FetchPHP is a modern HTTP client library for PHP, built on top of the Guzzle HTTP client, designed to mimic the behavior of JavaScript’s fetch API. Leveraging Matrix for true asynchronous capabilities with PHP Fibers, FetchPHP allows developers to use a JavaScript-like async/await syntax. FetchPHP also offers a fluent API inspired by Laravel's HTTP client, making request building both flexible and readable.

Whether you're building small APIs or large-scale systems with high concurrency needs, FetchPHP provides a powerful and efficient solution for managing HTTP requests in PHP.

Make sure to check out Matrix for more information on how FetchPHP is powered by PHP Fibers.

Full documentation can be found here

Why Choose FetchPHP Over Guzzle?

Guzzle is a well-established and widely-used HTTP client for PHP. It supports asynchronous requests using Promises. However, FetchPHP takes things further by offering true asynchronous task management through PHP Fibers, powered by Matrix. Here’s why FetchPHP stands out:

• True Async Task Management with Fibers: While Guzzle uses Promises for async operations, FetchPHP leverages PHP’s native Fibers (introduced in PHP 8.1) to provide true non-blocking concurrency. This gives developers fine-grained control over task execution, lifecycle management (e.g., pausing, resuming, retrying), and error handling.

• JavaScript-Like async/await Syntax: FetchPHP introduces a familiar async() syntax for developers who use JavaScript’s async/await functionality. This makes writing asynchronous PHP code more readable and intuitive.

• Fluent API: FetchPHP provides a fluent, chainable API similar to Laravel’s HTTP client, making constructing and managing HTTP requests easier. It’s more flexible and readable than Guzzle’s Promise-based API, which can feel rigid for managing complex tasks.

• Error Handling and Task Lifecycle Control: FetchPHP, powered by Matrix, allows for granular error management. Tasks can be paused, resumed, canceled, or retried dynamically, and errors can be handled through customizable handlers. Guzzle’s Promises manage errors in a less flexible way, usually through chained .then() and .catch() methods.

How FetchPHP's Async Task Management Differs from Guzzle

Here’s a breakdown of FetchPHP’s underlying async task management powered by Matrix compared to Guzzle’s Promise-based approach:

• Fiber-Based Concurrency: FetchPHP uses PHP Fibers to run tasks asynchronously. Fibers allow tasks to be paused, resumed, and canceled mid-execution, which isn’t possible with Guzzle’s Promises. This provides FetchPHP a true multi-tasking advantage.

• Task Lifecycle Management: FetchPHP allows you to start, pause, resume, cancel, and retry tasks using the Task class. Guzzle doesn’t offer built-in lifecycle management at this level. FetchPHP lets you track the status of a task (e.g., PENDING, RUNNING, PAUSED, COMPLETED, FAILED, CANCELED), giving more control over long-running or asynchronous processes.

• Custom Error Handling: FetchPHP offers a customizable ErrorHandler to manage retries, logging, and error resolution. This allows dynamic error handling and retrying tasks when needed, going beyond Guzzle’s Promises.

Note: The fetch() function allows for flexible HTTP request handling. When a URL is provided, it immediately sends the request. When no URL is provided, it returns a ClientHandler instance to enable further chaining for advanced request configuration.

Example: Managing Asynchronous Tasks with FetchPHP

<?php

use Fetch\Interfaces\Response as ResponseInterface;

$data = null;

// Asynchronously send a POST request using async/await syntax
async(fn () => fetch('https://example.com', [
    'method' => 'POST',
    'headers' => ['Content-Type' => 'application/json'],
    'body' => json_encode(['key' => 'value']),
]))
    ->then(fn (ResponseInterface $response) => $data = $response->json())  // Success handler
    ->catch(fn (\Throwable $e) => echo "Error: " . $e->getMessage());      // Error handler

// The async operation is managed with start, pause, resume, and cancel controls
$task = async(fn () => fetch('https://example.com', [
    'method' => 'POST',
    'headers' => ['Content-Type' => 'application/json'],
    'body' => json_encode(['key' => 'value']),
]));

// Manually control the task lifecycle as `then` and `catch` will automatically start the task
$task->start();

// Pause the task if needed
$task->pause();

// Resume the task
$task->resume();

// Cancel the task if required
$task->cancel();

// Retry the task if it fails
if ($task->getStatus() === TaskStatus::FAILED) {
    $task->retry();
}

// Get the result only if the task is completed successfully
$result = $task->getResult();

Lifecycle Control Example with FetchPHP

<?php

use Matrix\Task;
use Matrix\Enum\TaskStatus;

// Define a long-running task
$task = new Task(function () {
    return "Task completed!";
});

// Start the task
$task->start();

// Pause and resume the task dynamically
$task->pause();
$task->resume();

// Cancel the task if needed
$task->cancel();

// Retry the task if it fails
if ($task->getStatus() === TaskStatus::FAILED) {
    $task->retry();
}

$result = $task->getResult();

Why FetchPHP is Better for Asynchronous PHP

While Guzzle is a fantastic tool for making HTTP requests, FetchPHP brings modern PHP capabilities with PHP 8 Fibers, making it ideal for developers who need true asynchronous task management with a JavaScript-like syntax. FetchPHP is designed to make your code more flexible, readable, and efficient when managing complex HTTP operations, especially when concurrency and non-blocking I/O are crucial.

Installation

To install FetchPHP, run the following command:

composer require jerome/fetch-php

FetchPHP requires PHP 8.1 or above due to its use of Fibers for async tasks.

Core Features

• JavaScript-like fetch API for HTTP requests (synchronous and asynchronous).
• Fluent API for building requests with a readable and chainable syntax.
• Asynchronous support with PHP Fibers via the Matrix package.
• Built on Guzzle for robust and reliable HTTP client functionality.

Usage Examples

JavaScript-like Fetch API (Synchronous)

<?php

$response = fetch('https://example.com', [
    'method' => 'POST',
    'headers' => [
        'Content-Type' => 'application/json',
    ],
    'body' => json_encode(['key' => 'value']),
]);

$data = $response->json();

JavaScript-like Fetch API (Asynchronous)

<?php

use Fetch\Interfaces\Response as ResponseInterface;

$data = null;

// Asynchronously send a POST request using async/await syntax
async(fn () => fetch('https://example.com', [
    'method' => 'POST',
    'headers' => ['Content-Type' => 'application/json'],
    'body' => json_encode(['key' => 'value']),
]))
    ->then(fn (ResponseInterface $response) => $data = $response->json())  // Success handler
    ->catch(fn (\Throwable $e) => echo "Error: " . $e->getMessage());      // Error handler

Using the Fluent API

FetchPHP’s fluent API provides the following methods for building requests:

• withToken(): Attach a Bearer token to the request.
• withAuth(): Attach basic authentication credentials.
• withHeaders(): Add headers to the request.
• withBody(): Add a request body (e.g., JSON, form data).
• withProxy(): Specify a proxy server for the request.
• withCookies(): Attach cookies to the request.
• withQueryParameters(): Add query parameters to the request URL.
• timeout(): Set the timeout for the request.
• retry(): Set the number of retries and delay for failed requests.

Synchronous Example

<?php

$response = fetch()
    ->baseUri('https://example.com')
    ->withHeaders(['Content-Type' => 'application/json'])
    ->withBody(['key' => 'value'])
    ->withToken('fake-bearer-auth-token')
    ->post('/posts');

$data = $response->json();

Asynchronous Example

<?php

use Fetch\Interfaces\Response as ResponseInterface;

$data = null;

// Asynchronously send a POST request using the fluent API
async(fn () => fetch()
    ->baseUri('https://example.com')
    ->withHeaders(['Content-Type' => 'application/json'])
    ->withBody(['key' => 'value'])
    ->withToken('fake-bearer-auth-token')
    ->post('/posts'))
    ->then(fn (ResponseInterface $response) => $data = $response->json())  // Success handler
    ->catch(fn (\Throwable $e) => echo "Error: " . $e->getMessage());      // Error handler

Using the ClientHandler Class

The ClientHandler class is responsible for managing HTTP requests, including synchronous and asynchronous handling. You can use it directly for more advanced use cases:

Basic Example with ClientHandler

<?php

use Fetch\Http\ClientHandler;

$response = ClientHandler::handle('GET', 'https://example.com', [
    'headers' => ['Accept' => 'application/json']
]);

$data = $response->json();

Asynchronous Example with ClientHandler

<?php

use Fetch\Http\ClientHandler;

$data = null;

// Asynchronously manage a request using the ClientHandler
async(fn () => ClientHandler::handle('POST', 'https://example.com', [
    'headers' => ['Content-Type' => 'application/json'],
    'body' => json_encode(['key' => 'value']),
]))
    ->then(fn ($response) => $data = $response->json())
    ->catch(fn ($e) => echo "Error: " . $e->getMessage());

Request Options

FetchPHP accepts an array of options to configure requests:

• method: HTTP method (e.g., 'GET', 'POST', etc.). Default is 'GET'.
• headers: Array of HTTP headers.
• body: Request body for methods like POST, PUT, PATCH.
• json: JSON data to send as the request body.
• timeout: Timeout for the request in seconds.
• auth: Array for HTTP Basic or Digest authentication.
• proxy: Proxy server URL for routing requests.
• client: A custom Guzzle client instance.

Error Handling

Both synchronous and asynchronous requests handle errors gracefully. Here's how you can manage errors:

Synchronous Error Handling

<?php

$response = fetch('https://nonexistent-url.com');

if ($response->ok()) {
    echo $response->json();
} else {
    echo "Error: " . $response->statusText();
}

Asynchronous Error Handling

<?php

$response = async(fn () => fetch('https://nonexistent-url.com'))
    ->then(fn ($response) => $response->json())
    ->catch(fn ($e) => echo "Error: " . $e->getMessage());

echo $response;

Advanced Error Handling: Retry with Exponential Backoff

<?php

$response = async(fn () => fetch('https://api.example.com/resource'))
    ->then(fn ($response) => $response->json())
    ->catch(function (\Throwable $e) {
        // Implement retry logic with exponential backoff
        static $attempt = 1;
        if ($attempt <= 3) {
            sleep(pow(2, $attempt));  // Exponential backoff
            $attempt++;
            return retryRequest();  // Custom function to retry
        }
        return "Error: " . $e->getMessage();
    });

echo $response;

Proxy and Authentication Support

FetchPHP supports proxies and authentication out of the box:

Proxy Example

<?php

$response = fetch('https://example.com', [
    'proxy' => 'tcp://localhost:8080'
]);

// or

$response = fetch('https://example.com')
    ->withProxy('tcp://localhost:8080')
    ->get();

echo $response->statusText();

Authentication Example

<?php

$response = fetch('https://example.com/secure-endpoint', [
    'auth' => ['username', 'password']
]);

// or

$response = fetch('https://example.com')
    ->baseUri('https://example.com/')
    ->withAuth('username', 'password')
    ->get('/secure-endpoint');

echo $response->statusText();

License

This project is licensed under the MIT License - see the LICENSE.md file for details.

Contributing

Contributions are what make the open-source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

We’re currently looking for help in the following areas:

• Expanding test coverage for async task management
• Improving documentation for more advanced use cases
• Adding support for additional HTTP methods and protocols

To contribute:

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/amazing-feature)
3. Commit your Changes (git commit -m 'Add some amazing-feature')
4. Push to the Branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Authors

• [Jerome Thayananthajothy] - Initial work - Thavarshan

See also the list of contributors who participated in this project.

Acknowledgments

• Thanks to Guzzle HTTP for providing the underlying HTTP client that powers synchronous requests.