Need help with http?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

reactphp
475 Stars 103 Forks MIT License 646 Commits 14 Opened issues

Description

Event-driven, streaming HTTP client and server implementation for ReactPHP.

Services available

!
?

Need anything else?

Contributors list

HTTP

Build Status

Event-driven, streaming HTTP client and server implementation for ReactPHP.

This HTTP library provides re-usable implementations for an HTTP client and server based on ReactPHP's

Socket
and
EventLoop
components. Its client component allows you to send any number of async HTTP/HTTPS requests concurrently. Its server component allows you to build plaintext HTTP and secure HTTPS servers that accept incoming HTTP requests from HTTP clients (such as web browsers). This library provides async, streaming means for all of this, so you can handle multiple concurrent HTTP requests without blocking.

Table of contents

Quickstart example

Once installed, you can use the following code to access a HTTP webserver and send some simple HTTP GET requests:

$loop = React\EventLoop\Factory::create();
$client = new React\Http\Browser($loop);

$client->get('http://www.google.com/')->then(function (Psr\Http\Message\ResponseInterface $response) { var_dump($response->getHeaders(), (string)$response->getBody()); });

$loop->run();

This is an HTTP server which responds with

Hello World!
to every request.
$loop = React\EventLoop\Factory::create();

$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) { return new React\Http\Message\Response( 200, array( 'Content-Type' => 'text/plain' ), "Hello World!\n" ); });

$socket = new React\Socket\Server(8080, $loop); $server->listen($socket);

$loop->run();

See also the examples.

Client Usage

Request methods

Most importantly, this project provides a

Browser
object that offers several methods that resemble the HTTP protocol methods:

$browser->get($url, array $headers = array());
$browser->head($url, array $headers = array());
$browser->post($url, array $headers = array(), string|ReadableStreamInterface $contents = '');
$browser->delete($url, array $headers = array(), string|ReadableStreamInterface $contents = '');
$browser->put($url, array $headers = array(), string|ReadableStreamInterface $contents = '');
$browser->patch($url, array $headers = array(), string|ReadableStreamInterface $contents = '');

Each of these methods requires a

$url
and some optional parameters to send an HTTP request. Each of these method names matches the respective HTTP request method, for example the
get()
method sends an HTTP
GET
request.

You can optionally pass an associative array of additional

$headers
that will be sent with this HTTP request. Additionally, each method will automatically add a matching
Content-Length
request header if an outgoing request body is given and its size is known and non-empty. For an empty request body, if will only include a
Content-Length: 0
request header if the request method usually expects a request body (only applies to
POST
,
PUT
and
PATCH
HTTP request methods).

If you're using a streaming request body, it will default to using

Transfer-Encoding: chunked
unless you explicitly pass in a matching
Content-Length
request header. See also streaming request for more details.

By default, all of the above methods default to sending requests using the HTTP/1.1 protocol version. If you want to explicitly use the legacy HTTP/1.0 protocol version, you can use the

withProtocolVersion()
method. If you want to use any other or even custom HTTP request method, you can use the
request()
method.

Each of the above methods supports async operation and either fulfills with a PSR-7

ResponseInterface
or rejects with an

Exception
. Please see the following chapter about promises for more details.

Promises

Sending requests is async (non-blocking), so you can actually send multiple requests in parallel. The

Browser
will respond to each request with a PSR-7
ResponseInterface
message, the order is not guaranteed. Sending requests uses a Promise-based interface that makes it easy to react to when an HTTP request is completed (i.e. either successfully fulfilled or rejected with an error):
$browser->get($url)->then(
    function (Psr\Http\Message\ResponseInterface $response) {
        var_dump('Response received', $response);
    },
    function (Exception $error) {
        var_dump('There was an error', $error->getMessage());
    }
);

If this looks strange to you, you can also use the more traditional blocking API.

Keep in mind that resolving the Promise with the full response message means the whole response body has to be kept in memory. This is easy to get started and works reasonably well for smaller responses (such as common HTML pages or RESTful or JSON API requests).

You may also want to look into the streaming API:

  • If you're dealing with lots of concurrent requests (100+) or
  • If you want to process individual data chunks as they happen (without having to wait for the full response body) or
  • If you're expecting a big response body size (1 MiB or more, for example when downloading binary files) or
  • If you're unsure about the response body size (better be safe than sorry when accessing arbitrary remote HTTP endpoints and the response body size is unknown in advance).

Cancellation

The returned Promise is implemented in such a way that it can be cancelled when it is still pending. Cancelling a pending promise will reject its value with an Exception and clean up any underlying resources.

$promise = $browser->get($url);

$loop->addTimer(2.0, function () use ($promise) { $promise->cancel(); });

Timeouts

This library uses a very efficient HTTP implementation, so most HTTP requests should usually be completed in mere milliseconds. However, when sending HTTP requests over an unreliable network (the internet), there are a number of things that can go wrong and may cause the request to fail after a time. As such, this library respects PHP's

default_socket_timeout
setting (default 60s) as a timeout for sending the outgoing HTTP request and waiting for a successful response and will otherwise cancel the pending request and reject its value with an Exception.

Note that this timeout value covers creating the underlying transport connection, sending the HTTP request, receiving the HTTP response headers and its full response body and following any eventual redirects. See also redirects below to configure the number of redirects to follow (or disable following redirects altogether) and also streaming below to not take receiving large response bodies into account for this timeout.

You can use the

withTimeout()
method to pass a custom timeout value in seconds like this:

$browser = $browser->withTimeout(10.0);

$browser->get($url)->then(function (Psr\Http\Message\ResponseInterface $response) { // response received within 10 seconds maximum var_dump($response->getHeaders()); });

Similarly, you can use a bool

false
to not apply a timeout at all or use a bool
true
value to restore the default handling. See
withTimeout()
for more details.

If you're using a streaming response body, the time it takes to receive the response body stream will not be included in the timeout. This allows you to keep this incoming stream open for a longer time, such as when downloading a very large stream or when streaming data over a long-lived connection.

If you're using a streaming request body, the time it takes to send the request body stream will not be included in the timeout. This allows you to keep this outgoing stream open for a longer time, such as when uploading a very large stream.

Note that this timeout handling applies to the higher-level HTTP layer. Lower layers such as socket and DNS may also apply (different) timeout values. In particular, the underlying socket connection uses the same

default_socket_timeout
setting to establish the underlying transport connection. To control this connection timeout behavior, you can inject a custom
Connector
like this:
$browser = new React\Http\Browser(
    $loop,
    new React\Socket\Connector(
        $loop,
        array(
            'timeout' => 5
        )
    )
);

Authentication

This library supports HTTP Basic Authentication using the

Authorization: Basic …
request header or allows you to set an explicit
Authorization
request header.

By default, this library does not include an outgoing

Authorization
request header. If the server requires authentication, if may return a
401
(Unauthorized) status code which will reject the request by default (see also the
withRejectErrorResponse()
method
below).

In order to pass authentication details, you can simple pass the username and password as part of the request URL like this:

$promise = $browser->get('https://user:[email protected]/api');

Note that special characters in the authentication details have to be percent-encoded, see also

rawurlencode()
. This example will automatically pass the base64-encoded authentication details using the outgoing

Authorization: Basic …
request header. If the HTTP endpoint you're talking to requires any other authentication scheme, you can also pass this header explicitly. This is common when using (RESTful) HTTP APIs that use OAuth access tokens or JSON Web Tokens (JWT):
$token = 'abc123';

$promise = $browser->get( 'https://example.com/api', array( 'Authorization' => 'Bearer ' . $token ) );

When following redirects, the

Authorization
request header will never be sent to any remote hosts by default. When following a redirect where the
Location
response header contains authentication details, these details will be sent for following requests. See also redirects below.

Redirects

By default, this library follows any redirects and obeys

3xx
(Redirection) status codes using the
Location
response header from the remote server. The promise will be fulfilled with the last response from the chain of redirects.
$browser->get($url, $headers)->then(function (Psr\Http\Message\ResponseInterface $response) {
    // the final response will end up here
    var_dump($response->getHeaders());
});

Any redirected requests will follow the semantics of the original request and will include the same request headers as the original request except for those listed below. If the original request contained a request body, this request body will never be passed to the redirected request. Accordingly, each redirected request will remove any

Content-Length
and
Content-Type
request headers.

If the original request used HTTP authentication with an

Authorization
request header, this request header will only be passed as part of the redirected request if the redirected URL is using the same host. In other words, the
Authorizaton
request header will not be forwarded to other foreign hosts due to possible privacy/security concerns. When following a redirect where the
Location
response header contains authentication details, these details will be sent for following requests.

You can use the

withFollowRedirects()
method to control the maximum number of redirects to follow or to return any redirect responses as-is and apply custom redirection logic like this:

$browser = $browser->withFollowRedirects(false);

$browser->get($url)->then(function (Psr\Http\Message\ResponseInterface $response) { // any redirects will now end up here var_dump($response->getHeaders()); });

See also

withFollowRedirects()
for more details.

Blocking

As stated above, this library provides you a powerful, async API by default.

If, however, you want to integrate this into your traditional, blocking environment, you should look into also using clue/reactphp-block.

The resulting blocking code could look something like this:

use Clue\React\Block;

$loop = React\EventLoop\Factory::create(); $browser = new React\Http\Browser($loop);

$promise = $browser->get('http://example.com/');

try { $response = Block\await($promise, $loop); // response successfully received } catch (Exception $e) { // an error occured while performing the request }

Similarly, you can also process multiple requests concurrently and await an array of

Response
objects:
$promises = array(
    $browser->get('http://example.com/'),
    $browser->get('http://www.example.org/'),
);

$responses = Block\awaitAll($promises, $loop);

Please refer to clue/reactphp-block for more details.

Keep in mind the above remark about buffering the whole response message in memory. As an alternative, you may also see one of the following chapters for the streaming API.

Concurrency

As stated above, this library provides you a powerful, async API. Being able to send a large number of requests at once is one of the core features of this project. For instance, you can easily send 100 requests concurrently while processing SQL queries at the same time.

Remember, with great power comes great responsibility. Sending an excessive number of requests may either take up all resources on your side or it may even get you banned by the remote side if it sees an unreasonable number of requests from your side.

// watch out if array contains many elements
foreach ($urls as $url) {
    $browser->get($url)->then(function (Psr\Http\Message\ResponseInterface $response) {
        var_dump($response->getHeaders());
    });
}

As a consequence, it's usually recommended to limit concurrency on the sending side to a reasonable value. It's common to use a rather small limit, as doing more than a dozen of things at once may easily overwhelm the receiving side. You can use clue/reactphp-mq as a lightweight in-memory queue to concurrently do many (but not too many) things at once:

// wraps Browser in a Queue object that executes no more than 10 operations at once
$q = new Clue\React\Mq\Queue(10, null, function ($url) use ($browser) {
    return $browser->get($url);
});

foreach ($urls as $url) { $q($url)->then(function (Psr\Http\Message\ResponseInterface $response) { var_dump($response->getHeaders()); }); }

Additional requests that exceed the concurrency limit will automatically be enqueued until one of the pending requests completes. This integrates nicely with the existing Promise-based API. Please refer to clue/reactphp-mq for more details.

This in-memory approach works reasonably well for some thousand outstanding requests. If you're processing a very large input list (think millions of rows in a CSV or NDJSON file), you may want to look into using a streaming approach instead. See clue/reactphp-flux for more details.

Streaming response

All of the above examples assume you want to store the whole response body in memory. This is easy to get started and works reasonably well for smaller responses.

However, there are several situations where it's usually a better idea to use a streaming approach, where only small chunks have to be kept in memory:

  • If you're dealing with lots of concurrent requests (100+) or
  • If you want to process individual data chunks as they happen (without having to wait for the full response body) or
  • If you're expecting a big response body size (1 MiB or more, for example when downloading binary files) or
  • If you're unsure about the response body size (better be safe than sorry when accessing arbitrary remote HTTP endpoints and the response body size is unknown in advance).

You can use the

requestStreaming()
method to send an arbitrary HTTP request and receive a streaming response. It uses the same HTTP message API, but does not buffer the response body in memory. It only processes the response body in small chunks as data is received and forwards this data through ReactPHP's Stream API. This works for (any number of) responses of arbitrary sizes.

This means it resolves with a normal PSR-7

ResponseInterface
, which can be used to access the response message parameters as usual. You can access the message body as usual, however it now also implements ReactPHP's
ReadableStreamInterface
as well as parts of the PSR-7
StreamInterface
.

$browser->requestStreaming('GET', $url)->then(function (Psr\Http\Message\ResponseInterface $response) {
    $body = $response->getBody();
    assert($body instanceof Psr\Http\Message\StreamInterface);
    assert($body instanceof React\Stream\ReadableStreamInterface);

$body->on('data', function ($chunk) {
    echo $chunk;
});

$body->on('error', function (Exception $error) {
    echo 'Error: ' . $error->getMessage() . PHP_EOL;
});

$body->on('close', function () {
    echo '[DONE]' . PHP_EOL;
});

});

See also the stream download benchmark example and the stream forwarding example.

You can invoke the following methods on the message body:

$body->on($event, $callback);
$body->eof();
$body->isReadable();
$body->pipe(React\Stream\WritableStreamInterface $dest, array $options = array());
$body->close();
$body->pause();
$body->resume();

Because the message body is in a streaming state, invoking the following methods doesn't make much sense:

$body->__toString(); // ''
$body->detach(); // throws BadMethodCallException
$body->getSize(); // null
$body->tell(); // throws BadMethodCallException
$body->isSeekable(); // false
$body->seek(); // throws BadMethodCallException
$body->rewind(); // throws BadMethodCallException
$body->isWritable(); // false
$body->write(); // throws BadMethodCallException
$body->read(); // throws BadMethodCallException
$body->getContents(); // throws BadMethodCallException

Note how timeouts apply slightly differently when using streaming. In streaming mode, the timeout value covers creating the underlying transport connection, sending the HTTP request, receiving the HTTP response headers and following any eventual redirects. In particular, the timeout value does not take receiving (possibly large) response bodies into account.

If you want to integrate the streaming response into a higher level API, then working with Promise objects that resolve with Stream objects is often inconvenient. Consider looking into also using react/promise-stream. The resulting streaming code could look something like this:

use React\Promise\Stream;

function download(Browser $browser, string $url): React\Stream\ReadableStreamInterface { return Stream\unwrapReadable( $browser->requestStreaming('GET', $url)->then(function (Psr\Http\Message\ResponseInterface $response) { return $response->getBody(); }) ); }

$stream = download($browser, $url); $stream->on('data', function ($data) { echo $data; });

See also the

requestStreaming()
method for more details.

Streaming request

Besides streaming the response body, you can also stream the request body. This can be useful if you want to send big POST requests (uploading files etc.) or process many outgoing streams at once. Instead of passing the body as a string, you can simply pass an instance implementing ReactPHP's

ReadableStreamInterface
to the request methods like this:

$browser->post($url, array(), $stream)->then(function (Psr\Http\Message\ResponseInterface $response) {
    echo 'Successfully sent.';
});

If you're using a streaming request body (

React\Stream\ReadableStreamInterface
), it will default to using
Transfer-Encoding: chunked
or you have to explicitly pass in a matching
Content-Length
request header like so:
$body = new React\Stream\ThroughStream();
$loop->addTimer(1.0, function () use ($body) {
    $body->end("hello world");
});

$browser->post($url, array('Content-Length' => '11'), $body);

If the streaming request body emits an

error
event or is explicitly closed without emitting a successful
end
event first, the request will automatically be closed and rejected.

HTTP proxy

You can also establish your outgoing connections through an HTTP CONNECT proxy server by adding a dependency to clue/reactphp-http-proxy.

HTTP CONNECT proxy servers (also commonly known as "HTTPS proxy" or "SSL proxy") are commonly used to tunnel HTTPS traffic through an intermediary ("proxy"), to conceal the origin address (anonymity) or to circumvent address blocking (geoblocking). While many (public) HTTP CONNECT proxy servers often limit this to HTTPS port

443
only, this can technically be used to tunnel any TCP/IP-based protocol, such as plain HTTP and TLS-encrypted HTTPS.
$proxy = new Clue\React\HttpProxy\ProxyConnector(
    'http://127.0.0.1:8080',
    new React\Socket\Connector($loop)
);

$connector = new React\Socket\Connector($loop, array( 'tcp' => $proxy, 'dns' => false ));

$browser = new React\Http\Browser($loop, $connector);

See also the HTTP CONNECT proxy example.

SOCKS proxy

You can also establish your outgoing connections through a SOCKS proxy server by adding a dependency to clue/reactphp-socks.

The SOCKS proxy protocol family (SOCKS5, SOCKS4 and SOCKS4a) is commonly used to tunnel HTTP(S) traffic through an intermediary ("proxy"), to conceal the origin address (anonymity) or to circumvent address blocking (geoblocking). While many (public) SOCKS proxy servers often limit this to HTTP(S) port

80
and
443
only, this can technically be used to tunnel any TCP/IP-based protocol.
$proxy = new Clue\React\Socks\Client(
    'socks://127.0.0.1:1080',
    new React\Socket\Connector($loop)
);

$connector = new React\Socket\Connector($loop, array( 'tcp' => $proxy, 'dns' => false ));

$browser = new React\Http\Browser($loop, $connector);

See also the SOCKS proxy example.

SSH proxy

You can also establish your outgoing connections through an SSH server by adding a dependency to clue/reactphp-ssh-proxy.

Secure Shell (SSH) is a secure network protocol that is most commonly used to access a login shell on a remote server. Its architecture allows it to use multiple secure channels over a single connection. Among others, this can also be used to create an "SSH tunnel", which is commonly used to tunnel HTTP(S) traffic through an intermediary ("proxy"), to conceal the origin address (anonymity) or to circumvent address blocking (geoblocking). This can be used to tunnel any TCP/IP-based protocol (HTTP, SMTP, IMAP etc.), allows you to access local services that are otherwise not accessible from the outside (database behind firewall) and as such can also be used for plain HTTP and TLS-encrypted HTTPS.

$proxy = new Clue\React\SshProxy\SshSocksConnector('[email protected]:22', $loop);

$connector = new React\Socket\Connector($loop, array( 'tcp' => $proxy, 'dns' => false ));

$browser = new React\Http\Browser($loop, $connector);

See also the SSH proxy example.

Unix domain sockets

By default, this library supports transport over plaintext TCP/IP and secure TLS connections for the

http://
and
https://
URL schemes respectively. This library also supports Unix domain sockets (UDS) when explicitly configured.

In order to use a UDS path, you have to explicitly configure the connector to override the destination URL so that the hostname given in the request URL will no longer be used to establish the connection:

$connector = new React\Socket\FixedUriConnector(
    'unix:///var/run/docker.sock',
    new React\Socket\UnixConnector($loop)
);

$browser = new Browser($loop, $connector);

$client->get('http://localhost/info')->then(function (Psr\Http\Message\ResponseInterface $response) { var_dump($response->getHeaders(), (string)$response->getBody()); });

See also the Unix Domain Sockets (UDS) example.

Server Usage

Server

The

React\Http\Server
class is responsible for handling incoming connections and then processing each incoming HTTP request.

When a complete HTTP request has been received, it will invoke the given request handler function. This request handler function needs to be passed to the constructor and will be invoked with the respective request object and expects a response object in return:

$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    return new React\Http\Message\Response(
        200,
        array(
            'Content-Type' => 'text/plain'
        ),
        "Hello World!\n"
    );
});

Each incoming HTTP request message is always represented by the PSR-7

ServerRequestInterface
, see also following request chapter for more details.

Each outgoing HTTP response message is always represented by the PSR-7

ResponseInterface
, see also following response chapter for more details.

In order to start listening for any incoming connections, the

Server
needs to be attached to an instance of
React\Socket\ServerInterface
through the
listen()
method as described in the following chapter. In its most simple form, you can attach this to a
React\Socket\Server
in order to start a plaintext HTTP server like this:
$server = new React\Http\Server($loop, $handler);

$socket = new React\Socket\Server('0.0.0.0:8080', $loop); $server->listen($socket);

See also the

listen()
method and the hello world server example for more details.

By default, the

Server
buffers and parses the complete incoming HTTP request in memory. It will invoke the given request handler function when the complete request headers and request body has been received. This means the request object passed to your request handler function will be fully compatible with PSR-7 (http-message). This provides sane defaults for 80% of the use cases and is the recommended way to use this library unless you're sure you know what you're doing.

On the other hand, buffering complete HTTP requests in memory until they can be processed by your request handler function means that this class has to employ a number of limits to avoid consuming too much memory. In order to take the more advanced configuration out your hand, it respects setting from your

php.ini
to apply its default settings. This is a list of PHP settings this class respects with their respective default values:

memory_limit 128M
post_max_size 8M // capped at 64K

enable_post_data_reading 1 max_input_nesting_level 64 max_input_vars 1000

file_uploads 1 upload_max_filesize 2M max_file_uploads 20

In particular, the

post_max_size
setting limits how much memory a single HTTP request is allowed to consume while buffering its request body. This needs to be limited because the server can process a large number of requests concurrently, so the server may potentially consume a large amount of memory otherwise. To support higher concurrency by default, this value is capped at
64K
. If you assign a higher value, it will only allow
64K
by default. If a request exceeds this limit, its request body will be ignored and it will be processed like a request with no request body at all. See below for explicit configuration to override this setting.

By default, this class will try to avoid consuming more than half of your

memory_limit
for buffering multiple concurrent HTTP requests. As such, with the above default settings of
128M
max, it will try to consume no more than
64M
for buffering multiple concurrent HTTP requests. As a consequence, it will limit the concurrency to
1024
HTTP requests with the above defaults.

It is imperative that you assign reasonable values to your PHP ini settings. It is usually recommended to not support buffering incoming HTTP requests with a large HTTP request body (e.g. large file uploads). If you want to increase this buffer size, you will have to also increase the total memory limit to allow for more concurrent requests (set

memory_limit 512M
or more) or explicitly limit concurrency.

In order to override the above buffering defaults, you can configure the

Server
explicitly. You can use the
LimitConcurrentRequestsMiddleware
and
RequestBodyBufferMiddleware
(see below) to explicitly configure the total number of requests that can be handled at once like this:
$server = new React\Http\Server(
    $loop,
    new React\Http\Middleware\StreamingRequestMiddleware(),
    new React\Http\Middleware\LimitConcurrentRequestsMiddleware(100), // 100 concurrent buffering handlers
    new React\Http\Middleware\RequestBodyBufferMiddleware(2 * 1024 * 1024), // 2 MiB per request
    new React\Http\Middleware\RequestBodyParserMiddleware(),
    $handler
);

In this example, we allow processing up to 100 concurrent requests at once and each request can buffer up to

2M
. This means you may have to keep a maximum of
200M
of memory for incoming request body buffers. Accordingly, you need to adjust the
memory_limit
ini setting to allow for these buffers plus your actual application logic memory requirements (think
512M
or more).

Internally, this class automatically assigns these middleware handlers automatically when no

StreamingRequestMiddleware
is given. Accordingly, you can use this example to override all default settings to implement custom limits.

As an alternative to buffering the complete request body in memory, you can also use a streaming approach where only small chunks of data have to be kept in memory:

$server = new React\Http\Server(
    $loop,
    new React\Http\Middleware\StreamingRequestMiddleware(),
    $handler
);

In this case, it will invoke the request handler function once the HTTP request headers have been received, i.e. before receiving the potentially much larger HTTP request body. This means the request passed to your request handler function may not be fully compatible with PSR-7. This is specifically designed to help with more advanced use cases where you want to have full control over consuming the incoming HTTP request body and concurrency settings. See also streaming incoming request below for more details.

listen()

The

listen(React\Socket\ServerInterface $socket): void
method can be used to start listening for HTTP requests on the given socket server instance.

The given

React\Socket\ServerInterface
is responsible for emitting the underlying streaming connections. This HTTP server needs to be attached to it in order to process any connections and pase incoming streaming data as incoming HTTP request messages. In its most common form, you can attach this to a
React\Socket\Server
in order to start a plaintext HTTP server like this:

$server = new React\Http\Server($loop, $handler);

$socket = new React\Socket\Server('0.0.0.0:8080', $loop); $server->listen($socket);

See also hello world server example for more details.

This example will start listening for HTTP requests on the alternative HTTP port

8080
on all interfaces (publicly). As an alternative, it is very common to use a reverse proxy and let this HTTP server listen on the localhost (loopback) interface only by using the listen address
127.0.0.1:8080
instead. This way, you host your application(s) on the default HTTP port
80
and only route specific requests to this HTTP server.

Likewise, it's usually recommended to use a reverse proxy setup to accept secure HTTPS requests on default HTTPS port

443
(TLS termination) and only route plaintext requests to this HTTP server. As an alternative, you can also accept secure HTTPS requests with this HTTP server by attaching this to a
React\Socket\Server
using a secure TLS listen address, a certificate file and optional
passphrase
like this:
$server = new React\Http\Server($loop, $handler);

$socket = new React\Socket\Server('tls://0.0.0.0:8443', $loop, array( 'local_cert' => DIR . '/localhost.pem' )); $server->listen($socket);

See also hello world HTTPS example for more details.

Server Request

As seen above, the

Server
class is responsible for handling incoming connections and then processing each incoming HTTP request.

The request object will be processed once the request has been received by the client. This request object implements the PSR-7

ServerRequestInterface
which in turn extends the PSR-7
RequestInterface
and will be passed to the callback function like this.

$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    $body = "The method of the request is: " . $request->getMethod();
    $body .= "The requested path is: " . $request->getUri()->getPath();

return new React\Http\Message\Response(
    200,
    array(
        'Content-Type' => 'text/plain'
    ),
    $body
);

});

For more details about the request object, also check out the documentation of PSR-7

ServerRequestInterface
and PSR-7
RequestInterface
.

Request parameters

The

getServerParams(): mixed[]
method can be used to get server-side parameters similar to the
$_SERVER
variable. The following parameters are currently available:
  • REMOTE_ADDR
    The IP address of the request sender
  • REMOTE_PORT
    Port of the request sender
  • SERVER_ADDR
    The IP address of the server
  • SERVER_PORT
    The port of the server
  • REQUEST_TIME
    Unix timestamp when the complete request header has been received, as integer similar to
    time()
  • REQUEST_TIME_FLOAT
    Unix timestamp when the complete request header has been received, as float similar to
    microtime(true)
  • HTTPS
    Set to 'on' if the request used HTTPS, otherwise it won't be set
$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    $body = "Your IP is: " . $request->getServerParams()['REMOTE_ADDR'];

return new React\Http\Message\Response(
    200,
    array(
        'Content-Type' => 'text/plain'
    ),
    $body
);

});

See also whatsmyip server example.

Advanced: Note that address parameters will not be set if you're listening on a Unix domain socket (UDS) path as this protocol lacks the concept of host/port.

Query parameters

The

getQueryParams(): array
method can be used to get the query parameters similiar to the
$_GET
variable.
$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    $queryParams = $request->getQueryParams();

$body = 'The query parameter "foo" is not set. Click the following link ';
$body .= '<a href="https://github.com/reactphp/http/blob/master//?foo=bar">to use query parameter in your request</a>';

if (isset($queryParams['foo'])) {
    $body = 'The value of "foo" is: ' . htmlspecialchars($queryParams['foo']);
}

return new React\Http\Message\Response(
    200,
    array(
        'Content-Type' =&gt; 'text/html'
    ),
    $body
);

});

The response in the above example will return a response body with a link. The URL contains the query parameter

foo
with the value
bar
. Use
htmlentities
like in this example to prevent Cross-Site Scripting (abbreviated as XSS).

See also server query parameters example.

Request body

By default, the

Server
will buffer and parse the full request body in memory. This means the given request object includes the parsed request body and any file uploads.

As an alternative to the default buffering logic, you can also use the

StreamingRequestMiddleware
. Jump to the next chapter to learn more about how to process a streaming incoming request.

As stated above, each incoming HTTP request is always represented by the PSR-7

ServerRequestInterface
. This interface provides several methods that are useful when working with the incoming request body as described below.

The

getParsedBody(): null|array|object
method can be used to get the parsed request body, similar to PHP's
$_POST
variable
. This method may return a (possibly nested) array structure with all body parameters or a
null
value if the request body could not be parsed. By default, this method will only return parsed data for requests using
Content-Type: application/x-www-form-urlencoded
or
Content-Type: multipart/form-data
request headers (commonly used for
POST
requests for HTML form submission data).
$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    $name = $request->getParsedBody()['name'] ?? 'anonymous';

return new React\Http\Message\Response(
    200,
    array(),
    "Hello $name!\n"
);

});

See also form upload example for more details.

The

getBody(): StreamInterface
method can be used to get the raw data from this request body, similar to PHP's
php://input
stream
. This method returns an instance of the request body represented by the PSR-7
StreamInterface
. This is particularly useful when using a custom request body that will not otherwise be parsed by default, such as a JSON (
Content-Type: application/json
) or an XML (
Content-Type: application/xml
) request body (which is commonly used for
POST
,
PUT
or
PATCH
requests in JSON-based or RESTful/RESTish APIs).
$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    $data = json_decode((string)$request->getBody());
    $name = $data->name ?? 'anonymous';

return new React\Http\Message\Response(
    200,
    array('Content-Type' =&gt; 'application/json'),
    json_encode(['message' =&gt; "Hello $name!"])
);

});

See also JSON API server example for more details.

The

getUploadedFiles(): array
method can be used to get the uploaded files in this request, similar to PHP's
$_FILES
variable
. This method returns a (possibly nested) array structure with all file uploads, each represented by the PSR-7
UploadedFileInterface
. This array will only be filled when using the
Content-Type: multipart/form-data
request header (commonly used for
POST
requests for HTML file uploads).
$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    $files = $request->getUploadedFiles();
    $name = isset($files['avatar']) ? $files['avatar']->getClientFilename() : 'nothing';

return new React\Http\Message\Response(
    200,
    array(),
    "Uploaded $name\n"
);

});

See also form upload server example for more details.

The

getSize(): ?int
method can be used to get the size of the request body, similar to PHP's
$_SERVER['CONTENT_LENGTH']
variable. This method returns the complete size of the request body measured in number of bytes as defined by the message boundaries. This value may be
0
if the request message does not contain a request body (such as a simple
GET
request). This method operates on the buffered request body, i.e. the request body size is always known, even when the request does not specify a
Content-Length
request header or when using
Transfer-Encoding: chunked
for HTTP/1.1 requests.

Note: The

Server
automatically takes care of handling requests with the additional
Expect: 100-continue
request header. When HTTP/1.1 clients want to send a bigger request body, they MAY send only the request headers with an additional
Expect: 100-continue
request header and wait before sending the actual (large) message body. In this case the server will automatically send an intermediary
HTTP/1.1 100 Continue
response to the client. This ensures you will receive the request body without a delay as expected.

Streaming incoming request

If you're using the advanced

StreamingRequestMiddleware
, the request object will be processed once the request headers have been received. This means that this happens irrespective of (i.e. before) receiving the (potentially much larger) request body.

Note that this is non-standard behavior considered advanced usage. Jump to the previous chapter to learn more about how to process a buffered request body.

While this may be uncommon in the PHP ecosystem, this is actually a very powerful approach that gives you several advantages not otherwise possible:

  • React to requests before receiving a large request body, such as rejecting an unauthenticated request or one that exceeds allowed message lengths (file uploads).
  • Start processing parts of the request body before the remainder of the request body arrives or if the sender is slowly streaming data.
  • Process a large request body without having to buffer anything in memory, such as accepting a huge file upload or possibly unlimited request body stream.

The

getBody(): StreamInterface
method can be used to access the request body stream. In the streaming mode, this method returns a stream instance that implements both the PSR-7
StreamInterface
and the ReactPHP
ReadableStreamInterface
. However, most of the PSR-7
StreamInterface
methods have been designed under the assumption of being in control of a synchronous request body. Given that this does not apply to this server, the following PSR-7
StreamInterface
methods are not used and SHOULD NOT be called:
tell()
,
eof()
,
seek()
,
rewind()
,
write()
and
read()
. If this is an issue for your use case and/or you want to access uploaded files, it's highly recommended to use a buffered request body or use the
RequestBodyBufferMiddleware
instead. The ReactPHP
ReadableStreamInterface
gives you access to the incoming request body as the individual chunks arrive:
$server = new React\Http\Server(
    $loop,
    new React\Http\Middleware\StreamingRequestMiddleware(),
    function (Psr\Http\Message\ServerRequestInterface $request) {
        $body = $request->getBody();
        assert($body instanceof Psr\Http\Message\StreamInterface);
        assert($body instanceof React\Stream\ReadableStreamInterface);

    return new React\Promise\Promise(function ($resolve, $reject) use ($body) {
        $bytes = 0;
        $body-&gt;on('data', function ($data) use (&amp;$bytes) {
            $bytes += strlen($data);
        });

        $body-&gt;on('end', function () use ($resolve, &amp;$bytes){
            $resolve(new React\Http\Message\Response(
                200,
                array(
                    'Content-Type' =&gt; 'text/plain'
                ),
                "Received $bytes bytes\n"
            ));
        });

        // an error occures e.g. on invalid chunked encoded data or an unexpected 'end' event
        $body-&gt;on('error', function (\Exception $exception) use ($resolve, &amp;$bytes) {
            $resolve(new React\Http\Message\Response(
                400,
                array(
                    'Content-Type' =&gt; 'text/plain'
                ),
                "Encountered error after $bytes bytes: {$exception-&gt;getMessage()}\n"
            ));
        });
    });
}

);

The above example simply counts the number of bytes received in the request body. This can be used as a skeleton for buffering or processing the request body.

See also streaming request server example for more details.

The

data
event will be emitted whenever new data is available on the request body stream. The server also automatically takes care of decoding any incoming requests using
Transfer-Encoding: chunked
and will only emit the actual payload as data.

The

end
event will be emitted when the request body stream terminates successfully, i.e. it was read until its expected end.

The

error
event will be emitted in case the request stream contains invalid data for
Transfer-Encoding: chunked
or when the connection closes before the complete request stream has been received. The server will automatically stop reading from the connection and discard all incoming data instead of closing it. A response message can still be sent (unless the connection is already closed).

A

close
event will be emitted after an
error
or
end
event.

For more details about the request body stream, check out the documentation of ReactPHP

ReadableStreamInterface
.

The

getSize(): ?int
method can be used to get the size of the request body, similar to PHP's
$_SERVER['CONTENT_LENGTH']
variable. This method returns the complete size of the request body measured in number of bytes as defined by the message boundaries. This value may be
0
if the request message does not contain a request body (such as a simple
GET
request). This method operates on the streaming request body, i.e. the request body size may be unknown (
null
) when using
Transfer-Encoding: chunked
for HTTP/1.1 requests.
$server = new React\Http\Server(
    $loop,
    new React\Http\Middleware\StreamingRequestMiddleware(),
    function (Psr\Http\Message\ServerRequestInterface $request) {
        $size = $request->getBody()->getSize();
        if ($size === null) {
            $body = 'The request does not contain an explicit length.';
            $body .= 'This example does not accept chunked transfer encoding.';

        return new React\Http\Message\Response(
            411,
            array(
                'Content-Type' =&gt; 'text/plain'
            ),
            $body
        );
    }

    return new React\Http\Message\Response(
        200,
        array(
            'Content-Type' =&gt; 'text/plain'
        ),
        "Request body size: " . $size . " bytes\n"
    );
}

);

Note: The

Server
automatically takes care of handling requests with the additional
Expect: 100-continue
request header. When HTTP/1.1 clients want to send a bigger request body, they MAY send only the request headers with an additional
Expect: 100-continue
request header and wait before sending the actual (large) message body. In this case the server will automatically send an intermediary
HTTP/1.1 100 Continue
response to the client. This ensures you will receive the streaming request body without a delay as expected.

Request method

Note that the server supports any request method (including custom and non- standard ones) and all request-target formats defined in the HTTP specs for each respective method, including normal

origin-form
requests as well as proxy requests in
absolute-form
and
authority-form
. The
getUri(): UriInterface
method can be used to get the effective request URI which provides you access to individiual URI components. Note that (depending on the given
request-target
) certain URI components may or may not be present, for example the
getPath(): string
method will return an empty string for requests in
asterisk-form
or
authority-form
. Its
getHost(): string
method will return the host as determined by the effective request URI, which defaults to the local socket address if a HTTP/1.0 client did not specify one (i.e. no
Host
header). Its
getScheme(): string
method will return
http
or
https
depending on whether the request was made over a secure TLS connection to the target host.

The

Host
header value will be sanitized to match this host component plus the port component only if it is non-standard for this URI scheme.

You can use

getMethod(): string
and
getRequestTarget(): string
to check this is an accepted request and may want to reject other requests with an appropriate error code, such as
400
(Bad Request) or
405
(Method Not Allowed).

The

CONNECT
method is useful in a tunneling setup (HTTPS proxy) and not something most HTTP servers would want to care about. Note that if you want to handle this method, the client MAY send a different request-target than the
Host
header value (such as removing default ports) and the request-target MUST take precendence when forwarding.

Cookie parameters

The

getCookieParams(): string[]
method can be used to get all cookies sent with the current request.
$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    $key = 'react\php';

if (isset($request-&gt;getCookieParams()[$key])) {
    $body = "Your cookie value is: " . $request-&gt;getCookieParams()[$key];

    return new React\Http\Message\Response(
        200,
        array(
            'Content-Type' =&gt; 'text/plain'
        ),
        $body
    );
}

return new React\Http\Message\Response(
    200,
    array(
        'Content-Type' =&gt; 'text/plain',
        'Set-Cookie' =&gt; urlencode($key) . '=' . urlencode('test;more')
    ),
    "Your cookie has been set."
);

});

The above example will try to set a cookie on first access and will try to print the cookie value on all subsequent tries. Note how the example uses the

urlencode()
function to encode non-alphanumeric characters. This encoding is also used internally when decoding the name and value of cookies (which is in line with other implementations, such as PHP's cookie functions).

See also cookie server example for more details.

Invalid request

The

Server
class supports both HTTP/1.1 and HTTP/1.0 request messages. If a client sends an invalid request message, uses an invalid HTTP protocol version or sends an invalid
Transfer-Encoding
request header value, the server will automatically send a
400
(Bad Request) HTTP error response to the client and close the connection. On top of this, it will emit an
error
event that can be used for logging purposes like this:
$server->on('error', function (Exception $e) {
    echo 'Error: ' . $e->getMessage() . PHP_EOL;
});

Note that the server will also emit an

error
event if you do not return a valid response object from your request handler function. See also invalid response for more details.

Server Response

The callback function passed to the constructor of the

Server
is responsible for processing the request and returning a response, which will be delivered to the client.

This function MUST return an instance implementing PSR-7

ResponseInterface
object or a ReactPHP Promise which resolves with a PSR-7
ResponseInterface
object.

This projects ships a

Response
class which implements the PSR-7
ResponseInterface
. In its most simple form, you can use it like this:

$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    return new React\Http\Message\Response(
        200,
        array(
            'Content-Type' => 'text/plain'
        ),
        "Hello World!\n"
    );
});

We use this

Response
class throughout our project examples, but feel free to use any other implementation of the PSR-7
ResponseInterface
. See also the
Response
class
for more details.

Deferred response

The example above returns the response directly, because it needs no time to be processed. Using a database, the file system or long calculations (in fact every action that will take >=1ms) to create your response, will slow down the server. To prevent this you SHOULD use a ReactPHP Promise. This example shows how such a long-term action could look like:

$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) use ($loop) {
    return new Promise(function ($resolve, $reject) use ($loop) {
        $loop->addTimer(1.5, function() use ($resolve) {
            $response = new React\Http\Message\Response(
                200,
                array(
                    'Content-Type' => 'text/plain'
                ),
                "Hello world"
            );
            $resolve($response);
        });
    });
});

The above example will create a response after 1.5 second. This example shows that you need a promise, if your response needs time to created. The

ReactPHP Promise
will resolve in a
Response
object when the request body ends. If the client closes the connection while the promise is still pending, the promise will automatically be cancelled. The promise cancellation handler can be used to clean up any pending resources allocated in this case (if applicable). If a promise is resolved after the client closes, it will simply be ignored.

Streaming outgoing response

The

Response
class in this project supports to add an instance which implements the ReactPHP
ReadableStreamInterface
for the response body. So you are able stream data directly into the response body. Note that other implementations of the PSR-7
ResponseInterface
may only support strings.
$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) use ($loop) {
    $stream = new ThroughStream();

$timer = $loop-&gt;addPeriodicTimer(0.5, function () use ($stream) {
    $stream-&gt;write(microtime(true) . PHP_EOL);
});

$loop-&gt;addTimer(5, function() use ($loop, $timer, $stream) {
    $loop-&gt;cancelTimer($timer);
    $stream-&gt;end();
});

return new React\Http\Message\Response(
    200,
    array(
        'Content-Type' =&gt; 'text/plain'
    ),
    $stream
);

});

The above example will emit every 0.5 seconds the current Unix timestamp with microseconds as float to the client and will end after 5 seconds. This is just a example you could use of the streaming, you could also send a big amount of data via little chunks or use it for body data that needs to calculated.

If the request handler resolves with a response stream that is already closed, it will simply send an empty response body. If the client closes the connection while the stream is still open, the response stream will automatically be closed. If a promise is resolved with a streaming body after the client closes, the response stream will automatically be closed. The

close
event can be used to clean up any pending resources allocated in this case (if applicable).

Note that special care has to be taken if you use a body stream instance that implements ReactPHP's

DuplexStreamInterface
(such as the

ThroughStream
in the above example).

For most cases, this will simply only consume its readable side and forward (send) any data that is emitted by the stream, thus entirely ignoring the writable side of the stream. If however this is either a

101
(Switching Protocols) response or a
2xx
(Successful) response to a
CONNECT
method, it will also write data to the writable side of the stream. This can be avoided by either rejecting all requests with the
CONNECT
method (which is what most normal origin HTTP servers would likely do) or or ensuring that only ever an instance of ReactPHP's
ReadableStreamInterface
is used.

The

101
(Switching Protocols) response code is useful for the more advanced
Upgrade
requests, such as upgrading to the WebSocket protocol or implementing custom protocol logic that is out of scope of the HTTP specs and this HTTP library. If you want to handle the
Upgrade: WebSocket
header, you will likely want to look into using Ratchet instead. If you want to handle a custom protocol, you will likely want to look into the HTTP specs and also see examples #81 and #82 for more details. In particular, the
101
(Switching Protocols) response code MUST NOT be used unless you send an
Upgrade
response header value that is also present in the corresponding HTTP/1.1
Upgrade
request header value. The server automatically takes care of sending a
Connection: upgrade
header value in this case, so you don't have to.

The

CONNECT
method is useful in a tunneling setup (HTTPS proxy) and not something most origin HTTP servers would want to care about. The HTTP specs define an opaque "tunneling mode" for this method and make no use of the message body. For consistency reasons, this library uses a
DuplexStreamInterface
in the response body for tunneled application data. This implies that that a
2xx
(Successful) response to a
CONNECT
request can in fact use a streaming response body for the tunneled application data, so that any raw data the client sends over the connection will be piped through the writable stream for consumption. Note that while the HTTP specs make no use of the request body for
CONNECT
requests, one may still be present. Normal request body processing applies here and the connection will only turn to "tunneling mode" after the request body has been processed (which should be empty in most cases). See also HTTP CONNECT server example for more details.

Response length

If the response body size is known, a

Content-Length
response header will be added automatically. This is the most common use case, for example when using a
string
response body like this:
$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) {
    return new React\Http\Message\Response(
        200,
        array(
            'Content-Type' => 'text/plain'
        ),
        "Hello World!\n"
    );
});

If the response body size is unknown, a

Content-Length
response header can not be added automatically. When using a streaming outgoing response without an explicit
Content-Length
response header, outgoing HTTP/1.1 response messages will automatically use
Transfer-Encoding: chunked
while legacy HTTP/1.0 response messages will contain the plain response body. If you know the length of your streaming response body, you MAY want to specify it explicitly like this:
$server = new React\Http\Server($loop, function (Psr\Http\Message\ServerRequestInterface $request) use ($loop) {
    $stream = new ThroughStream();

$loop-&gt;addTimer(2.0, function () use ($stream) {
    $stream-&gt;end("Hello World!\n");
});

return new React\Http\Message\Response(
    200,
    array(
        'Content-Length' =&gt; '13',
        'Content-Type' =&gt; 'text/plain',
    ),
    $stream
);

});

Any response to a

HEAD
request and any response with a
1xx
(Informational),
204
(No Content) or
304
(Not Modified) status code will not include a message body as per the HTTP specs. This means that your callback does not have to take special care of this and any response body will simply be ignored.

Similarly, any

2xx
(Successful) response to a
CONNECT
request, any response with a
1xx
(Informational) or
204
(No Content) status code will not include a
Content-Length
or
Transfer-Encoding
header as these do not apply to these messages. Note that a response to a
HEAD
request and any response with a
304
(Not Modified) status code MAY include these headers even though the message does not contain a response body, because these header would apply to the message if the same request would have used an (unconditional)
GET
.

Invalid response

As stated above, each outgoing HTTP response is always represented by the PSR-7

ResponseInterface
. If your request handler function returns an invalid value or throws an unhandled

Exception
or
Throwable
, the server will automatically send a
500
(Internal Server Error) HTTP error response to the client. On top of this, it will emit an
error
event that can be used for logging purposes like this:
$server->on('error', function (Exception $e) {
    echo 'Error: ' . $e->getMessage() . PHP_EOL;
    if ($e->getPrevious() !== null) {
        echo 'Previous: ' . $e->getPrevious()->getMessage() . PHP_EOL;
    }
});

Note that the server will also emit an

error
event if the client sends an invalid HTTP request that never reaches your request handler function. See also invalid request for more details. Additionally, a streaming incoming request body can also emit an
error
event on the request body.

The server will only send a very generic

500
(Interval Server Error) HTTP error response without any further details to the client if an unhandled error occurs. While we understand this might make initial debugging harder, it also means that the server does not leak any application details or stack traces to the outside by default. It is usually recommended to catch any
Exception
or
Throwable
within your request handler function or alternatively use a
middleware
to avoid this generic error handling and create your own HTTP response message instead.

Default response headers

When a response is returned from the request handler function, it will be processed by the

Server
and then sent back to the client.

A

Server: ReactPHP/1
response header will be added automatically. You can add a custom
Server
response header like this:
$server = new React\Http\Server($loop, function (ServerRequestInterface $request) {
    return new React\Http\Message\Response(
        200,
        array(
            'Server' => 'PHP/3'
        )
    );
});

If you do not want to send this

Sever
response header at all (such as when you don't want to expose the underlying server software), you can use an empty string value like this:
$server = new React\Http\Server($loop, function (ServerRequestInterface $request) {
    return new React\Http\Message\Response(
        200,
        array(
            'Server' => ''
        )
    );
});

A

Date
response header will be added automatically with the current system date and time if none is given. You can add a custom
Date
response header like this:
$server = new React\Http\Server($loop, function (ServerRequestInterface $request) {
    return new React\Http\Message\Response(
        200,
        array(
            'Date' => gmdate('D, d M Y H:i:s \G\M\T')
        )
    );
});

If you do not want to send this

Date
response header at all (such as when you don't have an appropriate clock to rely on), you can use an empty string value like this:
$server = new React\Http\Server($loop, function (ServerRequestInterface $request) {
    return new React\Http\Message\Response(
        200,
        array(
            'Date' => ''
        )
    );
});

The

Server
class will automatically add the protocol version of the request, so you don't have to. For instance, if the client sends the request using the HTTP/1.1 protocol version, the response message will also use the same protocol version, no matter what version is returned from the request handler function.

Note that persistent connections (

Connection: keep-alive
) are currently not supported. As such, HTTP/1.1 response messages will automatically include a
Connection: close
header, irrespective of what header values are passed explicitly.

Middleware

As documented above, the

Server
accepts a single request handler argument that is responsible for processing an incoming HTTP request and then creating and returning an outgoing HTTP response.

Many common use cases involve validating, processing, manipulating the incoming HTTP request before passing it to the final business logic request handler. As such, this project supports the concept of middleware request handlers.

Custom middleware

A middleware request handler is expected to adhere the following rules:

  • It is a valid
    callable
    .
  • It accepts an instance implementing PSR-7
    ServerRequestInterface
    as first argument and an optional
    callable
    as second argument.
  • It returns either:
  • It calls
    $next($request)
    to continue processing the next middleware request handler or returns explicitly without calling
    $next
    to abort the chain.
    • The
      $next
      request handler (recursively) invokes the next request handler from the chain with the same logic as above and returns (or throws) as above.
    • The
      $request
      may be modified prior to calling
      $next($request)
      to change the incoming request the next middleware operates on.
    • The
      $next
      return value may be consumed to modify the outgoing response.
    • The
      $next
      request handler MAY be called more than once if you want to implement custom "retry" logic etc.

Note that this very simple definition allows you to use either anonymous functions or any classes that use the magic

__invoke()
method. This allows you to easily create custom middleware request handlers on the fly or use a class based approach to ease using existing middleware implementations.

While this project does provide the means to use middleware implementations, it does not aim to define how middleware implementations should look like. We realize that there's a vivid ecosystem of middleware implementations and ongoing effort to standardize interfaces between these with PSR-15 (HTTP Server Request Handlers) and support this goal. As such, this project only bundles a few middleware implementations that are required to match PHP's request behavior (see below) and otherwise actively encourages Third-Party Middleware implementations.

In order to use middleware request handlers, simply pass an array with all callables as defined above to the

Server
. The following example adds a middleware request handler that adds the current time to the request as a header (

Request-Time
) and a final request handler that always returns a 200 code without a body:
$server = new React\Http\Server(
    $loop,
    function (Psr\Http\Message\ServerRequestInterface $request, callable $next) {
        $request = $request->withHeader('Request-Time', time());
        return $next($request);
    },
    function (Psr\Http\Message\ServerRequestInterface $request) {
        return new React\Http\Message\Response(200);
    }
);

Note how the middleware request handler and the final request handler have a very simple (and similar) interface. The only difference is that the final request handler does not receive a

$next
handler.

Similarly, you can use the result of the

$next
middleware request handler function to modify the outgoing response. Note that as per the above documentation, the
$next
middleware request handler may return a PSR-7
ResponseInterface
directly or one wrapped in a promise for deferred resolution. In order to simplify handling both paths, you can simply wrap this in a
Promise\resolve()
call like this:
$server = new React\Http\Server(
    $loop,
    function (Psr\Http\Message\ServerRequestInterface $request, callable $next) {
        $promise = React\Promise\resolve($next($request));
        return $promise->then(function (ResponseInterface $response) {
            return $response->withHeader('Content-Type', 'text/html');
        });
    },
    function (Psr\Http\Message\ServerRequestInterface $request) {
        return new React\Http\Message\Response(200);
    }
);

Note that the

$next
middleware request handler may also throw an
Exception
(or return a rejected promise) as described above. The previous example does not catch any exceptions and would thus signal an error condition to the
Server
. Alternatively, you can also catch any
Exception
to implement custom error handling logic (or logging etc.) by wrapping this in a
Promise
like this:
$server = new React\Http\Server(
    $loop,
    function (Psr\Http\Message\ServerRequestInterface $request, callable $next) {
        $promise = new React\Promise\Promise(function ($resolve) use ($next, $request) {
            $resolve($next($request));
        });
        return $promise->then(null, function (Exception $e) {
            return new React\Http\Message\Response(
                500,
                array(),
                'Internal error: ' . $e->getMessage()
            );
        });
    },
    function (Psr\Http\Message\ServerRequestInterface $request) {
        if (mt_rand(0, 1) === 1) {
            throw new RuntimeException('Database error');
        }
        return new React\Http\Message\Response(200);
    }
);

Third-Party Middleware

While this project does provide the means to use middleware implementations (see above), it does not aim to define how middleware implementations should look like. We realize that there's a vivid ecosystem of middleware implementations and ongoing effort to standardize interfaces between these with PSR-15 (HTTP Server Request Handlers) and support this goal. As such, this project only bundles a few middleware implementations that are required to match PHP's request behavior (see middleware implementations) and otherwise actively encourages third-party middleware implementations.

While we would love to support PSR-15 directly in

react/http
, we understand that this interface does not specifically target async APIs and as such does not take advantage of promises for deferred responses. The gist of this is that where PSR-15 enforces a PSR-7
ResponseInterface
return value, we also accept a
PromiseInterface
. As such, we suggest using the external PSR-15 middleware adapter that uses on the fly monkey patching of these return values which makes using most PSR-15 middleware possible with this package without any changes required.

Other than that, you can also use the above middleware definition to create custom middleware. A non-exhaustive list of third-party middleware can be found at the middleware wiki. If you build or know a custom middleware, make sure to let the world know and feel free to add it to this list.

API

Browser

The

React\Http\Browser
is responsible for sending HTTP requests to your HTTP server and keeps track of pending incoming HTTP responses. It also registers everything with the main
EventLoop
.
$loop = React\EventLoop\Factory::create();

$browser = new React\Http\Browser($loop);

If you need custom connector settings (DNS resolution, TLS parameters, timeouts, proxy servers etc.), you can explicitly pass a custom instance of the

ConnectorInterface
:

$connector = new React\Socket\Connector($loop, array(
    'dns' => '127.0.0.1',
    'tcp' => array(
        'bindto' => '192.168.10.1:0'
    ),
    'tls' => array(
        'verify_peer' => false,
        'verify_peer_name' => false
    )
));

$browser = new React\Http\Browser($loop, $connector);

Note that the browser class is final and shouldn't be extended, it is likely to be marked final in a future release.

get()

The

get(string $url, array $headers = array()): PromiseInterface
method can be used to send an HTTP GET request.
$browser->get($url)->then(function (Psr\Http\Message\ResponseInterface $response) {
    var_dump((string)$response->getBody());
});

See also GET request client example.

post()

The

post(string $url, array $headers = array(), string|ReadableStreamInterface $contents = ''): PromiseInterface
method can be used to send an HTTP POST request.
$browser->post(
    $url,
    [
        'Content-Type' => 'application/json'
    ],
    json_encode($data)
)->then(function (Psr\Http\Message\ResponseInterface $response) {
    var_dump(json_decode((string)$response->getBody()));
});

See also POST JSON client example.

This method is also commonly used to submit HTML form data:

$data = [
    'user' => 'Alice',
    'password' => 'secret'
];

$browser->post( $url, [ 'Content-Type' => 'application/x-www-form-urlencoded' ], http_build_query($data) );

This method will automatically add a matching

Content-Length
request header if the outgoing request body is a
string
. If you're using a streaming request body (
ReadableStreamInterface
), it will default to using
Transfer-Encoding: chunked
or you have to explicitly pass in a matching
Content-Length
request header like so:
$body = new React\Stream\ThroughStream();
$loop->addTimer(1.0, function () use ($body) {
    $body->end("hello world");
});

$browser->post($url, array('Content-Length' => '11'), $body);

head()

The

head(string $url, array $headers = array()): PromiseInterface
method can be used to send an HTTP HEAD request.
$browser->head($url)->then(function (Psr\Http\Message\ResponseInterface $response) {
    var_dump($response->getHeaders());
});

patch()

The

patch(string $url, array $headers = array(), string|ReadableStreamInterface $contents = ''): PromiseInterface
method can be used to send an HTTP PATCH request.
$browser->patch(
    $url,
    [
        'Content-Type' => 'application/json'
    ],
    json_encode($data)
)->then(function (Psr\Http\Message\ResponseInterface $response) {
    var_dump(json_decode((string)$response->getBody()));
});

This method will automatically add a matching

Content-Length
request header if the outgoing request body is a
string
. If you're using a streaming request body (
ReadableStreamInterface
), it will default to using
Transfer-Encoding: chunked
or you have to explicitly pass in a matching
Content-Length
request header like so:
$body = new React\Stream\ThroughStream();
$loop->addTimer(1.0, function () use ($body) {
    $body->end("hello world");
});

$browser->patch($url, array('Content-Length' => '11'), $body);

put()

The

put(string $url, array $headers = array()): PromiseInterface
method can be used to send an HTTP PUT request.
$browser->put(
    $url,
    [
        'Content-Type' => 'text/xml'
    ],
    $xml->asXML()
)->then(function (Psr\Http\Message\ResponseInterface $response) {
    var_dump((string)$response->getBody());
});

See also PUT XML client example.

This method will automatically add a matching

Content-Length
request header if the outgoing request body is a
string
. If you're using a streaming request body (
ReadableStreamInterface
), it will default to using
Transfer-Encoding: chunked
or you have to explicitly pass in a matching
Content-Length
request header like so:
$body = new React\Stream\ThroughStream();
$loop->addTimer(1.0, function () use ($body) {
    $body->end("hello world");
});

$browser->put($url, array('Content-Length' => '11'), $body);

delete()

The

delete(string $url, array $headers = array()): PromiseInterface
method can be used to send an HTTP DELETE request.
$browser->delete($url)->then(function (Psr\Http\Message\ResponseInterface $response) {
    var_dump((string)$response->getBody());
});

request()

The

request(string $method, string $url, array $headers = array(), string|ReadableStreamInterface $body = ''): PromiseInterface
method can be used to send an arbitrary HTTP request.

The preferred way to send an HTTP request is by using the above request methods, for example the

get()
method to send an HTTP

GET
request.

As an alternative, if you want to use a custom HTTP request method, you can use this method:

$browser->request('OPTIONS', $url)->then(function (Psr\Http\Message\ResponseInterface $response) {
    var_dump((string)$response->getBody());
});

This method will automatically add a matching

Content-Length
request header if the size of the outgoing request body is known and non-empty. For an empty request body, if will only include a
Content-Length: 0
request header if the request method usually expects a request body (only applies to
POST
,
PUT
and
PATCH
).

If you're using a streaming request body (

ReadableStreamInterface
), it will default to using
Transfer-Encoding: chunked
or you have to explicitly pass in a matching
Content-Length
request header like so:
$body = new React\Stream\ThroughStream();
$loop->addTimer(1.0, function () use ($body) {
    $body->end("hello world");
});

$browser->request('POST', $url, array('Content-Length' => '11'), $body);

requestStreaming()

The

requestStreaming(string $method, string $url, array $headers = array(), string|ReadableStreamInterface $body = ''): PromiseInterface
method can be used to send an arbitrary HTTP request and receive a streaming response without buffering the response body.

The preferred way to send an HTTP request is by using the above request methods, for example the

get()
method to send an HTTP

GET
request. Each of these methods will buffer the whole response body in memory by default. This is easy to get started and works reasonably well for smaller responses.

In some situations, it's a better idea to use a streaming approach, where only small chunks have to be kept in memory. You can use this method to send an arbitrary HTTP request and receive a streaming response. It uses the same HTTP message API, but does not buffer the response body in memory. It only processes the response body in small chunks as data is received and forwards this data through ReactPHP's Stream API. This works for (any number of) responses of arbitrary sizes.

$browser->requestStreaming('GET', $url)->then(function (Psr\Http\Message\ResponseInterface $response) {
    $body = $response->getBody();
    assert($body instanceof Psr\Http\Message\StreamInterface);
    assert($body instanceof React\Stream\ReadableStreamInterface);

$body-&gt;on('data', function ($chunk) {
    echo $chunk;
});

$body-&gt;on('error', function (Exception $error) {
    echo 'Error: ' . $error-&gt;getMessage() . PHP_EOL;
});

$body-&gt;on('close', function () {
    echo '[DONE]' . PHP_EOL;
});

});

See also ReactPHP's

ReadableStreamInterface
and the streaming response for more details, examples and possible use-cases.

This method will automatically add a matching

Content-Length
request header if the size of the outgoing request body is known and non-empty. For an empty request body, if will only include a
Content-Length: 0
request header if the request method usually expects a request body (only applies to
POST
,
PUT
and
PATCH
).

If you're using a streaming request body (

ReadableStreamInterface
), it will default to using
Transfer-Encoding: chunked
or you have to explicitly pass in a matching
Content-Length
request header like so:
$body = new React\Stream\ThroughStream();
$loop->addTimer(1.0, function () use ($body) {
    $body->end("hello world");
});

$browser->requestStreaming('POST', $url, array('Content-Length' => '11'), $body);

withTimeout()

The

withTimeout(bool|number $timeout): Browser
method can be used to change the maximum timeout used for waiting for pending requests.

You can pass in the number of seconds to use as a new timeout value:

$browser = $browser->withTimeout(10.0);

You can pass in a bool

false
to disable any timeouts. In this case, requests can stay pending forever:
$browser = $browser->withTimeout(false);

You can pass in a bool

true
to re-enable default timeout handling. This will respects PHP's
default_socket_timeout
setting (default 60s):
$browser = $browser->withTimeout(true);

See also timeouts for more details about timeout handling.

Notice that the

Browser
is an immutable object, i.e. this method actually returns a new
Browser
instance with the given timeout value applied.

withFollowRedirects()

The

withTimeout(bool|int $followRedirects): Browser
method can be used to change how HTTP redirects will be followed.

You can pass in the maximum number of redirects to follow:

$browser = $browser->withFollowRedirects(5);

The request will automatically be rejected when the number of redirects is exceeded. You can pass in a

0
to reject the request for any redirects encountered:
$browser = $browser->withFollowRedirects(0);

$browser->get($url)->then(function (Psr\Http\Message\ResponseInterface $response) { // only non-redirected responses will now end up here var_dump($response->getHeaders()); });

You can pass in a bool

false
to disable following any redirects. In this case, requests will resolve with the redirection response instead of following the
Location
response header:
$browser = $browser->withFollowRedirects(false);

$browser->get($url)->then(function (Psr\Http\Message\ResponseInterface $response) { // any redirects will now end up here var_dump($response->getHeaderLine('Location')); });

You can pass in a bool

true
to re-enable default redirect handling. This defaults to following a maximum of 10 redirects:
$browser = $browser->withFollowRedirects(true);

See also redirects for more details about redirect handling.

Notice that the

Browser
is an immutable object, i.e. this method actually returns a new
Browser
instance with the given redirect setting applied.

withRejectErrorResponse()

The

withRejectErrorResponse(bool $obeySuccessCode): Browser
method can be used to change whether non-successful HTTP response status codes (4xx and 5xx) will be rejected.

You can pass in a bool

false
to disable rejecting incoming responses that use a 4xx or 5xx response status code. In this case, requests will resolve with the response message indicating an error condition:
$browser = $browser->withRejectErrorResponse(false);

$browser->get($url)->then(function (Psr\Http\Message\ResponseInterface $response) { // any HTTP response will now end up here var_dump($response->getStatusCode(), $response->getReasonPhrase()); });

You can pass in a bool

true
to re-enable default status code handling. This defaults to rejecting any response status codes in the 4xx or 5xx range with a
ResponseException
:
$browser = $browser->withRejectErrorResponse(true);

$browser->get($url)->then(function (Psr\Http\Message\ResponseInterface $response) { // any successful HTTP response will now end up here var_dump($response->getStatusCode(), $response->getReasonPhrase()); }, function (Exception $e) { if ($e instanceof React\Http\Message\ResponseException) { // any HTTP response error message will now end up here $response = $e->getResponse(); var_dump($response->getStatusCode(), $response->getReasonPhrase()); } else { var_dump($e->getMessage()); } });

Notice that the

Browser
is an immutable object, i.e. this method actually returns a new
Browser
instance with the given setting applied.

withBase()

The

withBase(string|null $baseUrl): Browser
method can be used to change the base URL used to resolve relative URLs to.

If you configure a base URL, any requests to relative URLs will be processed by first resolving this relative to the given absolute base URL. This supports resolving relative path references (like

../
etc.). This is particularly useful for (RESTful) API calls where all endpoints (URLs) are located under a common base URL.
$browser = $browser->withBase('http://api.example.com/v3/');

// will request http://api.example.com/v3/users $browser->get('users')->then(…);

You can pass in a

null
base URL to return a new instance that does not use a base URL:
$browser = $browser->withBase(null);

Accordingly, any requests using relative URLs to a browser that does not use a base URL can not be completed and will be rejected without sending a request.

This method will throw an

InvalidArgumentException
if the given
$baseUrl
argument is not a valid URL.

Notice that the

Browser
is an immutable object, i.e. the

withBase()
method actually returns a new
Browser
instance with the given base URL applied.

withProtocolVersion()

The

withProtocolVersion(string $protocolVersion): Browser
method can be used to change the HTTP protocol version that will be used for all subsequent requests.

All the above request methods default to sending requests as HTTP/1.1. This is the preferred HTTP protocol version which also provides decent backwards-compatibility with legacy HTTP/1.0 servers. As such, there should rarely be a need to explicitly change this protocol version.

If you want to explicitly use the legacy HTTP/1.0 protocol version, you can use this method:

$browser = $browser->withProtocolVersion('1.0');

$browser->get($url)->then(…);

Notice that the

Browser
is an immutable object, i.e. this method actually returns a new
Browser
instance with the new protocol version applied.

withResponseBuffer()

The

withResponseBuffer(int $maximumSize): Browser
method can be used to change the maximum size for buffering a response body.

The preferred way to send an HTTP request is by using the above request methods, for example the

get()
method to send an HTTP

GET
request. Each of these methods will buffer the whole response body in memory by default. This is easy to get started and works reasonably well for smaller responses.

By default, the response body buffer will be limited to 16 MiB. If the response body exceeds this maximum size, the request will be rejected.

You can pass in the maximum number of bytes to buffer:

$browser = $browser->withResponseBuffer(1024 * 1024);

$browser->get($url)->then(function (Psr\Http\Message\ResponseInterface $response) { // response body will not exceed 1 MiB var_dump($response->getHeaders(), (string) $response->getBody()); });

Note that the response body buffer has to be kept in memory for each pending request until its transfer is completed and it will only be freed after a pending request is fulfilled. As such, increasing this maximum buffer size to allow larger response bodies is usually not recommended. Instead, you can use the

requestStreaming()
method to receive responses with arbitrary sizes without buffering. Accordingly, this maximum buffer size setting has no effect on streaming responses.

Notice that the

Browser
is an immutable object, i.e. this method actually returns a new
Browser
instance with the given setting applied.

React\Http\Message

Response

The

React\Http\Message\Response
class can be used to represent an outgoing server response message.
$response = new React\Http\Message\Response(
    200,
    array(
        'Content-Type' => 'text/html'
    ),
    "Hello world!\n"
);

This class implements the PSR-7

ResponseInterface
which in turn extends the PSR-7
MessageInterface
.

Internally, this implementation builds on top of an existing incoming response message and only adds required streaming support. This base class is considered an implementation detail that may change in the future.

ServerRequest

The

React\Http\Message\ServerRequest
class can be used to respresent an incoming server request message.

This class implements the PSR-7

ServerRequestInterface
which extends the PSR-7
RequestInterface
which in turn extends the PSR-7
MessageInterface
.

This is mostly used internally to represent each incoming request message. Likewise, you can also use this class in test cases to test how your web application reacts to certain HTTP requests.

Internally, this implementation builds on top of an existing outgoing request message and only adds required server methods. This base class is considered an implementation detail that may change in the future.

ResponseException

The

React\Http\Message\ResponseException
is an
Exception
sub-class that will be used to reject a request promise if the remote server returns a non-success status code (anything but 2xx or 3xx). You can control this behavior via the
withRejectErrorResponse()
method
.

The

getCode(): int
method can be used to return the HTTP response status code.

The

getResponse(): ResponseInterface
method can be used to access its underlying response object.

React\Http\Middleware

StreamingRequestMiddleware

The

React\Http\Middleware\StreamingRequestMiddleware
can be used to process incoming requests with a streaming request body (without buffering).

This allows you to process requests of any size without buffering the request body in memory. Instead, it will represent the request body as a

ReadableStreamInterface
that emit chunks of incoming data as it is received:

$server = new React\Http\Server(array(
    new React\Http\Middleware\StreamingRequestMiddleware(),
    function (Psr\Http\Message\ServerRequestInterface $request) {
        $body = $request->getBody();
        assert($body instanceof Psr\Http\Message\StreamInterface);
        assert($body instanceof React\Stream\ReadableStreamInterface);

    return new React\Promise\Promise(function ($resolve) use ($body) {
        $bytes = 0;
        $body-&gt;on('data', function ($chunk) use (&amp;$bytes) {
            $bytes += \count($chunk);
        });
        $body-&gt;on('close', function () use (&amp;$bytes, $resolve) {
            $resolve(new React\Http\Message\Response(
                200,
                [],
                "Received $bytes bytes\n"
            ));
        });
    });
}

));

See also streaming incoming request for more details.

Additionally, this middleware can be used in combination with the

LimitConcurrentRequestsMiddleware
and
RequestBodyBufferMiddleware
(see below) to explicitly configure the total number of requests that can be handled at once:

$server = new React\Http\Server(array(
    new React\Http\Middleware\StreamingRequestMiddleware(),
    new React\Http\Middleware\LimitConcurrentRequestsMiddleware(100), // 100 concurrent buffering handlers
    new React\Http\Middleware\RequestBodyBufferMiddleware(2 * 1024 * 1024), // 2 MiB per request
    new React\Http\Middleware\RequestBodyParserMiddleware(),
    $handler
));

Internally, this class is used as a "marker" to not trigger the default request buffering behavior in the

Server
. It does not implement any logic on its own.

LimitConcurrentRequestsMiddleware

The

React\Http\Middleware\LimitConcurrentRequestsMiddleware
can be used to limit how many next handlers can be executed concurrently.

If this middleware is invoked, it will check if the number of pending handlers is below the allowed limit and then simply invoke the next handler and it will return whatever the next handler returns (or throws).

If the number of pending handlers exceeds the allowed limit, the request will be queued (and its streaming body will be paused) and it will return a pending promise. Once a pending handler returns (or throws), it will pick the oldest request from this queue and invokes the next handler (and its streaming body will be resumed).

The following example shows how this middleware can be used to ensure no more than 10 handlers will be invoked at once:

$server = new React\Http\Server(
    $loop,
    new React\Http\Middleware\LimitConcurrentRequestsMiddleware(10),
    $handler
);

Similarly, this middleware is often used in combination with the

RequestBodyBufferMiddleware
(see below) to limit the total number of requests that can be buffered at once:

$server = new React\Http\Server(
    $loop,
    new React\Http\Middleware\StreamingRequestMiddleware(),
    new React\Http\Middleware\LimitConcurrentRequestsMiddleware(100), // 100 concurrent buffering handlers
    new React\Http\Middleware\RequestBodyBufferMiddleware(2 * 1024 * 1024), // 2 MiB per request
    new React\Http\Middleware\RequestBodyParserMiddleware(),
    $handler
);

More sophisticated examples include limiting the total number of requests that can be buffered at once and then ensure the actual request handler only processes one request after another without any concurrency:

$server = new React\Http\Server(
    $loop,
    new React\Http\Middleware\StreamingRequestMiddleware(),
    new React\Http\Middleware\LimitConcurrentRequestsMiddleware(100), // 100 concurrent buffering handlers
    new React\Http\Middleware\RequestBodyBufferMiddleware(2 * 1024 * 1024), // 2 MiB per request
    new React\Http\Middleware\RequestBodyParserMiddleware(),
    new React\Http\Middleware\LimitConcurrentRequestsMiddleware(1), // only execute 1 handler (no concurrency)
    $handler
);

RequestBodyBufferMiddleware

One of the built-in middleware is the

React\Http\Middleware\RequestBodyBufferMiddleware
which can be used to buffer the whole incoming request body in memory. This can be useful if full PSR-7 compatibility is needed for the request handler and the default streaming request body handling is not needed. The constructor accepts one optional argument, the maximum request body size. When one isn't provided it will use
post_max_size
(default 8 MiB) from PHP's configuration. (Note that the value from your matching SAPI will be used, which is the CLI configuration in most cases.)

Any incoming request that has a request body that exceeds this limit will be accepted, but its request body will be discarded (empty request body). This is done in order to avoid having to keep an incoming request with an excessive size (for example, think of a 2 GB file upload) in memory. This allows the next middleware handler to still handle this request, but it will see an empty request body. This is similar to PHP's default behavior, where the body will not be parsed if this limit is exceeded. However, unlike PHP's default behavior, the raw request body is not available via

php://input
.

The

RequestBodyBufferMiddleware
will buffer requests with bodies of known size (i.e. with
Content-Length
header specified) as well as requests with bodies of unknown size (i.e. with
Transfer-Encoding: chunked
header).

All requests will be buffered in memory until the request body end has been reached and then call the next middleware handler with the complete, buffered request. Similarly, this will immediately invoke the next middleware handler for requests that have an empty request body (such as a simple

GET
request) and requests that are already buffered (such as due to another middleware).

Note that the given buffer size limit is applied to each request individually. This means that if you allow a 2 MiB limit and then receive 1000 concurrent requests, up to 2000 MiB may be allocated for these buffers alone. As such, it's highly recommended to use this along with the

LimitConcurrentRequestsMiddleware
(see above) to limit the total number of concurrent requests.

Usage:

$server = new React\Http\Server(
    $loop,
    new React\Http\Middleware\StreamingRequestMiddleware(),
    new React\Http\Middleware\LimitConcurrentRequestsMiddleware(100), // 100 concurrent buffering handlers
    new React\Http\Middleware\RequestBodyBufferMiddleware(16 * 1024 * 1024), // 16 MiB
    function (Psr\Http\Message\ServerRequestInterface $request) {
        // The body from $request->getBody() is now fully available without the need to stream it 
        return new React\Http\Message\Response(200);
    },
);

RequestBodyParserMiddleware

The

React\Http\Middleware\RequestBodyParserMiddleware
takes a fully buffered request body (generally from
RequestBodyBufferMiddleware
), and parses the form values and file uploads from the incoming HTTP request body.

This middleware handler takes care of applying values from HTTP requests that use

Content-Type: application/x-www-form-urlencoded
or
Content-Type: multipart/form-data
to resemble PHP's default superglobals
$_POST
and
$_FILES
. Instead of relying on these superglobals, you can use the
$request->getParsedBody()
and
$request->getUploadedFiles()
methods as defined by PSR-7.

Accordingly, each file upload will be represented as instance implementing the PSR-7

UploadedFileInterface
. Due to its blocking nature, the

moveTo()
method is not available and throws a
RuntimeException
instead. You can use
$contents = (string)$file->getStream();
to access the file contents and persist this to your favorite data store.
$handler = function (Psr\Http\Message\ServerRequestInterface $request) {
    // If any, parsed form fields are now available from $request->getParsedBody()
    $body = $request->getParsedBody();
    $name = isset($body['name']) ? $body['name'] : 'unnamed';

$files = $request-&gt;getUploadedFiles();
$avatar = isset($files['avatar']) ? $files['avatar'] : null;
if ($avatar instanceof Psr\Http\Message\UploadedFileInterface) {
    if ($avatar-&gt;getError() === UPLOAD_ERR_OK) {
        $uploaded = $avatar-&gt;getSize() . ' bytes';
    } elseif ($avatar-&gt;getError() === UPLOAD_ERR_INI_SIZE) {
        $uploaded = 'file too large';
    } else {
        $uploaded = 'with error';
    }
} else {
    $uploaded = 'nothing';
}

return new React\Http\Message\Response(
    200,
    array(
        'Content-Type' =&gt; 'text/plain'
    ),
    $name . ' uploaded ' . $uploaded
);

};

$server = new React\Http\Server( $loop, new React\Http\Middleware\StreamingRequestMiddleware(), new React\Http\Middleware\LimitConcurrentRequestsMiddleware(100), // 100 concurrent buffering handlers new React\Http\Middleware\RequestBodyBufferMiddleware(16 * 1024 * 1024), // 16 MiB new React\Http\Middleware\RequestBodyParserMiddleware(), $handler );

See also form upload server example for more details.

By default, this middleware respects the

upload_max_filesize
(default

2M
) ini setting. Files that exceed this limit will be rejected with an
UPLOAD_ERR_INI_SIZE
error. You can control the maximum filesize for each individual file upload by explicitly passing the maximum filesize in bytes as the first parameter to the constructor like this:
new React\Http\Middleware\RequestBodyParserMiddleware(8 * 1024 * 1024); // 8 MiB limit per file

By default, this middleware respects the

file_uploads
(default

1
) and
max_file_uploads
(default
20
) ini settings. These settings control if any and how many files can be uploaded in a single request. If you upload more files in a single request, additional files will be ignored and the
getUploadedFiles()
method returns a truncated array. Note that upload fields left blank on submission do not count towards this limit. You can control the maximum number of file uploads per request by explicitly passing the second parameter to the constructor like this:
new React\Http\Middleware\RequestBodyParserMiddleware(10 * 1024, 100); // 100 files with 10 KiB each

Note that this middleware handler simply parses everything that is already buffered in the request body. It is imperative that the request body is buffered by a prior middleware handler as given in the example above. This previous middleware handler is also responsible for rejecting incoming requests that exceed allowed message sizes (such as big file uploads). The

RequestBodyBufferMiddleware
used above simply discards excessive request bodies, resulting in an empty body. If you use this middleware without buffering first, it will try to parse an empty (streaming) body and may thus assume an empty data structure. See also
RequestBodyBufferMiddleware
for more details.

PHP's

MAX_FILE_SIZE
hidden field is respected by this middleware. Files that exceed this limit will be rejected with an
UPLOAD_ERR_FORM_SIZE
error.

This middleware respects the

max_input_vars
(default

1000
) and
max_input_nesting_level
(default
64
) ini settings.

Note that this middleware ignores the

enable_post_data_reading
(default

1
) ini setting because it makes little sense to respect here and is left up to higher-level implementations. If you want to respect this setting, you have to check its value and effectively avoid using this middleware entirely.

Install

The recommended way to install this library is through Composer. New to Composer?

This project follows SemVer. This will install the latest supported version:

$ composer require react/http:^1.1

See also the CHANGELOG for details about version upgrades.

This project aims to run on any platform and thus does not require any PHP extensions and supports running on legacy PHP 5.3 through current PHP 7+ and HHVM. It's highly recommended to use PHP 7+ for this project.

Tests

To run the test suite, you first need to clone this repo and then install all dependencies through Composer:

$ composer install

To run the test suite, go to the project root and run:

$ php vendor/bin/phpunit

The test suite also contains a number of functional integration tests that rely on a stable internet connection. If you do not want to run these, they can simply be skipped like this:

$ php vendor/bin/phpunit --exclude-group internet

License

MIT, see LICENSE file.

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.