How PHP Runs: Requests and Responses

Every order at a café is a fresh start. The barista makes your drink, hands it over, and forgets you. PHP works the same way.

In Mixing PHP and HTML, you saw PHP fill values into a page. But when exactly does that code run, and what happens to it afterward? Knowing this answers a lot of beginner puzzles, like "why did my variable forget its value?"

The café counter

Picture a busy café:

1. A customer walks up and orders: "A large hot chocolate, please."
2. The barista makes that one drink, following the recipe.
3. The barista hands it over, and the customer leaves.
4. The barista moves on to the next customer, with a clean counter, remembering nothing about the last one.

The web works like that:

1. The browser sends a request: "Please send me /menu.php."
2. The web server hands the request to PHP, which runs menu.php from top to bottom.
3. Everything the script sends out (with echo, or plain HTML outside the PHP tags) becomes the response, which goes back to the browser.
4. PHP forgets everything. All the script's variables are thrown away.

That cycle, one request in and one response out, happens for every page, every click, and every form someone submits.

What's in a request

A request is a small message with a few important parts:

• A method: usually GET ("send me this page") or POST ("here's some data I'm sending you," like a submitted form).
• A path: which page, like /menu.php.
• Sometimes a query string: extra details after a ? in the address, like /menu.php?size=large.

PHP puts these details in special variables for you. $_SERVER holds information about the request, and $_GET holds the query string's values:

<?php
// Visiting /menu.php?size=large in a browser:
echo $_SERVER["REQUEST_METHOD"]; // GET
echo $_GET["size"];             // large

You'll use these properly in Handling Forms: GET and POST.

What's in a response

The response is mostly the text your script produces: an HTML page. Along with it go a few headers, small labels about the response, like what kind of content it is. The most important is the status code, a number that says how it went:

Code	Means
200	OK, here's your page
404	Not found: there's no page at that address
500	Server error: something went wrong running the code

If your PHP script stops with a fatal error, the visitor may just see a blank page, or a 500 error, while the details are in the error message or the server's log. That's why the development settings you chose in Setting Up show every error: you want to see them while you're learning.

Fresh start, every time

Because PHP forgets everything after each request, a variable can't remember anything from one visit to the next. Consider this visit counter:

<?php
$visits = 0;
$visits = $visits + 1;
echo "You have visited this page $visits time(s).\n";

You have visited this page 1 time(s).

Refresh the page 10 times, and it says 1 every time. Each request starts the script from the top, so $visits starts at 0 again.

This isn't a flaw. It keeps each visitor's page separate, and it lets a server handle thousands of requests without them getting mixed up. When you do need to remember something, you store it somewhere that lasts:

• In a session or a cookie, to remember one visitor between pages, like "Maria is logged in." See Sessions and Cookies.
• In a file or a database, to remember things for everyone, like every sign-up. See Reading and Writing Files and Databases with PDO.

Two ways to run the same script

In this track, you run most scripts in the terminal with php index.php. On a website, a web server runs them for each request. It's the same PHP language, but a few things differ:

	Terminal (php index.php)	Web (php -S, or a real server)
Who asks	You, typing a command	A browser, sending a request
Where output goes	Your terminal window	The browser, as the page
Request details ($_GET, forms)	Empty: there's no request	Filled from the request
Line breaks (\n)	Start a new line	Only break the HTML source

PHP can even tell you which way it's running. php_sapi_name() returns "cli" in the terminal (command-line interface). With the built-in server, it returns "cli-server".

Try it

This script describes how it's being run. Predict what it prints when you run it in the terminal with php index.php.

Running PHP here

Running PHP right in the browser is coming to Syntaxia soon. For now, you can read and edit the code here. If you have PHP installed on your computer (see Setting Up), save it as index.php and run it with php index.php.

Check your prediction

Running as: cli
Counter: 2
Query values received: 0
No browser here, so there's no request.

$counter++ adds 1, so it ends at 2, and it'll be 2 on every run, never 4 or 6, because each run starts fresh. In the terminal, there's no request, so $_GET is empty. Run it through the built-in server instead, and the first line says cli-server, and the last line changes too.

Try it yourself

1. Save the Try it script, start the built-in server with php -S localhost:8000, and open http://localhost:8000/?a=1&b=2. What does the query values line say now?
2. Refresh the page a few times. Does the counter ever go above 2? Why not?
3. Open http://localhost:8000/nothing-here.php, a page that doesn't exist. What does the browser show, and which status code do you think that is?

Check your understanding

What happens to a PHP script's variables after it sends its response?

They are thrown away; the next request starts fresh

They are saved for the next visitor

They are sent to the browser

They are saved in a file automatically

What does the status code 404 mean?

The page loaded fine

The server crashed

There is no page at that address

The form was submitted

In the address /menu.php?size=large, where does PHP put size?

In $size automatically

In $_GET["size"]

In $_SERVER["size"]

Nowhere, PHP ignores it

Up next

You know how PHP runs. Now let's look closely at the values it works with, starting with Variables and Constants.