Web servers and browsers talk to each other using HTTP (Hypertext Transfer Protocol), a set of rules that govern how to request and retrieve data from a Web server. Most of the time, you don ’ t need to involve into the workings of HTTP, because HTTP communication happens automatically whenever a visitor visits your Web page or PHP script.
However, occasionally it can be useful to understand some of the processes involved in HTTP, because PHP lets you have some control over these processes.
For example, if you understand how HTTP response headers work, you can use your PHP script to create your own custom response headers allowing the script to display an image, or redirect the browser to a new page.
Understanding HTTP Requests
Whenever a browser wants to display a Web page or other resource (such as an image file) that is stored on a Web server, the browser first connects to the server (usually via port 80, the HTTP port), and then sends various pieces of information to the server, known as the request message . The request message consists of the following sections, in order:
The request line:
This tells the Web server which resource (URL) the browser wants to retrieve a list of HTTP headers: These optional lines of text allow the browser to send additional information to the server, such as cookies and which character sets the browser can accept an empty line:
This is required after the request line and any headers An optional message body: This might contain, for example, form data sent via the POST method.
Each line in the message must end with a carriage return character followed by a line feed character. The request line is the most important part of the request, because it tells the server which resource to send back to the browser. Here ’ s a typical request line:
GET /about/index.php HTTP/1.1
The request line consists of three parts:
- The request method ( GET in this case)
- the URL to retrieve (/about/index.php ), and the version of HTTP to use (most modern browsers work with HTTP/1.1 ).
- Other request methods include POST (for sending large amounts of form data) and HEAD (similar to GET but asks the server to return just the response headers, rather than the actual content).
Note : Many HTTP request headers can be sent from a browser to a server.
So a complete browser request (assuming there ’ s no request body) might look like this:
GET /about/index.php HTTP/1.1
Host: www.example.com
Accept: text/html, application/xml
Accept-Charset: ISO-8859-1,utf-8
Accept-Encoding: gzip,deflate
Accept-Language: en-gb,en
Cookie: name=fred
Referer: www.example.com
User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-GB;
rv:1.9.0.5) Gecko/2008120121 Firefox/3.0.5
Exploring HTTP Responses
When the Web server receives an HTTP request from the browser, it sends back an HTTP response.
Typically this includes the content that the browser requested, along with some information about the content, but it may instead return an error (if the content couldn ’ t be found, for example).
As with requests, a response typically contains up to four sections:
- The status line: This tells the Web browser the status of the request.
- A list of HTTP headers: These optional headers contain extra information about the response,
such as the type and length of the returned content3 - An empty line: This is required after the request line and any headers
- An optional message body: Usually this contains the returned content, such as the Web page
markup or encoded image data.
An HTTP status line consists of a status code and a corresponding reason phrase (that is, a plain English version of the status code). Here ’ s an example response from a Web server after a browser has requested an HTML page:
HTTP/1.x 200 OK Date: Mon, 05 Jan 2009 10:19:52 GMT Server: Apache/2.0.59 (Unix) PHP/5.2.5 DAV/2 X-Powered-By: PHP/5.2.5 Content-Length: 395 Keep-Alive: timeout=15, max=96 Connection: Keep-Alive Content-Type: text/html < !DOCTYPE html PUBLIC “-//W3C//DTD XHTML 1.0 Strict//EN” “http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd” > < html xmlns=”http://www.w3.org/1999/xhtml” xml:lang=”en” lang=”en” > < head > < title > About Us < /title > < link rel=”stylesheet” type=”text/css” href=”common.css” / > < /head > < body > < h1 > About Us < /h1 > < p > We specialize in widgets for all occasions. < /p > < /body > < /html
Notice that the response consists of the status line, followed by various response headers, followed by a blank line and, finally, the requested content (in this case a Web page).
Once the browser receives a Web page, it usually makes additional requests for any other resources referenced in the page, such as the common.css style sheet file in this example, or any images embedded in the page.
Therefore when a visitor views a Web page, several HTTP requests and responses may be initiated.
Modifying an HTTP Response
Because the PHP engine interacts with the Web server, your PHP scripts can influence the HTTP response headers sent by the server. This can be very useful.
To get the Web server to send a custom HTTP header as part of its response, you use PHP ’ s header() function. This simply takes the header line to output, then injects this line into the response headers:
header( “Server: Never you mind” );
By default, header() replaces any HTTP header field with the same name. In the example just shown, if the response already contains a Server header, that header is replaced with the one passed into header() .
However, some HTTP header fields can be included more than once in the response. If you ’ d like to include the same header several times, pass in false as the second argument to header() :
header( “Set-Cookie: name=Fred; expires=Mon, 05-Jan-2009 10:22:21 GMT; path=/; domain=.example.com”); header( “Set-Cookie: age=33; expires=Mon, 05-Jan-2009 10:22:21 GMT; path=/; domain=.example.com”, false );
Generally speaking, when you pass a header line to header() , PHP faithfully injects the header line
as – is into the response. However, there are two special cases:
1. If the header string starts with HTTP/ , PHP assumes you want to set the status line, rather than
add or replace a header line. This allows you to set your own HTTP status lines:
// Nothing to see here, move along
header( “HTTP/1.1 404 Not Found” );
2. If you pass in a Location header string, PHP automatically sends a 302 Found status line as
well as the Location header:
// Redirect to the login page
header( “Location: http://www.example.com/login.php” );
3. This makes it easy to do page redirects (as you saw in Chapter 9). If you ’ d rather send a different
status line, simply specify the status line as well:
header( “HTTP/1.1 301 Moved Permanently” );
header( “Location: http://www.example.com/newpage.php” );
header() is very useful if your PHP script needs to send anything other than an HTML Web page. For
example, say you have a report.pdf file on the Web server, and you want to send this to the browser.
You could write the following:
< ?php
header( “Content-Type: application/pdf” );
readfile( “report.pdf” );
? >
The first line tells the Web browser to expect a PDF document rather than a regular Web page. The
second line reads the PDF file on the server ’ s hard drive and outputs its contents to the Web browser, which can then save or display the PDF.
Usually it ’ s up to the browser as to whether it displays the file in the browser itself, or offers to save it to the user ’ s hard disk. You can use a Content – Disposition: Attachment header to suggest to the browser that the file should be saved rather than displayed and, optionally, to suggest a filename for the saved file:
< ?php header( “Content-Type: application/pdf” ); header( ‘Content-Disposition: attachment; filename=”Latest Report.pdf”’ ); readfile( “report.pdf” ); ? >
By the way, you need to make sure you don ’ t send anything to the browser before calling header() .
This includes HTML markup, or even blank lines, before your opening < ?php tag.
This is because, once PHP has received a request to send some content to the browser, it sends the HTTP headers first (because the headers need to be sent at the start of the response).
Therefore, by the time your header() call is executed, the content is already being sent, and it ’ s too late to send any more headers. (If you fall foul of this, then your header isn ’ t sent and PHP generates a Cannot modify header information -headers already sent warning.)
Getting Information from the Web Server
Each time a PHP script is run, the Web server software makes a wealth of useful information available to the PHP engine. Such information includes details about the server itself, as well as details of the script being executed.
You can access all of this information in your PHP script through the $_SERVER superglobal array. For example, to display the IP address of the visitor ’ s computer (or proxy server) you might use:
echo “Your IP address is: “ . $_SERVER[“REMOTE_ADDR”];
Because Web servers come in all shapes and sizes, the information available depends very much on your particular server setup. Having said that, there ’ s usually a core list of values that are always present.
The following simple script outputs all of the values in the $_SERVER superglobal array:
< !DOCTYPE html PUBLIC “-//W3C//DTD XHTML 1.0 Strict//EN” “http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd” > < html xmlns=”http://www.w3.org/1999/xhtml” xml:lang=”en” lang=”en” > < head > < title > Server and script details < /title > < link rel=”stylesheet” type=”text/css” href=”common.css” / > < /head > < body > < h1 > Server and script details < /h1 > < pre > < ?php print_r( $_SERVER ); ? > < /pre > < /body > < /html >
This is all about the Working with HTTP in PHP.