Automatic Retries**: Built-in visibility timeout management
+- **Streaming Support**: Handle large payloads efficiently
+- **Customizable Serialization**: Use built-in transports (JSON, Buffer, Stream) or create your own
 
 ## Installation
 
@@ -17,71 +18,61 @@ A TypeScript client library for interacting with the Vercel Queue Service API wi
 npm install @vercel/queue
 ```
 
+The package includes:
+
+- **Main Library**: Queue client and utilities for production and development
+- **CLI Tool**: `npx vercel-queue-local-init` for local development handler initialization
+
 ## Quick Start
 
-For local development, you'll need to pull your Vercel environment variables (including the OIDC token):
+For local development, you'll need to set up your Vercel project:
 
 ```bash
 # Install Vercel CLI if you haven't already
 npm i -g vercel
 
+# Link your project to Vercel
+vc link
+
 # Pull environment variables from your Vercel project
 vc env pull
 ```
 
-Publishing and consuming messages on a queue
+## Local Development
 
-```typescript
-// index.ts
-import { QueueClient, createTopic, JsonTransport } from "@vercel/queue";
+**Queues just work locally.** After you have setup your Vercel project, when you `send()` messages in development mode, they automatically trigger your handlers locally - no external queue infrastructure needed.
 
-// Create a client - automatically authenticated using the OIDC token
-const client = QueueClient.fromVercelFunction();
+### Next.js Lazy Loading
 
-// Create a topic with JSON serialization (default)
-const topic = createTopic<{ message: string; timestamp: number }>(
-  client,
-  "my-topic",
-);
+For Next.js API routes (or others that are lazy-loaded), run this simple command to initialize handlers:
 
-// Publish a message
-await topic.publish({
-  message: "Hello, World!",
-  timestamp: Date.now(),
-});
-
-// Create a consumer group
-const consumer = topic.consumerGroup("my-processors");
+```bash
+npx vercel-queue-local-init
+```
 
-// Process messages continuously with cancellation support
-const controller = new AbortController();
+That's it! The script reads your `vercel.json`, finds your queue handlers, and triggers Next.js to load them.
 
-// Start processing (blocks until aborted or error)
-try {
-  await consumer.subscribe(controller.signal, async (message) => {
-    console.log("Received:", message.payload.message);
-    console.log("Timestamp:", new Date(message.payload.timestamp));
-  });
-} catch (error) {
-  console.error("Processing stopped due to error:", error);
-}
+### Example Workflow
 
-// Stop processing from elsewhere in your code
-// controller.abort();
-```
+```bash
+# Start your dev server
+npm run dev
 
-Run the script
+# Initialize handlers (only needed for frameworks that lazy load routes in dev)
+npx vercel-queue-local-init
 
-```bash
-# Using dotenv to load the OIDC token
-dotenv -e .env.local node index.ts
+# Send messages - they process locally automatically!
 ```
 
-## Usage with Vercel
+### CLI Options
 
-When deploying on Vercel, rather than having a persistent server subscribed to a queue, Vercel can trigger a callback route when a message is ready for consumption.
+```bash
+# Custom port
+npx vercel-queue-local-init --port 3001
 
-To demonstrate using queues on Vercel, let's use a Next.js app. You can use an existing app or create one using [create-next-app](https://nextjs.org/docs/app/api-reference/cli/create-next-app).
+# Different config file
+npx vercel-queue-local-init --config ./my-vercel.json
+```
 
 ### TypeScript Configuration
 
@@ -91,800 +82,293 @@ Update your `tsconfig.json` to use `"bundler"` module resolution for proper pack
 {
   "compilerOptions": {
     "moduleResolution": "bundler"
-    // ... other options
   }
 }
 ```
 
-### Publishing messages to a queue
+### Publishing Messages
 
-Create a new server function to publish messages
+The `send` function can be used anywhere in your codebase to publish messages to a queue:
 
 ```typescript
-// app/action.ts
-"use server";
-
-import { QueueClient, createTopic } from "@vercel/queue";
-
-export async function publishTestMessage(message: string) {
-  // Initialize a queue client
-  const client = await QueueClient.fromVercelFunction();
-
-  // Create a topic with JSON serialization (default)
-  const topic = createTopic<{ message: string; timestamp: number }>(
-    client,
-    "my-topic",
-  );
-
-  // Publish the message
-  const { messageId } = await topic.publish(
-    { message, timestamp: Date.now() },
-    {
-      // Provide a callback URL to invoke a consumer when the message is ready to be processed
-      callbacks: {
-        webhook: {
-          url: process.env.VERCEL_PROJECT_PRODUCTION_URL
-            ? `https://${process.env.VERCEL_PROJECT_PRODUCTION_URL}/api/queue/handle`
-            : "http://localhost:3000/api/queue/handle",
-        },
-      },
-    },
-  );
-
-  console.log(`Published message ${messageId}`);
-}
-```
+import { send } from "@vercel/queue";
 
-Now wire up the server function to your app
-
-```jsx
-// app/some/page.tsx
-"use client";
-import { publishTestMessage } from "./actions";
-
-export default function Button() {
-  return (
-    // ...
-    <Button onClick={() => publishTestMessage("Hello world")} >
-      Publish Test Message
-    </a>
-  );
-}
-```
-
-### Consuming the queue
-
-Instead of running a persistent server that subscribes to the queue, we use the callback functionality of Vercel queues to consume messages on the fly, when a message is ready to be processed.
-
-```typescript
-// app/api/queue/handle/route.ts
-import { QueueClient, Topic, parseCallbackRequest } from "@vercel/queue";
-import { NextRequest } from "next/server";
-
-// Handle Vercel Queue callback requests
-export async function POST(request: NextRequest) {
-  try {
-    // Parse the queue callback information
-    const { queueName, consumerGroup, messageId } =
-      parseCallbackRequest(request);
-
-    // Create client
-    const client = await QueueClient.fromVercelFunction();
-
-    // Create topic and consumer group from the callback info
-    const topic = new Topic(client, queueName);
-    const cg = topic.consumerGroup(consumerGroup);
-
-    // Process the message
-    await cg.receiveMessage(messageId, async (message) => {
-      const payload = message.payload as { message: string; timestamp: number };
-      console.log(
-        `Received message "${payload.message}" (Sent at: ${payload.timestamp})`,
-      );
-    });
-
-    return Response.json({ status: "success" });
-  } catch (error) {
-    console.error("Webhook error:", error);
-    return Response.json(
-      { error: "Failed to process webhook" },
-      { status: 500 },
-    );
-  }
-}
-```
-
-> ![NOTE]
-> A single webhook handle can be used to process messages across various queues and consumer groups. Use the values of `queueName` and `consumerGroup` from `parseCallbackRequest()` to dynamically handle different code paths:
->
-> ```typescript
-> if (queueName === "upload-queue") {
->   processImageQueue(consumerGroup, message);
-> }
-> // ...
-> function processImageQueue(consumerGroup, message) {
->   if (consumerGroup === "compress") {
->     // handle image compression
->   }
->   // ...
-> }
-> // ...
-> ```
->
-> We are building an SDK to make queues and workflow easier to use. Reach out if you're interested.
-
-## Key Features
-
-### Streaming Support
-
-Handle large files and data streams without loading them into memory:
-
-```typescript
-import { StreamTransport } from "@vercel/queue";
+// Send a message to a topic
+await send("my-topic", {
+  message: "Hello world",
+});
 
-const videoTopic = createTopic<ReadableStream<Uint8Array>>(
-  client,
-  "video-processing",
-  new StreamTransport(),
+// With additional options
+await send(
+  "my-topic",
+  {
+    message: "Hello world",
+  },
+  {
+    idempotencyKey: "unique-key", // Optional: prevent duplicate messages
+    retentionSeconds: 3600, // Optional: override retention time (defaults to 24 hours)
+  },
 );
-
-// Process large video files efficiently
-const processor = videoTopic.consumerGroup("processors");
-await processor.subscribe(signal, async (message) => {
-  const videoStream = message.payload;
-  // Process stream chunk by chunk
-  const reader = videoStream.getReader();
-  while (true) {
-    const { done, value } = await reader.read();
-    if (done) break;
-    await processChunk(value);
-  }
-});
 ```
 
-### Consumer Groups
-
-Multiple consumers can process messages from the same topic in parallel:
+Example usage in an API route:
 
 ```typescript
-// Multiple workers in the same group - they share/split messages
-const worker1 = topic.consumerGroup("workers");
-const worker2 = topic.consumerGroup("workers"); // Same group name
-// worker1 and worker2 will receive different messages (load balancing)
-
-// Different consumer groups - each gets copies of ALL messages
-const analytics = topic.consumerGroup("analytics");
-const webhooks = topic.consumerGroup("webhooks");
-// analytics and webhooks will both receive every message
-```
+// app/api/send-message/route.ts
+import { send } from "@vercel/queue";
 
-## Architecture
-
-- **Topics**: Named message channels with configurable serialization
-- **Consumer Groups**: Named groups of consumers that process messages in parallel
-  - `subscribe()`: Continuously process messages with automatic polling
-  - `receiveMessage()`: Process a specific message by ID
-  - `receiveNextMessage()`: Process the next available message (one-shot)
-  - `handleMessage()`: Process message metadata only (without payload)
-- **Transports**: Pluggable serialization/deserialization for different data types
-- **Streaming**: Memory-efficient processing of large payloads
-- **Visibility Timeouts**: Automatic message lifecycle management
+export async function POST(request: Request) {
+  const body = await request.json();
 
-## Performance
-
-The multipart parser is optimized for high-throughput scenarios:
-
-- **Streaming**: Messages are yielded immediately as headers are parsed
-- **Memory Efficient**: No buffering of complete payloads
-- **Fast Parsing**: Native Buffer operations for ~50% performance improvement
-- **Scalable**: Can handle arbitrarily large responses without memory constraints
-
-## Serialization (Transport) System
-
-The queue client supports customizable serialization through the `Transport` interface with **streaming support** for memory-efficient processing. Transport can be configured at the **topic level** when creating a topic, or at the **consumer group level** when creating a consumer group.
-
-### Built-in Transports
-
-#### JsonTransport (Default)
-
-Buffers data for JSON parsing - suitable for structured data that fits in memory.
-
-```typescript
-import { JsonTransport, createTopic } from "@vercel/queue";
+  const { messageId } = await send("my-topic", {
+    message: body.message,
+  });
 
-const topic = createTopic<{ data: any }>(
-  client,
-  "json-topic",
-  new JsonTransport(),
-);
-// or simply (JsonTransport is the default):
-const topic = createTopic<{ data: any }>(client, "json-topic");
+  return Response.json({ messageId });
+}
 ```
 
-#### BufferTransport
-
-Buffers the entire payload into memory as a Buffer - suitable for binary data that fits in memory.
-
-```typescript
-import { BufferTransport, createTopic } from "@vercel/queue";
+### Consuming Messages
 
-const topic = createTopic<Buffer>(
-  client,
-  "binary-topic",
-  new BufferTransport(),
-);
-const binaryData = Buffer.from("Binary data", "utf8");
-await topic.publish(binaryData);
-```
+Messages are consumed using API routes that Vercel automatically triggers when messages are available.
 
-#### StreamTransport
+#### 1. Create API Routes
 
-**True streaming support** - passes ReadableStream directly without buffering. Ideal for large files and memory-efficient processing.
+The recommended approach is to handle multiple topics and consumers in a single API route to keep your `vercel.json` configuration simple:
 
 ```typescript
-import { StreamTransport, createTopic } from "@vercel/queue";
+// app/api/queue/route.ts
+import { handleCallback } from "@vercel/queue";
+
+export const POST = handleCallback({
+  // Single topic with one consumer
+  "my-topic": {
+    "my-consumer": async (message, metadata) => {
+      // metadata includes: { messageId, deliveryCount, createdAt }
+      console.log("Processing message:", message);
+
+      // If this throws an error, the message will be automatically retried
+      await processMessage(message);
+    },
+  },
 
-const topic = createTopic<ReadableStream<Uint8Array>>(
-  client,
-  "streaming-topic",
-  new StreamTransport(),
-);
+  // Multiple consumers for different purposes
+  "order-events": {
+    fulfillment: async (order, metadata) => {
+      // By default, errors will trigger automatic retries
+      // But you can control retry timing if needed:
+      if (!isSystemReady()) {
+        // Override default retry with a 5 minute delay
+        return { timeoutSeconds: 300 };
+      }
 
-// Send large file as stream without loading into memory
-const fileStream = new ReadableStream<Uint8Array>({
-  start(controller) {
-    // Read file in chunks
-    for (const chunk of readFileInChunks("large-file.bin")) {
-      controller.enqueue(chunk);
-    }
-    controller.close();
+      await processOrder(order);
+    },
+    analytics: async (order, metadata) => {
+      try {
+        await trackOrder(order);
+      } catch (error) {
+        // Optional: Custom exponential backoff instead of default retry timing
+        const timeoutSeconds = Math.pow(2, metadata.deliveryCount) * 60;
+        return { timeoutSeconds };
+      }
+    },
   },
 });
-
-await topic.publish(fileStream);
 ```
 
-### Custom Transport
+While you can split handlers into separate routes if needed (e.g., for code organization or deployment flexibility), consolidating them in one route is recommended for simpler configuration.
 
-You can create your own serialization format by implementing the `Transport` interface:
+#### 2. Configure vercel.json
 
-```typescript
-import { Transport } from "@vercel/queue";
+Configure which topics and consumers your API route handles:
 
-interface Transport<T = unknown> {
-  serialize(value: T): Buffer | ReadableStream<Uint8Array>;
-  deserialize(stream: ReadableStream<Uint8Array>): Promise<T>;
-  contentType: string;
+```json
+{
+  "functions": {
+    "app/api/queue/route.ts": {
+      "experimentalTriggers": [
+        {
+          "type": "queue/v1beta",
+          "topic": "my-topic",
+          "consumer": "my-consumer",
+          "maxAttempts": 3,
+          "retryAfterSeconds": 60,
+          "initialDelaySeconds": 0
+        },
+        {
+          "type": "queue/v1beta",
+          "topic": "order-events",
+          "consumer": "fulfillment"
+        },
+        {
+          "type": "queue/v1beta",
+          "topic": "order-events",
+          "consumer": "analytics",
+          "maxAttempts": 5,
+          "retryAfterSeconds": 300
+        }
+      ]
+    }
+  }
 }
 ```
 
-### Choosing the Right Transport
-
-| Use Case               | Recommended Transport | Memory Usage | Performance |
-| ---------------------- | --------------------- | ------------ | ----------- |
-| Small JSON objects     | `JsonTransport`       | Low          | High        |
-| Binary files < 100MB   | `BufferTransport`     | Medium       | High        |
-| Large files > 100MB    | `StreamTransport`     | Very Low     | Medium      |
-| Real-time data streams | `StreamTransport`     | Very Low     | High        |
-| Custom protocols       | Custom implementation | Varies       | Varies      |
-
-## API Reference
-
-### QueueClient
-
-```typescript
-const client = new QueueClient({
-  token: string;
-  baseUrl?: string; // defaults to 'https://vqs.vercel.sh'
-});
-```
-
-### Topic
-
-```typescript
-const topic = createTopic<T>(client, topicName, transport?);
-
-// Publish a message (uses topic's transport)
-await topic.publish(payload, options?);
-
-// Create a consumer group (can override transport)
-const consumer = topic.consumerGroup<U>(groupName, options?);
-```
-
-### ConsumerGroup
-
-```typescript
-// Start continuous processing (blocks until signal is aborted or error occurs)
-await consumer.subscribe(signal, handler, options?);
+### Key Concepts
 
-// Process a specific message by ID
-await consumer.receiveMessage(messageId, handler);
+- **Topics**: Named message channels that can have multiple consumer groups
+- **Consumer Groups**: Named groups of consumers that process messages in parallel
+  - Different consumer groups for the same topic each get a copy of every message
+  - Multiple consumers in the same group share/split messages for load balancing
+- **Automatic Triggering**: Vercel triggers your API routes when messages are available
+- **Message Processing**: Your API routes receive message metadata via headers
+- **Configuration**: The `vercel.json` file tells Vercel which routes handle which topics/consumers
 
-// Process the next available message
-await consumer.receiveNextMessage(handler);
+## Advanced Features
 
-// Handle a specific message by ID without payload
-await consumer.handleMessage(messageId, handler);
-```
+### Serialization (Transport) System
 
-### Message Handler
+The queue client supports customizable serialization through the `Transport` interface:
 
-```typescript
-// Handler function signature
-type MessageHandler<T> = (
-  message: Message<T>,
-) => Promise<MessageHandlerResult> | MessageHandlerResult;
+#### Built-in Transports
 
-// Handler result types
-type MessageHandlerResult = void | MessageTimeoutResult;
+1. **JsonTransport (Default)**: For structured data that fits in memory
+2. **BufferTransport**: For binary data that fits in memory
+3. **StreamTransport**: For large files and memory-efficient processing
 
-interface MessageTimeoutResult {
-  timeoutSeconds: number; // seconds before message becomes available again
-}
-```
-
-### Transport Interface
+Example:
 
 ```typescript
-interface Transport<T = unknown> {
-  serialize(value: T): Buffer | ReadableStream<Uint8Array>;
-  deserialize(stream: ReadableStream<Uint8Array>): Promise<T>;
-  contentType: string;
-}
-```
-
-### Callback Utilities
+import { send, JsonTransport } from "@vercel/queue";
 
-```typescript
-// Parse queue callback request headers
-function parseCallbackRequest(request: Request): CallbackMessageOptions;
-
-// Callback options type
-interface CallbackMessageOptions {
-  queueName: string;
-  consumerGroup: string;
-  messageId: string;
-}
+// JsonTransport is the default
+await send("json-topic", { data: "example" });
 
-// Error thrown for invalid callback requests
-class InvalidCallbackError extends Error;
+// Explicit transport configuration
+await send(
+  "json-topic",
+  { data: "example" },
+  { transport: new JsonTransport() },
+);
 ```
 
-## Examples
-
-### Basic JSON Processing
-
-```typescript
-interface UserEvent {
-  userId: string;
-  action: string;
-  timestamp: number;
-}
-
-const userTopic = createTopic<UserEvent>(client, "user-events");
-
-await userTopic.publish({
-  userId: "123",
-  action: "login",
-  timestamp: Date.now(),
-});
-
-const consumer = userTopic.consumerGroup("processors");
-const controller = new AbortController();
-
-try {
-  await consumer.subscribe(controller.signal, async (message) => {
-    console.log(
-      `User ${message.payload.userId} performed ${message.payload.action}`,
-    );
-  });
-} catch (error) {
-  console.error("Processing error:", error);
-}
-
-// Stop processing when needed
-// controller.abort();
-```
+### Transport Selection Guide
 
-### Processing Specific Messages by ID
+| Use Case             | Recommended Transport | Memory Usage | Performance |
+| -------------------- | --------------------- | ------------ | ----------- |
+| Small JSON objects   | JsonTransport         | Low          | High        |
+| Binary files < 100MB | BufferTransport       | Medium       | High        |
+| Large files > 100MB  | StreamTransport       | Very Low     | Medium      |
+| Real-time streams    | StreamTransport       | Very Low     | High        |
 
-```typescript
-const userTopic = createTopic<{ userId: string; action: string }>(
-  client,
-  "user-events",
-);
-const consumer = userTopic.consumerGroup("processors");
+## Error Handling
 
-// Process a specific message if you know its ID
-const messageId = "01234567-89ab-cdef-0123-456789abcdef";
+The queue client provides specific error types:
 
-try {
-  await consumer.receiveMessage(messageId, async (message) => {
-    console.log(`Processing specific message: ${message.messageId}`);
-    console.log(
-      `User ${message.payload.userId} performed ${message.payload.action}`,
-    );
-  });
-  console.log("Message processed successfully");
-} catch (error) {
-  if (error.message.includes("not found or not available")) {
-    console.log("Message was already processed or does not exist");
-  } else if (error.message.includes("FIFO ordering violation")) {
-    console.log("FIFO queue requires processing messages in order");
-  } else {
-    console.error("Error processing message:", error);
-  }
-}
-```
+- **`QueueEmptyError`**: No messages available (204)
+- **`MessageLockedError`**: Message temporarily locked (423)
+- **`MessageNotFoundError`**: Message doesn't exist (404)
+- **`MessageNotAvailableError`**: Message exists but unavailable (409)
+- **`MessageCorruptedError`**: Message data corrupted
+- **`BadRequestError`**: Invalid parameters (400)
+- **`UnauthorizedError`**: Authentication failure (401)
+- **`ForbiddenError`**: Access denied (403)
+- **`InternalServerError`**: Server errors (500+)
 
-### Processing Next Available Message
+Example error handling:
 
 ```typescript
-const workTopic = createTopic<{ taskType: string; data: any }>(
-  client,
-  "work-queue",
-);
-const worker = workTopic.consumerGroup("workers");
+import {
+  BadRequestError,
+  ForbiddenError,
+  InternalServerError,
+  UnauthorizedError,
+} from "@vercel/queue";
 
-// Process the next available message (one-shot processing)
 try {
-  await worker.receiveNextMessage(async (message) => {
-    console.log(`Processing task: ${message.payload.taskType}`);
-    await processTask(message.payload.taskType, message.payload.data);
-  });
-  console.log("Message processed successfully");
+  await send("my-topic", payload);
 } catch (error) {
-  if (error instanceof QueueEmptyError) {
-    console.log("No messages available");
-  } else if (error instanceof MessageLockedError) {
-    console.log("Next message is locked (FIFO queue)");
-    if (error.retryAfter) {
-      console.log(`Retry after ${error.retryAfter} seconds`);
-    }
-  } else {
-    console.error("Error processing message:", error);
+  if (error instanceof UnauthorizedError) {
+    console.log("Invalid token - refresh authentication");
+  } else if (error instanceof ForbiddenError) {
+    console.log("Environment mismatch - check configuration");
+  } else if (error instanceof BadRequestError) {
+    console.log("Invalid parameters:", error.message);
+  } else if (error instanceof InternalServerError) {
+    console.log("Server error - retry with backoff");
   }
 }
-
-// You can also use it with timeout results
-await worker.receiveNextMessage(async (message) => {
-  if (!canProcessTaskType(message.payload.taskType)) {
-    // Return timeout to retry later
-    return { timeoutSeconds: 60 };
-  }
-
-  await processTask(message.payload.taskType, message.payload.data);
-});
 ```
 
-### Timing Out Messages
+## Advanced Usage
 
-```typescript
-const workTopic = createTopic<{ taskType: string; data: any }>(
-  client,
-  "work-queue",
-);
-const worker = workTopic.consumerGroup("workers");
-const controller = new AbortController();
+### Direct Message Processing
 
-try {
-  await worker.subscribe(controller.signal, async (message) => {
-    const { taskType, data } = message.payload;
-
-    // Check if we can process this task type right now
-    if (taskType === "heavy-computation" && isSystemOverloaded()) {
-      // Return timeout to retry later (5 minutes)
-      return { timeoutSeconds: 300 };
-    }
-
-    // Check if we have required resources
-    if (taskType === "external-api" && !isExternalServiceAvailable()) {
-      // Return timeout to retry in 1 minute
-      return { timeoutSeconds: 60 };
-    }
-
-    // Process the message normally
-    console.log(`Processing ${taskType} task`);
-    await processTask(taskType, data);
-    // Message will be automatically deleted on successful completion
-  });
-} catch (error) {
-  console.error("Worker processing error:", error);
-}
-
-// Example with exponential backoff
-const backoffController = new AbortController();
-
-try {
-  await worker.subscribe(backoffController.signal, async (message) => {
-    const maxRetries = 3;
-    const deliveryCount = message.deliveryCount;
-
-    try {
-      await processMessage(message.payload);
-      // Successful processing - message will be deleted
-    } catch (error) {
-      if (deliveryCount < maxRetries) {
-        // Exponential backoff: 2^deliveryCount minutes
-        const timeoutSeconds = Math.pow(2, deliveryCount) * 60;
-        console.log(
-          `Retrying message in ${timeoutSeconds} seconds (attempt ${deliveryCount})`,
-        );
-        return { timeoutSeconds: timeoutSeconds };
-      } else {
-        // Max retries reached, let the message fail and be deleted
-        console.error("Max retries reached, message will be discarded:", error);
-        throw error;
-      }
-    }
-  });
-} catch (error) {
-  console.error("Backoff processing error:", error);
-}
-```
-
-### Complete Example: Video Processing Pipeline
-
-Here's a comprehensive example showing a video processing pipeline that processes videos with FFmpeg and stores the results in Vercel Blob:
+> **Note**: The `receive` function is not intended for use in Vercel deployments. It's designed for use in the Vercel Sandbox environment or alternative server setups where you need direct message processing control.
 
 ```typescript
-import { QueueClient, createTopic, StreamTransport } from "@vercel/queue";
-import { spawn } from "child_process";
-import ffmpeg from "ffmpeg-static";
-import { put } from "@vercel/blob";
+// Process next available message
+await receive<T>(topicName, consumerGroup, handler);
 
-const client = new QueueClient({
-  token: "your-vercel-oidc-token",
+// Process specific message by ID
+await receive<T>(topicName, consumerGroup, handler, {
+  messageId: "message-id"
 });
 
-// Input topic with unoptimized videos
-const unoptimizedVideosTopic = createTopic<ReadableStream<Uint8Array>>(
-  client,
-  "unoptimized-videos",
-  new StreamTransport(),
-);
-
-// Output topic for optimized videos
-const optimizedVideosTopic = createTopic<ReadableStream<Uint8Array>>(
-  client,
-  "optimized-videos",
-  new StreamTransport(),
-);
-
-// Step 1: Process videos with FFmpeg
-const videoProcessor = unoptimizedVideosTopic.consumerGroup("processors");
-const processingController = new AbortController();
-
-try {
-  await videoProcessor.subscribe(
-    processingController.signal,
-    async (message) => {
-      const inputVideoStream = message.payload;
-      console.log("Processing video...");
-
-      if (!ffmpeg) {
-        throw new Error("FFmpeg not available");
-      }
-
-      // Create optimized video stream using FFmpeg
-      const optimizedStream = new ReadableStream<Uint8Array>({
-        start(controller) {
-          const ffmpegProcess = spawn(
-            ffmpeg,
-            [
-              "-i",
-              "pipe:0", // Input from stdin
-              "-c:v",
-              "libvpx-vp9", // Video codec
-              "-c:a",
-              "libopus", // Audio codec
-              "-crf",
-              "23", // Quality
-              "-f",
-              "webm", // Output format
-              "pipe:1", // Output to stdout
-            ],
-            { stdio: ["pipe", "pipe", "pipe"] },
-          );
-
-          // Pipe input stream to FFmpeg
-          const reader = inputVideoStream.getReader();
-          const pipeInput = async () => {
-            while (true) {
-              const { done, value } = await reader.read();
-              if (done) {
-                ffmpegProcess.stdin?.end();
-                break;
-              }
-              ffmpegProcess.stdin?.write(value);
-            }
-          };
-          pipeInput();
-
-          // Stream FFmpeg output
-          ffmpegProcess.stdout?.on("data", (chunk) => {
-            controller.enqueue(new Uint8Array(chunk));
-          });
-
-          ffmpegProcess.on("close", (code) => {
-            if (code === 0) {
-              controller.close();
-            } else {
-              controller.error(new Error(`FFmpeg failed with code ${code}`));
-            }
-          });
-        },
-      });
-
-      // Publish optimized video to next topic
-      await optimizedVideosTopic.publish(optimizedStream);
-      console.log("Video optimized and published");
-    },
-  );
-} catch (error) {
-  console.error("Video processing error:", error);
-}
-
-// Step 2: Store optimized videos in Vercel Blob
-const blobUploader = optimizedVideosTopic.consumerGroup("blob-uploaders");
-const uploadController = new AbortController();
+// Process message with options
+await receive<T>(topicName, consumerGroup, handler, {
+  messageId?: string; // Process specific message by ID
+  skipPayload?: boolean; // Skip payload download (requires messageId)
+  transport?: Transport<T>; // Custom transport (defaults to JsonTransport)
+  visibilityTimeoutSeconds?: number; // Message visibility timeout
+  refreshInterval?: number; // Refresh interval for long-running operations
+});
 
-try {
-  await blobUploader.subscribe(uploadController.signal, async (message) => {
-    const optimizedVideo = message.payload;
+// Handler function signature
+type MessageHandler<T = unknown> = (
+  message: T,
+  metadata: MessageMetadata
+) => Promise<MessageHandlerResult> | MessageHandlerResult;
 
-    // Upload to Vercel Blob storage
-    const filename = `optimized-${Date.now()}.webm`;
-    const blob = await put(filename, optimizedVideo, {
-      access: "public",
-      contentType: "video/webm",
-    });
+// Handler result types
+type MessageHandlerResult = void | MessageTimeoutResult;
 
-    console.log(`Video uploaded to blob: ${blob.url} (${blob.size} bytes)`);
-  });
-} catch (error) {
-  console.error("Blob upload error:", error);
+interface MessageTimeoutResult {
+  timeoutSeconds: number; // seconds before message becomes available again
 }
-
-// Graceful shutdown
-process.on("SIGINT", () => {
-  processingController.abort();
-  uploadController.abort();
-});
 ```
 
-## Error Handling
-
-The queue client provides specific error types for different failure scenarios:
-
-### Error Types
-
-- **`QueueEmptyError`**: Thrown when attempting to receive messages from an empty queue (204 status)
-
-  - Only thrown when directly using `client.receiveMessages()`
-  - `ConsumerGroup.subscribe()` handles this error internally and continues polling
-
-- **`MessageLockedError`**: Thrown when a message is temporarily locked (423 status)
+## Limits
 
-  - Contains optional `retryAfter` property with seconds to wait before retry
-  - For `receiveMessages()` on FIFO queues: the next message in sequence is locked
-  - For `receiveMessageById()`: the requested message is locked
-  - `ConsumerGroup.subscribe()` handles this error internally when polling
+- **Message Throughput**: Each topic can handle up to 1,000 messages per second
+- **Payload Size**: Maximum payload size is 4.5MB (this limit will be increased soon)
+- **Number of Topics**: No limit on the number of topics you can create
 
-- **`MessageNotFoundError`**: Message doesn't exist (404 status)
+### Scaling Beyond Limits
 
-- **`MessageNotAvailableError`**: Message exists but isn't available for processing (409 status)
+If you need more than 1,000 messages per second, you can create multiple topics (e.g., user-specific or shard-based topics) and handle them with a single consumer using wildcards in your `vercel.json`:
 
-- **`FifoOrderingViolationError`**: FIFO queue ordering violation (409 status with nextMessageId)
-
-  - Contains `nextMessageId` property indicating which message to process first
-
-- **`FailedDependencyError`**: FIFO ordering violation when receiving by ID (424 status)
-
-  - Contains `nextMessageId` property indicating which message must be processed first
-  - Similar to `FifoOrderingViolationError` but specifically for receive-by-ID operations
-
-- **`MessageCorruptedError`**: Message data is corrupted or can't be parsed
-
-- **`BadRequestError`**: Invalid request parameters (400 status)
-
-  - Invalid queue names, FIFO limit violations, missing required parameters
-
-- **`UnauthorizedError`**: Authentication failure (401 status)
-
-  - Missing or invalid authentication token
-
-- **`ForbiddenError`**: Access denied (403 status)
-
-  - Queue environment doesn't match token environment
-
-- **`InternalServerError`**: Server-side errors (500+ status codes)
-  - Unexpected server errors, service unavailable, etc.
-
-### Error Handling Examples
-
-```typescript
-import {
-  QueueEmptyError,
-  MessageLockedError,
-  FifoOrderingViolationError,
-  FailedDependencyError,
-  BadRequestError,
-  UnauthorizedError,
-  ForbiddenError,
-  InternalServerError,
-} from "@vercel/queue";
-
-// Handle empty queue or locked messages
-try {
-  for await (const message of client.receiveMessages(options, transport)) {
-    // Process messages
-  }
-} catch (error) {
-  if (error instanceof QueueEmptyError) {
-    console.log("Queue is empty, retry later");
-  } else if (error instanceof MessageLockedError) {
-    console.log("Next message in FIFO queue is locked");
-    if (error.retryAfter) {
-      console.log(`Retry after ${error.retryAfter} seconds`);
+```json
+{
+  "functions": {
+    "app/api/queue/route.ts": {
+      "experimentalTriggers": [
+        {
+          "type": "queue/v1beta",
+          "topic": "user-*",
+          "consumer": "processor"
+        }
+      ]
     }
   }
 }
+```
 
-// Handle locked message with retry
-try {
-  await consumer.receiveMessage(messageId, handler);
-} catch (error) {
-  if (error instanceof MessageLockedError) {
-    console.log("Message is locked by another consumer");
-    if (error.retryAfter) {
-      console.log(`Retry after ${error.retryAfter} seconds`);
-      setTimeout(() => retry(), error.retryAfter * 1000);
-    }
-  } else if (error instanceof FailedDependencyError) {
-    // FIFO ordering violation for receive by ID
-    console.log(`Must process ${error.nextMessageId} first`);
-  }
-}
+This allows you to:
 
-// Handle authentication and authorization errors
-try {
-  await topic.publish(payload);
-} catch (error) {
-  if (error instanceof UnauthorizedError) {
-    console.log("Invalid token - refresh authentication");
-  } else if (error instanceof ForbiddenError) {
-    console.log("Environment mismatch - check token/queue configuration");
-  } else if (error instanceof BadRequestError) {
-    console.log("Invalid parameters:", error.message);
-  } else if (error instanceof InternalServerError) {
-    console.log("Server error - retry with backoff");
-  }
-}
-
-// Complete error handling pattern
-function handleQueueError(error: unknown): void {
-  if (error instanceof QueueEmptyError || error instanceof MessageLockedError) {
-    // Transient errors - safe to retry
-    console.log("Temporary condition, will retry");
-  } else if (
-    error instanceof UnauthorizedError ||
-    error instanceof ForbiddenError
-  ) {
-    // Authentication/authorization errors - need to fix configuration
-    console.log("Auth error - check credentials");
-  } else if (error instanceof BadRequestError) {
-    // Client error - fix the request
-    console.log("Invalid request:", error.message);
-  } else if (error instanceof InternalServerError) {
-    // Server error - implement exponential backoff
-    console.log("Server error - retry with backoff");
-  } else {
-    // Unknown error
-    console.error("Unexpected error:", error);
-  }
-}
-```
+- Create topics like `user-1`, `user-2`, etc.
+- Process messages from all user topics with a single handler
+- Each topic gets its own 1,000 messages per second quota
 
 ## License