tools of the trade
Main Page
PHP Editor
Perl Editor
CSS Editor
MySQL Client
Apache GUI
GCC Builder
Join our mailing list
Privacy policy
This is a tutorial on basic PHP written by Martin Geisler. He wrote it because he wanted to give something back to the OpenSource community. PHP (http://www.php.net/) is free, and this is his way of saying thanks!

The tutorial is aimed at users who have no previous experience with programming in PHP or any other language. But there's no guarantee that this goal has been reached. It's very likely that there are places where the user is assumed to know things he or she won't know at this point of learning PHP.

MartinGeisler wrote this tutorial after having used PHP for some time. Because of that, some sections of the tutorial might be difficult to understand --- it's hard to remember afterwards how difficult things were. So if you have any problems with parts of the tutorial, then please mail him and give your advice. Or even better: edit the page directly in your browser. The pages in the tutorial (except this one) can be edited freely so please use this chance to improve the text. The only requirement is that you choose a proper WikiWord as your login name --- no registration is necessary, this is just a way for you to sign your change.


  • Intro gives you a quick introduction to PHP.
  • Installation explains how to install PHP.
  • Action shows you actual PHP code.
  • Variables describes how variables work in PHP.
  • Control explains the control-structures in PHP.
  • Functions tells you how you group your code.

PHP stands for PHP: Hypertext Pre-processor. The funny abbreviation follows the style set by Richard Stallmann when he founded GNU (GNU's not Unix!) As the name says, it's a pre-processor for hypertext, which is just another word for what most people call webpages. Since it's a preprocessor, it runs on the remote web server and processes the webpages before they are sent to the browser. This makes it a so-called server-side scripting language. The fact that it runs on the server has several benefits, and some drawbacks. Lets take the benefits first:
  • On the server you can have access to things like a database. This means that you can make a script that sorts through large amounts of data, without the client having to download them first.
  • It's only the output from the script that is sent to the Client, not the script itself. That means that you can make the script invisible from the end-user. That makes PHP-scripts browser-neutral; they don't depend on some capability of the browser. It also means that you don't have to worry that someone else can steal your carefully crafted script. It's not like when you make a JavaScript --- everybody is able to read it in the source of the webpage. It has to be this way with client-side scripting, or else the client would be unable to get the source of the script, and therefore unable to do any scripting.
  • You can make your own programs for use in your scripts. You could implement part of the script in C, and then call the program from your script to make it run faster. PHP is a parsed language, meaning that there are no compiled binaries. Every time someone requests a page with PHP-code, the parser looks through the page and executes any PHP-statements it might find. Fortunately this is a very fast process, but you might want to speed things up if you have a very complicated script.

    When you make a C-program, you compile the source and then run the resulting binary. This makes PHP slower than an equivalent C-program.

As I said, there are also some drawbacks:

  • By executing everything on the server, you put more strain on it. With many concurrent requests, and large complex scripts, the server might not be able to handle it. But this isn't a real concern because the parser in PHP is so quick. And if your server still can't cope with the number of visitors, then you should be able to generate some revenue from banners on your site, and then pay for a bigger server :-)
  • The pages can't do anything themselves --- you need the server to do the magic. This means that the pages will lose some of their functionality if your visitors decide to save them to their computer.

    You could of course still put some JavaScript in your pages. This is a very powerful combination between server- and client-side scripting. You could use PHP to fetch some values from a database, and then setup your variables in the JavaScript with these values.

In my opinion the benefits clearly outweigh the drawbacks (that's the reason I'm using PHP :-). And I'm not the only one --- an increasing number of websites are using PHP.

Some Background Information

PHP is actually a rather simple language, despite its great powers. It's a very young language, so the developers have had the chance to learn from previous language's mistakes and implement their strengths. Much of the syntax is borrowed from C. This is reflected in the different conditional statements, the loop-structures, the Boolean operators, and the assignment of variables. Since C is probably the most common programming language today, this should make PHP easy to pick up. Even if you don't have any previous experience with C (like me) you should be able to learn it quickly. I certainly have, and now that I have started to look at C, I can see how much of the syntax they have borrowed. I can just apply the techniques I've learned from PHP when I program in C. Of course this is only true as long as you only write simple programs in C, since C is a "real" programming language, suitable for writing operating-systems in or the like.

But another thing that makes PHP easy to learn is it's relaxed way of dealing with the types of variables. Its very simple: you don't have to think of the types of variables at all! If you assign a number to a variable, then it just works. When you later try to output the variable to the browser it also just works. PHP takes care of converting the variable from an integer-type to a string-type, on the fly and automatically. To make matters even simpler, you don't even declare your variables --- you just assign a value to them, and then they are ready. If you are an experienced programmer (and especially if you program in Pascal) you might say: "This can't be real! There's no syntax in this language. When you just declare variables like that you don't have any control --- you don't even know what type a variable is!" All I can say to this is: It really doesn't matter --- most of the time you're interested in the value of the variable, not its type. And if you really want, then you can indeed find out what type a given variable is. It could come in handy if you want to check to see if a parameter to a function really is of a given type.

Because PHP is meant to be used with webpages it has a lot of functions to deal with text. Generally you can say that PHP is specially designed to deal with webpages, and doing so quickly and efficiently. Because of that most of the built-in functions are simple and straightforward to use.

Being web-oriented, PHP also contains all the functions you'll need to do things on the Internet. There are functions for connecting to remote webservers, checking mail via POP3 or IMAP, or url encoding strings to protect special characters.

Together with a good manual, you have all the help you'll need. And if you get stuck --- then you can count on the community. There has been a lot written about PHP on the net (you are reading just such a thing now :-) and there's countless mailings-lists you can subscribe to.

Now I've told you about how easy PHP is, and how many things you can do with it. Let's see it in action then.

Here we'll go through how you would install PHP.

Installing under Linux

The first thing you'll have to do, is to grab the source for PHP from . Then grab the source for the Apache webserver from http://httpd.apache.org/download.cgi

Instructions can be found at http://www.php.net/manual/en/install.unix.php.

Installing under Windows

Instructions can be found at http://www.php.net/manual/en/install.windows.php.

Another possibility is to download and run PHPTriad from http://www.phpgeek.com/. This installs the very popular triad of programs: PHP, MySql and Apache. It should be very cool and very easy.

As in all good tutorials, we start with the (in)famous HelloWorld example. Here's the code:

<title>Hello World</title>
// This is a comment that extends to the end of the line.

/* This is multi-line comment. Comments are ignored by PHP,
   but provide valuable information to people reading the source. */

echo "Hello World!\n";

You should save this piece of code in a file with the extension .php. If that doesn't work, then try other extensions like .php3 or .phtml. If you can't figure out the right extension, then ask whoever installed the webserver. If that's you, then have a look at the file called httpd.conf - if you're using Apache that is...

It should contain a line that looks like this:

AddType application/x-httpd-php .php

Anyway --- this line informs Apache that files ending with .php are PHP scripts. Thanks go to Jon-o Addleman jaddle@po-box.mcgill.ca for suggesting that I should write this piece.

Back to the example, which is somewhat silly. But it can be used to teach you a few things anyway.

Most of the code is just plain HTML. Only the part between <?php and ?> is PHP-code. You use either <?php or just <? to start the code and ?> to stop it again. The parser works within these two tags. Between the <?php and ?> you put a number of statements, each of which must be ended by a ; (a semi-colon).

Between the two tags there is a comment and one statement, echo "Hello World";. The comment will be completely ignored, and the statement just prints the string "Hello World" into the webpage, which is then sent to the browser. After the parser has processed the page, the final output that is sent to the browser looks this way:



<title>Hello World</title>




Hello World



As you can see, there's no trace of the PHP-code left in the final output. In the following code-snippets I won't include all the HTML code, but remember that you'll have to use HTML if you want your PHP scripts to work on the net. You'll also have to include the right HTML codes in the strings you output from PHP, if you're outputting more than just very simple strings.

You should think of a PHP-script as a normal web page written in HTML, since most of the page is still just plain-old HTML. But between the start- and end-tags (<?php or <? and ?>) the PHP-parsed spices the page up with possibly dynamic content. 

Before I show you any more examples I had better introduce to you the different kinds of variables we have in PHP. Variables are what you use to store something for later use. You might want to refer to your age, so it would be nice to define the age only once, and then just refer to it later. When your age changes, you just update the definition of your age, and then all other calculations that refer to the age are also updated. All the names of the variables must start with the $ character. After that there must be a letter. The rest of the name can consist of both letters and numbers. Here are three examples of valid names for a variable: $name, $address2 and $colour_30. As you can see, you can also use the _ (underscore) character in your names. It is often used as a replacement for spaces, since they are not allowed.

To use a variable you have to assign a value to it. It is done this way:

$name = 'Martin';
The name of the variable is on the left of the = sign and the new value of the variable is on the right. Here we assign the value of 18 to the variable $age: $age = 18;
There are several different types of variables. You have already met the integer and string types. Lets start with the integer type.


Dealing with numbers is easy with PHP. You can just use them as you like. All the normal rules about precedence apply, so you can write something like this:

$a = 4;
$b = 7;
$c = 2 + 3 * $a + 5 * $b // This is evaluated as 2 + (3 * 4) + (5 * 7) = 49
As you can see, everything works as you would expect. Any variables are automatically substituted with their values. This is done before the new value is stored in the variable on the left side of the =-sign. This means that you can do something like this: $a = 5;
$b = 10;
$a = (2 * $a + 5)/$b; // Evaluated as $a = (2 * 5 + 5)/10 = 1.5


To assign a string to a variable, you must use quotes. You can choose between single quotes (') or double quotes ("). The kind of quote you should choose depends on the string you work with in a given situation. There are some differences between the two types of quotes, which the following code should demonstrate:

$first_name = 'Martin';
$greeting1  = "Hello, my first name is $first_name.";
$last_name  = "Geisler";
$greeting2  = 'Hello, my last name is $last_name.';
This code produces the following output:
Hello, my first name is Martin.

Hello, my last name is $last_name.

"Why was it only the first string that worked properly?", you might ask. The answer is that when using double quotes, PHP performs what is known as variable-expansion. That means that PHP expands (substitutes) $first_name with it's value, just as you saw it when using numbers. The result is that the string stored in $greeting1 is "Hello, my name is Martin." (the quotes are not part of the string stored in $greeting1). When assigning to $greeting2 using single quotes, PHP doesn't substitute any variables and $greeting2 ends up with "Hello, my last name is $last_name.".

But there's more magic with the double quotes. They will also expand other special characters, such as newline-charecters (\n). This mean that the following string:

echo "Martin\nGeisler";
will look like this in the source of the page:


You've met the echo statement before (in the Hello World example) and it just outputs the string after it to the webbrowser. As I said earlier there's no trace of the PHP code, but notice that there's a linebreak between the two words. (The linebreak can only be seen in the source of the page, because the webbrowser treats the linebreak as a normal space.) You should also notice that there's no space before or after the \n-character. Because you use the \-character to indicate that the next character is special, you have to write \\ when you want to have a single \. The backslash (\) is called an escape-character, since it makes the characters escape from their normal role. The escape-character also has to be escaped.

If you want to have a $-charactar in a string enclosed in double-quotes, you'll need to escape it, or else PHP will look for a variable, since all variables start with a $.

This string will give you a little problem:

$name = "Martin "gimpster" Geisler";

If you try it out, you'll get an error. That's because the parser expects the string to stop after the second double quote, just before the word gimpster. The rest of the line is just garbage to the parser so it complains loud. Perhaps it would help if we escaped the double quote? Exactly! The following string is correct:

$name = "Martin \"gimpster\" Geisler";

The same applies if we had been using single quotes throughout the string. But there's another way to solve the problem. You could quote the string with single quotes and then use as many double quotes within the string as you would like. Or you could do the inverse, and enclose the string in double quotes and use single quotes in the string. This solves a problem you would have when outputting HTML-code with lots of double quotes:

echo '
<a href="mailto:gimpster@gimpster.com">
  <img src="image.gif" width="16" height="16" alt="me" border="0">

If we had quoted the string with double quotes, we would have had to escape all the other double quotes. But on the other hand, in the string above PHP wouldn't do any variable-substitution. Also, the escape-characters will not be understood, so you can't write \n in a string enclosed by single quotes. Instead you can put the newlines directly into the code, as I've done in the example. The final output will still contain those newlines.

It's only when you're evaluating expressions (such as in assignments and echo statements) that the difference between single and double quotes comes into play. After the assignment is done, you wont be able to tell how the string was produced. Consider this code:

$first_name = 'Martin';
$last_name = "Geisler";
$var1 = "$first_name $last_name";
$var2 = 'Martin Geisler';

The two variables $var1 and $var2 will both contain the string Martin Geisler after the evaluation, without any sign of all the different kinds of quotes that were used in the construction and PHP will not make any distinction between the values stored in the two variables.

If you want to concatenate (add together) two strings you use the .-character (a dot or full stop) like this:

$first_name = 'Martin';
$last_name = 'Geisler';
$full_name = $first_name.$last_name;
Now $full_name will contain the string "MartinGeisler". That's probably not what you wanted, it would be better with a space between the two words. To do this, you execute the following code: $full_name = $first_name . ' ' . $last_name;

Notice that the space between the variable-name and the dot is ignored --- it's the string with the space that's important. You use the dot (full-stop) each time you have two strings that you need to put after one another.

But you've also seen how this could be solved using variable-substitution. This is often easier to read, just remember to use double-quotes around the string. So this code gives the same result:

$full_name = "$first_name $last_name";

When using variable substitution you might run into a little problem. The problem occurs when you want to substitute a variable in a string and have text immediately after the variable. You have to delimit the name of the variable somehow. Using curly braces is the solution as shown in this code:

$noun = 'car';
"A $noun, two ${noun}s";

Here we're playing with English words in plural form. The first occurrence of $noun is substituted correctly, even when it's followed immediately by a comma because you cannot have a comma in the name of a variable. But you could certainly have a variable called $nouns, so we have to use curly braces to indicate that we want the variable called $noun and not some nonexistent variable called $nouns.


Arrays are a single variable that can hold a number of different values at the same time. To access the different values you specify them by using brackets after the name. It works like this:

$name[0] = 'Martin';

$name[1] = 'Geisler';

(There's no syntax hightlighting in the above example because it involves square brackets with numbers in them, which is also used for footnotes in PhpWiki. So some of the following examples will be in plain black & white.)

Now we have made an array, and filled in the two first slots. Notice that the numbering starts at zero and not one. We can recall them like this:

echo $name[0] . ' ' . $name[1];

If you don't care which slot you get when you assign to an array, then use a pair of empty brackets after the name:

$name[] = 'Martin';
$name[] = 'gimpster';
$name[] = 'Geisler';

This filled the first three slots of the array, numbered 0, 1, and 2. When you want to retrieve the information again, you must specify which slot you want access to. But you can also identify the different slots by name. This makes everything much more readable and easy to understand:

$personal_data['first name'] = 'Martin';
$personal_data['nickname'] = 'gimpster';
$personal_data['last name'] = 'Geisler';
$personal_data['age'] = 18;
'Hello ' . $personal_data['first name'] .
'! We know that you are ' . $personal_data['age'] .
' years old.';

Notice how I broke the last line to make it shorter and how I used the dot operator to glue the strings together with the variables. The parser will ignore all whitespace, and the result is:

Hello Martin! We know that you are 18 years old.

Also note how I split the string up into its individual pieces. This is often the easiest way to deal with associative arrays. Another possibility is to use variable substitution with curly braces like this:

echo "Hello {$personal_data['first name']}! " .
"We know that you are {$personal_data['age']} years old.";

This code gives the same result. You can put complex expressions inside such a set of curly braces. This includes accessing elements in multidimensional arrays and object properties.

Please do use descriptive names for the slots in your arrays. It makes the code much more clear, and helps to indicate what the different slots are intended for.

Now that we have seen how to assign values to variables, let's have a look at the different control structures there are in PHP:

You use control-structures to control the flow of execution through your program. If you didn't have any of these control-structures, all your scripts would always do the same. So let's get on with it.

The if-statement

We start with the if-statement. You use the if-statement to execute different parts of your script based on the truth-value of an expression. The expression could be something like $a > 5. This tests to see if the variable $a is greater than 5. If so, the expression is true, and the next part of the if-statement is executed. If the expression is false, then the optional else-part of the if-statement is executed. In a script it would look like this:

if ($a < 5) {
$b = $a + 10;
} else {
$b = $a * 2;

If the part between the parenthesis, (), is true, then the statements between the first pair of curly braces, {...} is executed. Else the part in the curly brackets after the else keyword is executed. You don't have to provide a else-part, if you don't then nothing will be executed if the test fails.

You can have all kinds of statement between the parenthesis and the curly braces. PHP is a language where almost every statement has a value, a value that can be used together with other values to build big expressions.

Take an assignment for example. The statement $b = $a actually has a value, namely the value of $a. This means that you can use this value in another assignment, like this:

$c = ($b = $a);

The assignment in the parenthesis has the value of $a which is then assigned to $c. You can actually drop the parenthesis; the result is the same: all three variables have the value of $a. This allows you to understand a statement like this:

/* Initialize counters: */
$i = $j = $k = $l = 0;

To compare the value of a variable up against other variables of numbers, you have a number of different comparison-operators:

 Operator   Meaning 
 ==   Is equal to 
 !=   Is not equal to 
 <   Is less than 
 <=   Is less than or equal to 
 >   Is greater than 
 >=   Is greater than or equal to 
 &&, and   The first and the second is true 
 , or   The first or the second or both is true 

(Note: problems with the WikiMarkup; you actually need to use || and not as the or operator.)

Using these operators you can construct every Boolean expression you'll ever need. Take for example the following if-statement:

if ($a <= $b && ($array[$a] <= $array[$b] || $array[$a+$b] > $array[$a-$b])) {
/* Do something useful - this is a comment, it won't be executed */

This if-statement first tests to see if the variable $a is smaller or equal to the variable $b. If true, the next part is executed to see if it is also true. This happens because we're using &&, which means that the first statement and the second statement should be true.

The second part uses $a and $b as indexes in an array. If the first part is true, then the second part won't be executed. We're just interested in whether the entire test is true, which it is, if either the first or the second test is true. So there's no need to evaluate the second test, if the first is already true. This means that you can execute the following statement:

mysql_connect() or die('Could not connect to the database!');

The function mysql_connect() tries to make a connection to a MySQL database. It returns true if it was able to make a connection. PHP evaluates the line as a Boolean expression. That means that it starts by executing mysql_connect(). If that returns true, then PHP skips the next statement, because we used or. The second statement makes PHP stop with an error message, if the first statement fails. As you can see, you can make it as complicated as you want.

There are other control-structures:

The for-loop

The for-loop is useful when you need to have the same task repeated a number of times. The for-loop works by first initializing a variable, then performing a test, and then - if the test is true - executing some code. After that code has been executed, additional code is run to increment the variable and then execute the main code again. An example will probably help:

for ($i = 0; $i < 10; $i++) {

"Hmm, now he introduced several new things again" you might think. And you're perfectly right. First, let's have a look at the for-loop.

It first assigns a starting-value of 0 to $i, then it checks to see if $i is smaller than 10, and, if so, the echo-statement is executed. At the end of an iteration, (one pass of the loop) the $i++ statement is executed.

This statement is a shorthand of $i = $i + 1;. It simply increments the variable i by 1. Similarly, you could write $n--; that would decrement the variable $n by 1.

The main code is executed while $i is less than 10, so our little for-loop simply prints the numbers from 0 to 9 into our webpage.

The while-loop

The while-loop is very similar to the for-loop. It works like this:

while (something is true) {
some statements

It is used when you don't know in advance how many iterations you'll need. If you look close, then you'll see that each for-loop can be made into a while-loop and that each while-loop can be made into a for-loop. This is because a for-loop like this (where a, b, c, and d are statements):

for (a; b; c) {

works completely like this while-loop:

while (
b) {

You'll often see a while-loop being used to run though an array like this:

while (list($key, $val) = each($array)) {
"<p>$key => $val</p>\n";

The condition in the while-loop uses several functions and an assignment. Each time the function each() is called, it returns an array with the key and value from $array. When there are no elements left in $array, each() returns false - this is what makes the whole condition false and makes while-loop stop.

The list() function can be used on the left side of an assignment to extract values from an array. You can use list() like this, although it would be much easier to assign the variables one at a time:

list($a, $b, $c, $d) = array(1, 2, 3, 4); // $a == 1, $b == 2, $c == 3, $d == 4

So the condition in the while-loop prepares $key and $val for the body of the loop. When all key-value pairs have been examined, the loop terminates because each() return false, which then becomes the value of the assignment, and therefore the value of the condition.

After you've run through the array like this, each() will return false if you use it on the same array. You'll have to call reset($array) to make each() start from the beginning again.

Here is an example with reset() and each() that will build a table that shows the keys and values of an array:

$array = array('Foo' => 42,
'Bar' => 3.14,
'Baz' => 2.71);

// Start the table
echo "<table>\n";
// Print the first row with headers
echo "<tr>\n";
"  <th>Key</th>\n";
"  <th>Value</th>\n";

// Ensure that the each() function starts from the start:
while (list(
$key, $value) = each($array)) {
// Print a single row
echo "<tr>\n";
// End the table
echo "</table>\n";

This code will produce a table like this one:

 Key   Value 
 Foo   42 
 Bar   3.14 
 Baz   2.71 

The important thing for you to notice in the above code is the way HTML tables can be built up in PHP. Tables in HTML start with the opening tag, and then comes any number of rows, each row consists of any number of cells. Finally you close the table again.

So when building HTML tables using PHP, remember that they always start in the upper left-hand corner (the cell that says "Key" in this example) and then continue to the right along the same row: then come the rows below in the same way, left-to-right, until you finally end up in the lower right-hand corner (that's the cell that says "2.71" here).

Whenever you need to use a certain piece of code over and over again, it is very useful to put it in a function. By putting your code into functions, you make the code easy to reuse. Whenever you locate an error in your code, you only have to fix it once: in the function. Using functions also makes your code easier to read and understand, especially when you choose sensible names for functions. Let's start with a simple function:

function foo($arg) {
  if (
$arg < 10) {
"$arg is smaller than 10";
  } else {
"$arg is greater than or equal to 10";

This piece of code declares a function called foo. It takes one argument (that's the bit between parenthesis called $arg above), and it echoes different things, based on the value of that argument. You "call" the foo function lower down the page, in your other code like this:

foo(15); /* Echoes "15 is greater than or equal to 10" */

Isn't that simple? You can also make functions that don't have any arguments, just add an empty pair of parenthesis after the name of the function, both when you declare the function and when you use it.

If your version of PHP is below 4 (use the function phpinfo() to learn all sorts of things about PHP) you have to declare your functions before you use them. That isn't necessary with version 4 and above. You really should use at least version 4, as it is the latest at the time of writing.

You often need your functions to return a value you can use in your programs, and here is how to do that: simply use the return statement.

function bar() {
  return (
time() - mktime(0, 0, 0, 1, 21, 1982))/86400;

This function has no arguments and it returns the number of days since the 21st of January, 1982 (my birthday :-) It uses a number of different built-in functions to get the time now, and then subtract the time I was born, and finally divide the time with 86400, the number of seconds in a day.

Under UNIX all times are measured as the number of seconds since the beginning of the Epoch, which was in 1970. So when you deal with times as I just did, you are just dealing with rather large integers (around 950,000,000). That makes it very easy to find the time between two events.

Note that before we return the number, we could round it with the round()-function, which will round the number to an integer: more on round() below!

In the body of the page you call into action that bar function like this :

$days_ages = bar();

If instead you wanted to round the floating-point number to a certain number of decimal places, then it's easy to make your own function:

function smart_round($float, $dec) {
  if (
$dec == 0) {
  } else {
floor($float * pow(10, $dec) + 0.5) / pow(10, $dec);

$a = smart_round(2.317, 2); /* Now $a is equal to 2.32 */

This smart_round function returns $float rounded to $dec decimals. The php-function:round has been changed in recent versions of PHP to allow the use of that second, precision, $dec argument.

A new feature in PHP4 makes it possible to use an entire file as a function. This makes it possible to save all your really cool functions in different files, and then use them in different scripts. This expands on the concept of code-reuse, as I talked about earlier. All this is possible thanks to the improved include language-construct: include() can now use a value returned from an external file, instead of calling a function declared in the same file.

Let's look at this in action! We have the file foo.inc which we want to use within the file script.php. This is the code for foo.inc:

/*  You must write the included files within PHP,
   in other words within <?php and ?> tags. This is because the assumption is always made that
   included files are not part of any PHP code.  */

return (time() - mktime(0, 0, 0, 1, 21, 1982))/86400;

Then you could write the following in script.php:

$days_aged = include('foo.inc');

That works just like the other example, except that you have reused the code from an external source.

Now that you've learned the basics of PHP, you should be able to write your own PHP-scripts


PHP Tutorial
Written by Martin Geisler