PHP HttpClient

PHP-HttpClient

Veröffentlicht 

von wappler

 

am 24. Dezember 2007

 

in PHP

.Tags:php.

转自http://www.wappler.eu/php-httpclient/

PHP-HttpClient based on Simon Willisons HttpClient Version 0.9.

HttpClient is a client class for the HTTP protocol. It can be used to interact with another web server from within a PHP script. As well as retrieving information from a server, HttpClient can interact with a server via POST or GET. It can therefore be used as part of any script that needs to communicate with an application running on another site.

Zu Deutsch: PHP-HttpClient ist eine Klasse, mit der man einen Browser in PHP simulieren kann. So kann man Inhalte von anderen Webseiten holen und in die eigene Webseite einbinden. Ich musste die Version von Simon Willison ein wenig erweitern, um mehrere Dateien per POST übertragen zu können.

Features

• HTTP 1.0
• Supports Cookies
• Custom User-Agent
• Can automatically handle redirections
• GET, POST and multiple File-Uploads via POST
• Debugging
• Supports password-protected sites

Download

Download HttpClient.class.php

Usage

Grabbing an HTML page (static method)

$pageContents = HttpClient::quickGet('http://example.com/');

Posting a form and grabbing the response (static method)

$pageContents = HttpClient::quickPost('http://example.com/someForm', array(
'name' => 'Some Name',
'email' => '[e-mail]'
));

The static methods are easy to use, but seriously limit the functionality of the class as you cannot access returned headers or use facilities such as cookies or authentication.
A simple GET request using the class

$client = new HttpClient('example.com');
if (!$client->get('/')) {
die('An error occurred: '.$client->getError());
}
$pageContents = $client->getContent();

A GET request with debugging turned on

$client = new HttpClient('example.com');
$client->setDebug(true);
if (!$client->get('/')) {
die('An error occurred: '.$client->getError());
}
$pageContents = $client->getContent();

A GET request demonstrating automatic redirection

$client = new HttpClient('www.amazon.com');
$client->setDebug(true);
if (!$client->get('/')) {
die('An error occurred: '.$client->getError());
}
$pageContents = $client->getContent();

Check to see if a page exists

$client = new HttpClient('example.com');
$client->setDebug(true);
if (!$client->get('/thispagedoesnotexist')) {
die('An error occurred: '.$client->getError());
}
if ($client->getStatus() == '404') {
echo 'Page does not exist!';
}
$pageContents = $client->getContent();

Fake the User Agent string

$client = new HttpClient('example.com');
$client->setDebug(true);
$client->setUserAgent('Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.3a) Gecko/20021207');
if (!$client->get('/')) {
die('An error occurred: '.$client->getError());
}
$pageContents = $client->getContent();

Log in to a site via a login form, then request a page

In this example, it is assumed that correctly logging in will result in the server sending a sesssion cookie of some sort. This cookie is resent by the HttpClient class automatically, so a request to a page that requires users to be logged in will now work.

$client = new HttpClient('example.com');
$client->post('/login.php', array(
'username' => 'Simon',
'password' => 'ducks'
));
if (!$client->get('/private.php')) {
die('An error occurred: '.$client->getError());
}
$pageContents = $client->getContent();

Using HTTP authorisation

Note that the class uses the American spelling ‘authorization’ to fit with the HTTP specification.

$client = new HttpClient('example.com');
$client->setAuthorization('Username', 'Password');
if (!$client->get('/')) {
die('An error occurred: '.$client->getError());
}
$pageContents = $client->getContent();

Print out the headers from a response

$client = new HttpClient('example.com');
if (!$client->get('/')) {
die('An error occurred: '.$client->getError());
}
print_r($client->getHeaders());

Setting the maximum number of redirects

$client = new HttpClient('www.amazon.com');
$client->setDebug(true);
$client->setMaxRedirects(3);
$client->get('/');