Files
doc-en/language/control-structures.xml
Hartmut Holzgraefe 9410c19fda added version attribute to <?xml tags
git-svn-id: https://svn.php.net/repository/phpdoc/en/trunk@61864 c90b9560-bf6c-de11-be94-00142212c4b1
2001-11-10 21:49:43 +00:00

1450 lines
47 KiB
XML

<?xml version="1.0" encoding="iso-8859-1"?>
<!-- $Revision: 1.42 $ -->
<chapter id="control-structures">
<title>Control Structures</title>
<simpara>
Any PHP script is built out of a series of statements. A statement
can be an assignment, a function call, a loop, a conditional
statement of even a statement that does nothing (an empty
statement). Statements usually end with a semicolon. In addition,
statements can be grouped into a statement-group by encapsulating a
group of statements with curly braces. A statement-group is a
statement by itself as well. The various statement types are
described in this chapter.
</simpara>
<sect1 id="control-structures.if">
<title><literal>if</literal></title>
<para>
The <literal>if</literal> construct is one of the most important
features of many languages, PHP included. It allows for
conditional execution of code fragments. PHP features an
<literal>if</literal> structure that is similar to that of C:
<informalexample>
<programlisting>
if (expr)
statement
</programlisting>
</informalexample>
</para>
<simpara>
As described in the section about expressions, expr is evaluated
to its truth value. If <replaceable>expr</replaceable> evaluates
to &true;, PHP will execute statement, and if it
evaluates to &false; - it'll ignore it.
</simpara>
<para>
The following example would display <computeroutput>a is bigger
than b</computeroutput> if <replaceable>$a</replaceable> is bigger
than <replaceable>$b</replaceable>:
<informalexample>
<programlisting role="php">
if ($a &gt; $b)
print "a is bigger than b";
</programlisting>
</informalexample>
</para>
<para>
Often you'd want to have more than one statement to be executed
conditionally. Of course, there's no need to wrap each statement
with an <literal>if</literal> clause. Instead, you can group
several statements into a statement group. For example, this code
would display <computeroutput>a is bigger than b</computeroutput>
if <replaceable>$a</replaceable> is bigger than
<replaceable>$b</replaceable>, and would then assign the value of
<replaceable>$a</replaceable> into <replaceable>$b</replaceable>:
<informalexample>
<programlisting role="php">
if ($a &gt; $b) {
print "a is bigger than b";
$b = $a;
}
</programlisting>
</informalexample>
</para>
<simpara>
If statements can be nested indefinitely within other
<literal>if</literal> statements, which provides you with complete
flexibility for conditional execution of the various parts of your
program.
</simpara>
</sect1>
<sect1 id="control-structures.else">
<title><literal>else</literal></title>
<para>
Often you'd want to execute a statement if a certain condition is
met, and a different statement if the condition is not met. This
is what <literal>else</literal> is for. <literal>else</literal>
extends an <literal>if</literal> statement to execute a statement
in case the expression in the <literal>if</literal> statement
evaluates to &false;. For example, the following
code would display <computeroutput>a is bigger than
b</computeroutput> if <replaceable>$a</replaceable> is bigger than
<replaceable>$b</replaceable>, and <computeroutput>a is NOT bigger
than b</computeroutput> otherwise:
<informalexample>
<programlisting role="php">
if ($a &gt; $b) {
print "a is bigger than b";
} else {
print "a is NOT bigger than b";
}
</programlisting>
</informalexample>
The <literal>else</literal> statement is only executed if the
<literal>if</literal> expression evaluated to
&false;, and if there were any
<literal>elseif</literal> expressions - only if they evaluated to
&false; as well (see <link
linkend="control-structures.elseif">elseif</link>).
</para>
</sect1>
<sect1 id="control-structures.elseif">
<title><literal>elseif</literal></title>
<para>
<literal>elseif</literal>, as its name suggests, is a combination
of <literal>if</literal> and <literal>else</literal>. Like
<literal>else</literal>, it extends an <literal>if</literal>
statement to execute a different statement in case the original
<literal>if</literal> expression evaluates to
&false;. However, unlike
<literal>else</literal>, it will execute that alternative
expression only if the <literal>elseif</literal> conditional
expression evaluates to &true;. For example, the
following code would display <computeroutput>a is bigger than
b</computeroutput>, <computeroutput>a equal to b</computeroutput>
or <computeroutput>a is smaller than b</computeroutput>:
<informalexample>
<programlisting role="php">
if ($a &gt; $b) {
print "a is bigger than b";
} elseif ($a == $b) {
print "a is equal to b";
} else {
print "a is smaller than b";
}
</programlisting>
</informalexample>
</para>
<simpara>
There may be several <literal>elseif</literal>s within the same
<literal>if</literal> statement. The first
<literal>elseif</literal> expression (if any) that evaluates to
&true; would be executed. In PHP, you can also
write 'else if' (in two words) and the behavior would be identical
to the one of 'elseif' (in a single word). The syntactic meaning
is slightly different (if you're familiar with C, this is the same
behavior) but the bottom line is that both would result in exactly
the same behavior.
</simpara>
<simpara>
The <literal>elseif</literal> statement is only executed if the
preceding <literal>if</literal> expression and any preceding
<literal>elseif</literal> expressions evaluated to
&false;, and the current
<literal>elseif</literal> expression evaluated to
&true;.
</simpara>
</sect1>
<sect1 id="control-structures.alternative-syntax">
<title>Alternative syntax for control structures</title>
<para>
PHP offers an alternative syntax for some of its control
structures; namely, <literal>if</literal>,
<literal>while</literal>, <literal>for</literal>,
<literal>foreach</literal>, and <literal>switch</literal>.
In each case, the basic form of the alternate syntax is to change
the opening brace to a colon (:) and the closing brace to
<literal>endif;</literal>, <literal>endwhile;</literal>,
<literal>endfor;</literal>, <literal>endforeach;</literal>, or
<literal>endswitch;</literal>, respectively.
<informalexample>
<programlisting role="php">
&lt;?php if ($a == 5): ?&gt;
A is equal to 5
&lt;?php endif; ?&gt;
</programlisting>
</informalexample>
</para>
<simpara>
In the above example, the HTML block "A = 5" is nested within an
<literal>if</literal> statement written in the alternative syntax.
The HTML block would be displayed only if $a is equal to 5.
</simpara>
<para>
The alternative syntax applies to <literal>else</literal> and
<literal>elseif</literal> as well. The following is an
<literal>if</literal> structure with <literal>elseif</literal> and
<literal>else</literal> in the alternative format:
<informalexample>
<programlisting role="php">
if ($a == 5):
print "a equals 5";
print "...";
elseif ($a == 6):
print "a equals 6";
print "!!!";
else:
print "a is neither 5 nor 6";
endif;
</programlisting>
</informalexample>
</para>
<para>
See also <link linkend="control-structures.while">while</link>,
<link linkend="control-structures.for">for</link>, and <link
linkend="control-structures.if">if</link> for further examples.
</para>
</sect1>
<sect1 id="control-structures.while">
<title><literal>while</literal></title>
<para>
<literal>while</literal> loops are the simplest type of loop in
PHP. They behave just like their C counterparts. The basic form
of a <literal>while</literal> statement is:
<informalexample>
<programlisting>
while (expr) statement
</programlisting>
</informalexample>
</para>
<simpara>
The meaning of a <literal>while</literal> statement is simple. It
tells PHP to execute the nested statement(s) repeatedly, as long
as the <literal>while</literal> expression evaluates to
&true;. The value of the expression is checked
each time at the beginning of the loop, so even if this value
changes during the execution of the nested statement(s), execution
will not stop until the end of the iteration (each time PHP runs
the statements in the loop is one iteration). Sometimes, if the
<literal>while</literal> expression evaluates to
&false; from the very beginning, the nested
statement(s) won't even be run once.
</simpara>
<para>
Like with the <literal>if</literal> statement, you can group
multiple statements within the same <literal>while</literal> loop
by surrounding a group of statements with curly braces, or by
using the alternate syntax:
<informalexample>
<programlisting>
while (expr): statement ... endwhile;
</programlisting>
</informalexample>
</para>
<para>
The following examples are identical, and both print numbers from
1 to 10:
<informalexample>
<programlisting role="php">
/* example 1 */
$i = 1;
while ($i &lt;= 10) {
print $i++; /* the printed value would be
$i before the increment
(post-increment) */
}
/* example 2 */
$i = 1;
while ($i &lt;= 10):
print $i;
$i++;
endwhile;
</programlisting>
</informalexample>
</para>
</sect1>
<sect1 id="control-structures.do.while">
<title><literal>do..while</literal></title>
<simpara>
<literal>do..while</literal> loops are very similar to
<literal>while</literal> loops, except the truth expression is
checked at the end of each iteration instead of in the beginning.
The main difference from regular <literal>while</literal> loops is
that the first iteration of a <literal>do..while</literal> loop is
guarenteed to run (the truth expression is only checked at the end
of the iteration), whereas it's may not necessarily run with a
regular <literal>while</literal> loop (the truth expression is
checked at the beginning of each iteration, if it evaluates to
&false; right from the beginning, the loop
execution would end immediately).
</simpara>
<para>
There is just one syntax for <literal>do..while</literal> loops:
<informalexample>
<programlisting role="php">
$i = 0;
do {
print $i;
} while ($i>0);
</programlisting>
</informalexample>
</para>
<simpara>
The above loop would run one time exactly, since after the first
iteration, when truth expression is checked, it evaluates to
&false; ($i is not bigger than 0) and the loop
execution ends.
</simpara>
<para>
Advanced C users may be familiar with a different usage of the
<literal>do..while</literal> loop, to allow stopping execution in
the middle of code blocks, by encapsulating them with
<literal>do..while</literal>(0), and using the <link
linkend="control-structures.break"><literal>break</literal></link>
statement. The following code fragment demonstrates this:
<informalexample>
<programlisting role="php">
do {
if ($i &lt; 5) {
print "i is not big enough";
break;
}
$i *= $factor;
if ($i &lt; $minimum_limit) {
break;
}
print "i is ok";
...process i...
} while(0);
</programlisting>
</informalexample>
</para>
<simpara>
Don't worry if you don't understand this right away or at all.
You can code scripts and even powerful scripts without using this
`feature'.
</simpara>
</sect1>
<sect1 id="control-structures.for">
<title><literal>for</literal></title>
<para>
<literal>for</literal> loops are the most complex loops in PHP.
They behave like their C counterparts. The syntax of a
<literal>for</literal> loop is:
<informalexample>
<programlisting>
for (expr1; expr2; expr3) statement
</programlisting>
</informalexample>
</para>
<simpara>
The first expression (<replaceable>expr1</replaceable>) is
evaluated (executed) once unconditionally at the beginning of the
loop.
</simpara>
<simpara>
In the beginning of each iteration,
<replaceable>expr2</replaceable> is evaluated. If it evaluates to
&true;, the loop continues and the nested
statement(s) are executed. If it evaluates to
&false;, the execution of the loop ends.
</simpara>
<simpara>
At the end of each iteration, <replaceable>expr3</replaceable> is
evaluated (executed).
</simpara>
<simpara>
Each of the expressions can be empty.
<replaceable>expr2</replaceable> being empty means the loop should
be run indefinitely (PHP implicitly considers it as
&true;, like C). This may not be as useless as
you might think, since often you'd want to end the loop using a
conditional <link
linkend="control-structures.break"><literal>break</literal></link>
statement instead of using the <literal>for</literal> truth
expression.
</simpara>
<para>
Consider the following examples. All of them display numbers from
1 to 10:
<informalexample>
<programlisting role="php">
/* example 1 */
for ($i = 1; $i &lt;= 10; $i++) {
print $i;
}
/* example 2 */
for ($i = 1;;$i++) {
if ($i &gt; 10) {
break;
}
print $i;
}
/* example 3 */
$i = 1;
for (;;) {
if ($i &gt; 10) {
break;
}
print $i;
$i++;
}
/* example 4 */
for ($i = 1; $i &lt;= 10; print $i, $i++);
</programlisting>
</informalexample>
</para>
<simpara>
Of course, the first example appears to be the nicest one (or
perhaps the fourth), but you may find that being able to use empty
expressions in <literal>for</literal> loops comes in handy in many
occasions.
</simpara>
<para>
PHP also supports the alternate "colon syntax" for
<literal>for</literal> loops.
<informalexample>
<programlisting>
for (expr1; expr2; expr3): statement; ...; endfor;
</programlisting>
</informalexample>
</para>
<para>
Other languages have a <literal>foreach</literal> statement to
traverse an array or hash. PHP 3 has no such construct; PHP 4 does
(see <link
linkend="control-structures.foreach">foreach</link>). In PHP 3, you
can combine <link linkend="control-structures.while">while</link>
with the <function>list</function> and <function>each</function>
functions to achieve the same effect. See the documentation for
these functions for an example.
</para>
</sect1>
<sect1 id="control-structures.foreach">
<title><literal>foreach</literal></title>
<para>
PHP 4 (not PHP 3) includes a <literal>foreach</literal> construct,
much like Perl and some other languages. This simply gives an easy
way to iterate over arrays. There are two syntaxes; the second is
a minor but useful extension of the first:
<informalexample>
<programlisting>
foreach(array_expression as $value) statement
foreach(array_expression as $key =&gt; $value) statement
</programlisting>
</informalexample>
</para>
<simpara>
The first form loops over the array given by
<literal>array_expression</literal>. On each loop, the value of
the current element is assigned to <literal>$value</literal> and
the internal array pointer is advanced by one (so on the next
loop, you'll be looking at the next element).
</simpara>
<simpara>
The second form does the same thing, except that the current
element's key will be assigned to the variable
<literal>$key</literal> on each loop.
</simpara>
<para>
<note>
<para>
When <literal>foreach</literal> first starts executing, the
internal array pointer is automatically reset to the first element
of the array. This means that you do not need to call
<function>reset</function> before a <literal>foreach</literal>
loop.
</para>
</note>
</para>
<para>
<note>
<para>
Also note that <literal>foreach</literal> operates on a copy of
the specified array, not the array itself, therefore the array
pointer is not modified as with the <function>each</function>
construct and changes to the array element returned are not
reflected in the original array.
</para>
</note>
</para>
<note>
<para>
<literal>foreach</literal> does not support the ability to
suppress error messages using '@'.
</para>
</note>
<para>
You may have noticed that the following are functionally
identical:
<informalexample>
<programlisting role="php">
reset ($arr);
while (list(, $value) = each ($arr)) {
echo "Value: $value&lt;br&gt;\n";
}
foreach ($arr as $value) {
echo "Value: $value&lt;br&gt;\n";
}
</programlisting>
</informalexample>
The following are also functionally identical:
<informalexample>
<programlisting role="php">
reset ($arr);
while (list($key, $value) = each ($arr)) {
echo "Key: $key; Value: $value&lt;br&gt;\n";
}
foreach ($arr as $key =&gt; $value) {
echo "Key: $key; Value: $value&lt;br&gt;\n";
}
</programlisting>
</informalexample>
</para>
<para>
Some more examples to demonstrate usages:
<informalexample>
<programlisting role="php">
/* foreach example 1: value only */
$a = array (1, 2, 3, 17);
foreach ($a as $v) {
print "Current value of \$a: $v.\n";
}
/* foreach example 2: value (with key printed for illustration) */
$a = array (1, 2, 3, 17);
$i = 0; /* for illustrative purposes only */
foreach($a as $v) {
print "\$a[$i] =&gt; $v.\n";
}
/* foreach example 3: key and value */
$a = array (
"one" =&gt; 1,
"two" =&gt; 2,
"three" =&gt; 3,
"seventeen" =&gt; 17
);
foreach($a as $k =&gt; $v) {
print "\$a[$k] =&gt; $v.\n";
}
/* foreach example 4: multi-dimensional arrays */
$a[0][0] = "a";
$a[0][1] = "b";
$a[1][0] = "y";
$a[1][1] = "z";
foreach($a as $v1) {
foreach ($v1 as $v2) {
print "$v2\n";
}
}
/* foreach example 5: dynamic arrays */
foreach(array(1, 2, 3, 4, 5) as $v) {
print "$v\n";
}
</programlisting>
</informalexample>
</para>
</sect1>
<sect1 id="control-structures.break">
<title><literal>break</literal></title>
<simpara>
<literal>break</literal> ends execution of the current
<literal>for</literal>, <literal>foreach</literal>
<literal>while</literal>, <literal>do..while</literal> or
<literal>switch</literal> structure.
</simpara>
<simpara>
<literal>break</literal> accepts an optional numeric argument
which tells it how many nested enclosing structures are to be
broken out of.
</simpara>
<para>
<informalexample>
<programlisting role="php">
$arr = array ('one', 'two', 'three', 'four', 'stop', 'five');
while (list (, $val) = each ($arr)) {
if ($val == 'stop') {
break; /* You could also write 'break 1;' here. */
}
echo "$val&lt;br&gt;\n";
}
/* Using the optional argument. */
$i = 0;
while (++$i) {
switch ($i) {
case 5:
echo "At 5&lt;br&gt;\n";
break 1; /* Exit only the switch. */
case 10:
echo "At 10; quitting&lt;br&gt;\n";
break 2; /* Exit the switch and the while. */
default:
break;
}
}
</programlisting>
</informalexample>
</para>
</sect1>
<sect1 id="control-structures.continue">
<title><literal>continue</literal></title>
<simpara>
<literal>continue</literal> is used within looping structures to
skip the rest of the current loop iteration and continue execution
at the beginning of the next iteration.
</simpara>
<simpara>
<literal>continue</literal> accepts an optional numeric argument
which tells it how many levels of enclosing loops it should skip
to the end of.
</simpara>
<para>
<informalexample>
<programlisting role="php">
while (list ($key, $value) = each ($arr)) {
if (!($key % 2)) { // skip odd members
continue;
}
do_something_odd ($value);
}
$i = 0;
while ($i++ &lt; 5) {
echo "Outer&lt;br&gt;\n";
while (1) {
echo "&nbsp;&nbsp;Middle&lt;br&gt;\n";
while (1) {
echo "&nbsp;&nbsp;Inner&lt;br&gt;\n";
continue 3;
}
echo "This never gets output.&lt;br&gt;\n";
}
echo "Neither does this.&lt;br&gt;\n";
}
</programlisting>
</informalexample>
</para>
</sect1>
<sect1 id="control-structures.switch">
<title><literal>switch</literal></title>
<simpara>
The <literal>switch</literal> statement is similar to a series of
IF statements on the same expression. In many occasions, you may
want to compare the same variable (or expression) with many
different values, and execute a different piece of code depending
on which value it equals to. This is exactly what the
<literal>switch</literal> statement is for.
</simpara>
<para>
The following two examples are two different ways to write the
same thing, one using a series of <literal>if</literal>
statements, and the other using the <literal>switch</literal>
statement:
<informalexample>
<programlisting role="php">
if ($i == 0) {
print "i equals 0";
}
if ($i == 1) {
print "i equals 1";
}
if ($i == 2) {
print "i equals 2";
}
switch ($i) {
case 0:
print "i equals 0";
break;
case 1:
print "i equals 1";
break;
case 2:
print "i equals 2";
break;
}
</programlisting>
</informalexample>
</para>
<para>
It is important to understand how the <literal>switch</literal>
statement is executed in order to avoid mistakes. The
<literal>switch</literal> statement executes line by line
(actually, statement by statement). In the beginning, no code is
executed. Only when a <literal>case</literal> statement is found
with a value that matches the value of the
<literal>switch</literal> expression does PHP begin to execute the
statements. PHP continues to execute the statements until the end
of the <literal>switch</literal> block, or the first time it sees
a <literal>break</literal> statement. If you don't write a
<literal>break</literal> statement at the end of a case's
statement list, PHP will go on executing the statements of the
following case. For example:
<informalexample>
<programlisting role="php">
switch ($i) {
case 0:
print "i equals 0";
case 1:
print "i equals 1";
case 2:
print "i equals 2";
}
</programlisting>
</informalexample>
</para>
<simpara>
Here, if $i equals to 0, PHP would execute all of the print
statements! If $i equals to 1, PHP would execute the last two
print statements, and only if $i equals to 2, you'd get the
'expected' behavior and only 'i equals 2' would be displayed. So,
it's important not to forget <literal>break</literal> statements
(even though you may want to avoid supplying them on purpose under
certain circumstances).
</simpara>
<simpara>
In a <literal>switch</literal> statement, the condition is
evaluated only once and the result is compared to each
<literal>case</literal> statement. In an <literal>elseif</literal>
statement, the condition is evaluated again. If your condition is
more complicated than a simple compare and/or is in a tight loop,
a <literal>switch</literal> may be faster.
</simpara>
<para>
The statement list for a case can also be empty, which simply
passes control into the statement list for the next case.
<informalexample>
<programlisting role="php">
switch ($i) {
case 0:
case 1:
case 2:
print "i is less than 3 but not negative";
break;
case 3:
print "i is 3";
}
</programlisting>
</informalexample>
</para>
<para>
A special case is the default case. This case matches anything
that wasn't matched by the other cases, and should be the last
<literal>case</literal> statement. For example:
<informalexample>
<programlisting role="php">
switch ($i) {
case 0:
print "i equals 0";
break;
case 1:
print "i equals 1";
break;
case 2:
print "i equals 2";
break;
default:
print "i is not equal to 0, 1 or 2";
}
</programlisting>
</informalexample>
</para>
<para>
The <literal>case</literal> expression may be any expression that
evaluates to a simple type, that is, integer or floating-point
numbers and strings. Arrays or objects cannot be used here unless
they are dereferenced to a simple type.
</para>
<para>
The alternative syntax for control structures is supported with
switches. For more information, see <link
linkend="control-structures.alternative-syntax">Alternative syntax
for control structures</link> .
<informalexample>
<programlisting role="php">
switch ($i):
case 0:
print "i equals 0";
break;
case 1:
print "i equals 1";
break;
case 2:
print "i equals 2";
break;
default:
print "i is not equal to 0, 1 or 2";
endswitch;
</programlisting>
</informalexample>
</para>
</sect1>
<sect1 id="control-structures.declare">
<title><literal>declare</literal></title>
<para>
The <literal>declare</literal> construct is used to
set execution directives for a block of code.
The syntax of <literal>declare</literal> is similiar to
the syntax of other flow control constructs:
<informalexample>
<programlisting>
declare (directive) statement
</programlisting>
</informalexample>
</para>
<para>
The <literal>directive</literal> section allows the
behavior of the <literal>declare</literal> block to
be set.
Currently only one directive is recognized: the
<literal>ticks</literal> directive. (See below for more
information on the
<link linkend="control-structures.declare.ticks">ticks</link>
directive)
</para>
<para>
The <literal>statement</literal> part of the
<literal>declare</literal> block will be executed - how
it is executed and what side-effects occur during execution
may depend on the directive set in the
<literal>directive</literal> block.
</para>
<sect2 id="control-structures.declare.ticks">
<title>Ticks</title>
<para>A tick is an event that occurs for every
<replaceable>N</replaceable> low-level statements executed
by the parser within the <literal>declare</literal> block.
The value for <replaceable>N</replaceable> is specified
using <literal>ticks=<replaceable>N</replaceable></literal>
within the <literal>declare</literal> blocks's
<literal>directive</literal> section.
</para>
<para>
The event(s) that occurs on each tick is specified using the
<function>register_tick_function</function>. See the example
below for more details. Note that more than one event can occur
for each tick.
</para>
<para>
<example>
<title>Profile a section of PHP code</title>
<programlisting role="php">
&lt;pre&gt;
&lt;?php
// A function that records the time when it is called
function profile ($dump = FALSE)
{
static $profile;
// Return the times stored in profile, then erase it
if ($dump) {
$temp = $profile;
unset ($profile);
return ($temp);
}
$profile[] = microtime ();
}
// Set up a tick handler
register_tick_function("profile");
// Initialize the function before the declare block
profile ();
// Run a block of code, throw a tick every 2nd statement
declare (ticks=2) {
for ($x = 1; $x &lt; 50; ++$x) {
echo similar_text (md5($x), md5($x*$x)), "&lt;br&gt;";
}
}
// Display the data stored in the profiler
print_r (profile (TRUE));
?&gt;
&lt;/pre&gt;
</programlisting>
</example>
The example profiles the PHP code within the 'declare'
block, recording the time at which every second low-level
statement in the block was executed. This information can
then be used to find the slow areas within particular
segments of code. This process can be performed using other
methods: using ticks is more convenient and easier to
implement.
</para>
<simpara>
Ticks are well suited for debugging, implementing simple
multitasking, backgrounded I/O and many other tasks.
</simpara>
<simpara>
See also <function>register_tick_function</function> and
<function>unregister_tick_function</function>.
</simpara>
</sect2>
</sect1>
<sect1 id="function.require">
<title><function>require</function></title>
<simpara>
The <function>require</function> statement replaces itself with
the specified file, much like the C preprocessor's
<literal>#include</literal> works.
</simpara>
<simpara>
If "URL fopen wrappers" are enabled in PHP (which they are in the
default configuration), you can specify the file to be
<function>require</function>ed using an URL instead of a local
pathname. See <link linkend="features.remote-files">Remote
files</link> and <function>fopen</function> for more information.
</simpara>
<simpara>
An important note about how this works is that when a file is
<function>include</function>ed or <function>require</function>ed,
parsing drops out of PHP mode and into HTML mode at the beginning
of the target file, and resumes PHP mode again at the end. For
this reason, any code inside the target file which should be
executed as PHP code must be enclosed within <link
linkend="language.basic-syntax.phpmode">valid PHP start and end
tags</link>.
</simpara>
<simpara>
<function>require</function> is not actually a function in PHP;
rather, it is a language construct. It is subject to some
different rules than functions are. For instance,
<function>require</function> is not subject to any containing
control structures. For another, it does not return any value;
attempting to read a return value from a
<function>require</function> call results in a parse error.
</simpara>
<simpara>
Unlike <function>include</function>, <function>require</function>
will <emphasis>always</emphasis> read in the target file,
<emphasis>even if the line it's on never executes</emphasis>. If
you want to conditionally include a file, use
<function>include</function>. The conditional statement won't
affect the <function>require</function>. However, if the line on
which the <function>require</function> occurs is not executed,
neither will any of the code in the target file be executed.
</simpara>
<simpara>
Similarly, looping structures do not affect the behaviour of
<function>require</function>. Although the code contained in the
target file is still subject to the loop, the
<function>require</function> itself happens only once.
</simpara>
<para>
This means that you can't put a <function>require</function>
statement inside of a loop structure and expect it to include the
contents of a different file on each iteration. To do that, use an
<function>include</function> statement.
<informalexample>
<programlisting role="php">
require ('header.inc');
</programlisting>
</informalexample>
</para>
<simpara>
When a file is <function>require</function>ed, the code it
contains inherits the variable scope of the line on which the
<function>require</function> occurs. Any variables available at
that line in the calling file will be available within the called
file. If the <function>require</function> occurs inside a
function within the calling file, then all of the code contained
in the called file will behave as though it had been defined
inside that function.
</simpara>
<para>
If the <function>require</function>ed file is called via HTTP
using the fopen wrappers, and if the target server interprets the
target file as PHP code, variables may be passed to the
<function>require</function>ed file using an URL request string as
used with HTTP GET. This is not strictly speaking the same thing
as <function>require</function>ing the file and having it inherit
the parent file's variable scope; the script is actually being run
on the remote server and the result is then being included into
the local script.
<informalexample>
<programlisting role="php">
/* This example assumes that someserver is configured to parse .php
* files and not .txt files. Also, 'works' here means that the variables
* $varone and $vartwo are available within the require()ed file. */
/* Won't work; file.txt wasn't handled by someserver. */
require ("http://someserver/file.txt?varone=1&amp;vartwo=2");
/* Won't work; looks for a file named 'file.php?varone=1&amp;vartwo=2'
* on the local filesystem. */
require ("file.php?varone=1&amp;vartwo=2");
/* Works. */
require ("http://someserver/file.php?varone=1&amp;vartwo=2");
$varone = 1;
$vartwo = 2;
require ("file.txt"); /* Works. */
require ("file.php"); /* Works. */
</programlisting>
</informalexample>
</para>
<simpara>
In PHP 3, it is possible to execute a <literal>return</literal>
statement inside a <function>require</function>ed file, as long as
that statement occurs in the global scope of the
<function>require</function>ed file. It may not occur within any
block (meaning inside braces ({}). In PHP 4, however, this ability
has been discontinued. If you need this functionality, see
<function>include</function>.
</simpara>
<simpara>
See also <function>include</function>, <function>require_once</function>,
<function>include_once</function>, <function>readfile</function>,
and <function>virtual</function>.
</simpara>
</sect1>
<sect1 id="function.include">
<title><function>include</function></title>
<simpara>
The <function>include</function> statement includes and evaluates
the specified file.
</simpara>
<simpara>
If "URL fopen wrappers" are enabled in PHP (which they are in the
default configuration), you can specify the file to be
<function>include</function>ed using an URL instead of a local
pathname. See <link linkend="features.remote-files">Remote
files</link> and <function>fopen</function> for more information.
</simpara>
<simpara>
An important note about how this works is that when a file is
<function>include</function>ed or <function>require</function>ed,
parsing drops out of PHP mode and into HTML mode at the beginning
of the target file, and resumes again at the end. For this reason,
any code inside the target file which should be executed as PHP
code must be enclosed within <link
linkend="language.basic-syntax.phpmode">valid PHP start and end
tags</link>.
</simpara>
<para>
This happens each time the <function>include</function> statement
is encountered, so you can use an <function>include</function>
statement within a looping structure to include a number of
different files.
<informalexample>
<programlisting role="php">
$files = array ('first.inc', 'second.inc', 'third.inc');
for ($i = 0; $i &lt; count($files); $i++) {
include $files[$i];
}
</programlisting>
</informalexample>
</para>
<para>
<function>include</function> differs from
<function>require</function> in that the include statement is
re-evaluated each time it is encountered (and only when it is
being executed), whereas the <function>require</function>
statement is replaced by the required file when it is first
encountered, whether the contents of the file will be evaluated or
not (for example, if it is inside an <link
linkend="control-structures.if">if</link> statement whose
condition evaluated to &false;).
</para>
<para>
Because <function>include</function> is a special language
construct, you must enclose it within a statement block if it is
inside a conditional block.
<informalexample>
<programlisting role="php">
/* This is WRONG and will not work as desired. */
if ($condition)
include($file);
else
include($other);
/* This is CORRECT. */
if ($condition) {
include($file);
} else {
include($other);
}
</programlisting>
</informalexample>
</para>
<simpara>
In both PHP 3 and PHP 4, it is possible to execute a
<literal>return</literal> statement inside an
<function>include</function>ed file, in order to terminate
processing in that file and return to the script which called
it. Some differences in the way this works exist, however. The
first is that in PHP 3, the <literal>return</literal> may not
appear inside a block unless it's a function block, in which case
the <literal>return</literal> applies to that function and not the
whole file. In PHP 4, however, this restriction does not
exist. Also, PHP 4 allows you to return values from
<function>include</function>ed files. You can take the value of
the <function>include</function> call as you would a normal
function. This generates a parse error in PHP 3.
</simpara>
<example>
<title><function>include</function> in PHP 3 and PHP 4</title>
<para>
Assume the existence of the following file (named
<filename>test.inc</filename>) in the same directory as the main
file:
<programlisting role="php">
&lt;?php
echo "Before the return &lt;br&gt;\n";
if (1) {
return 27;
}
echo "After the return &lt;br&gt;\n";
?&gt;
</programlisting>
</para>
<para>
Assume that the main file (<filename>main.html</filename>)
contains the following:
<programlisting role="php">
&lt;?php
$retval = include ('test.inc');
echo "File returned: '$retval'&lt;br&gt;\n";
?&gt;
</programlisting>
</para>
<para>
When <filename>main.html</filename> is called in PHP 3, it will
generate a parse error on line 2; you can't take the value of an
<function>include</function> in PHP 3. In PHP 4, however, the
result will be:
<screen>
Before the return
File returned: '27'
</screen>
</para>
<para>
Now, assume that <filename>main.html</filename> has been altered
to contain the following:
<programlisting role="php">
&lt;?php
include ('test.inc');
echo "Back in main.html&lt;br&gt;\n";
?&gt;
</programlisting>
</para>
<para>
In PHP 4, the output will be:
<screen>
Before the return
Back in main.html
</screen>
However, PHP 3 will give the following output:
<screen>
Before the return
27Back in main.html
Parse error: parse error in /home/torben/public_html/phptest/main.html on line 5
</screen>
</para>
<para>
The above parse error is a result of the fact that the
<literal>return</literal> statement is enclosed in a non-function
block within <filename>test.inc</filename>. When the return is
moved outside of the block, the output is:
<screen>
Before the return
27Back in main.html
</screen>
</para>
<para>
The spurious '27' is due to the fact that PHP 3 does not support
<literal>return</literal>ing values from files like that.
</para>
</example>
<simpara>
When a file is <function>include</function>ed, the code it
contains inherits the variable scope of the line on which the
<function>include</function> occurs. Any variables available at
that line in the calling file will be available within the called
file. If the <function>include</function> occurs inside a
function within the calling file, then all of the code contained
in the called file will behave as though it had been defined
inside that function.
</simpara>
<para>
If the <function>include</function>ed file is called via HTTP
using the fopen wrappers, and if the target server interprets the
target file as PHP code, variables may be passed to the
<function>include</function>ed file using an URL request string as
used with HTTP GET. This is not strictly speaking the same thing
as <function>include</function>ing the file and having it inherit
the parent file's variable scope; the script is actually being run
on the remote server and the result is then being included into
the local script.
<informalexample>
<programlisting role="php">
/* This example assumes that someserver is configured to parse .php
* files and not .txt files. Also, 'works' here means that the variables
* $varone and $vartwo are available within the include()ed file. */
/* Won't work; file.txt wasn't handled by someserver. */
include ("http://someserver/file.txt?varone=1&amp;vartwo=2");
/* Won't work; looks for a file named 'file.php?varone=1&amp;vartwo=2'
* on the local filesystem. */
include ("file.php?varone=1&amp;vartwo=2");
/* Works. */
include ("http://someserver/file.php?varone=1&amp;vartwo=2");
$varone = 1;
$vartwo = 2;
include ("file.txt"); /* Works. */
include ("file.php"); /* Works. */
</programlisting>
</informalexample>
</para>
<simpara>
See also <function>require</function>, <function>require_once</function>,
<function>include_once</function>, <function>readfile</function>,
and <function>virtual</function>.
</simpara>
</sect1>
<sect1 id="function.require-once">
<title><function>require_once</function></title>
<para>
The <function>require_once</function> statement replaces
itself with the specified file, much like the C preprocessor's
<literal>#include</literal> works, and in that respect is
similar to the <function>require</function> statement. The main
difference is that in an inclusion chain, the use of
<function>require_once</function> will assure that the code is
added to your script only once, and avoid clashes with variable
values or function names that can happen.
</para>
<para>
For example, if you create the following 2 include files
<literal>utils.inc</literal> and <literal>foolib.inc</literal>
<example>
<title>utils.inc</title>
<programlisting role="php">
&lt;?php
define("PHPVERSION", floor(phpversion()));
echo "GLOBALS ARE NICE\n";
function goodTea()
{
return "Oolong tea tastes good!";
}
?&gt;
</programlisting>
</example>
<example>
<title>foolib.inc</title>
<programlisting role="php">
&lt;?php
require ("utils.inc");
function showVar($var)
{
if (PHPVERSION == 4) {
print_r($var);
} else {
var_dump($var);
}
}
// bunch of other functions ...
?&gt;
</programlisting>
</example>
And then you write a script <literal>cause_error_require.php</literal>
<example>
<title>cause_error_require.php</title>
<programlisting role="php">
&lt;?php
require("foolib.inc");
/* the following will generate an error */
require("utils.inc");
$foo = array("1",array("complex","quaternion"));
echo "this is requiring utils.inc again which is also\n";
echo "required in foolib.inc\n";
echo "Running goodTea: ".goodTea()."\n";
echo "Printing foo: \n";
showVar($foo);
?&gt;
</programlisting>
</example>
When you try running the latter one, the resulting ouptut will be (using
PHP 4.01pl2):
<informalexample>
<programlisting>
GLOBALS ARE NICE
GLOBALS ARE NICE
Fatal error: Cannot redeclare goodTea() in utils.inc on line 5
</programlisting>
</informalexample>
By modifying <literal>foolib.inc</literal> and
<literal>cause_errror_require.php</literal>
to use <function>require_once</function>
instead of <function>require</function> and renaming the
last one to <literal>avoid_error_require_once.php</literal>, we have:
<example>
<title>foolib.inc (fixed)</title>
<programlisting role="php">
...
require_once("utils.inc");
function showVar($var)
{
...
</programlisting>
</example>
<example>
<title>avoid_error_require_once.php</title>
<programlisting role="php">
...
require_once("foolib.inc");
require_once("utils.inc");
$foo = array("1",array("complex","quaternion"));
...
</programlisting>
</example>
And when running the latter, the output will be (using PHP 4.0.1pl2):
<informalexample>
<programlisting>
GLOBALS ARE NICE
this is requiring globals.inc again which is also
required in foolib.inc
Running goodTea: Oolong tea tastes good!
Printing foo:
Array
(
[0] =&gt; 1
[1] =&gt; Array
(
[0] =&gt; complex
[1] =&gt; quaternion
)
)
</programlisting>
</informalexample>
</para>
<para>
Also note that, analogous to the behavior of the
<literal>#include</literal> of the C preprocessor, this statement
acts at "compile time", e.g. when the script is parsed and before it
is executed, and should not be used for parts of the script that need
to be inserted dynamically during its execution. You should use
<function>include_once</function> or <function>include</function>
for that purpose.
</para>
<para>
For more examples on using <function>require_once</function> and
<function>include_once</function>, look at the PEAR code included in
the latest PHP source code distributions.
</para>
<para>
See also: <function>require</function>,
<function>include</function>, <function>include_once</function>,
<function>get_required_files</function>,
<function>get_included_files</function>, <function>readfile</function>,
and <function>virtual</function>.
</para>
</sect1>
<sect1 id="function.include-once">
<title><function>include_once</function></title>
<para>
The <function>include_once</function> statement includes and evaluates
the specified file during the execution of the script.
This is a behavior similar to the <function>include</function> statement,
with the important difference that if the code from a file has already
been included, it will not be included again.
</para>
<para>
As mentioned in the <function>require_once</function> description, the
<function>include_once</function> should be used in the cases in which
the same file might be included and evaluated more than once during a
particular execution of a script, and you want to be sure that it is
included exactly once to avoid problems with function redefinitions,
variable value reassignments, etc.
</para>
<para>
For more examples on using <function>require_once</function> and
<function>include_once</function>, look at the PEAR code included in
the latest PHP source code distributions.
</para>
<para>
<function>include_once</function> was added in PHP 4.0.1pl2
</para>
<para>
See also: <function>require</function>,
<function>include</function>, <function>require_once</function>,
<function>get_required_files</function>,
<function>get_included_files</function>, <function>readfile</function>,
and <function>virtual</function>.
</para>
</sect1>
</chapter>
<!-- Keep this comment at the end of the file
Local variables:
mode: sgml
sgml-omittag:t
sgml-shorttag:t
sgml-minimize-attributes:nil
sgml-always-quote-attributes:t
sgml-indent-step:1
sgml-indent-data:t
sgml-parent-document:nil
sgml-default-dtd-file:"../../manual.ced"
sgml-exposed-tags:nil
sgml-local-catalogs:nil
sgml-local-ecat-files:nil
End:
vim600: syn=xml fen fdm=syntax fdl=2 si
vim: et tw=78 syn=sgml
vi: ts=1 sw=1
-->