Add a tip regrading array iteration with for (Thanks for the help Pierre)

git-svn-id: https://svn.php.net/repository/phpdoc/en/trunk@265700 c90b9560-bf6c-de11-be94-00142212c4b1

language/control-structures.xml

@@ -1,5 +1,5 @@
 <?xml version="1.0" encoding="iso-8859-1"?>
-<!-- $Revision: 1.165 $ -->
+<!-- $Revision: 1.166 $ -->
  <chapter xml:id="language.control-structures" xmlns="http://docbook.org/ns/docbook" xmlns:xlink="http://www.w3.org/1999/xlink">
   <title>Control Structures</title>
 
@@ -524,7 +524,60 @@ for (expr1; expr2; expr3):
 endfor;
 ]]>
      </programlisting>
-     </informalexample>
+    </informalexample>
+   </para>
+   <simpara>
+    Its a common thing to many users to iterate though arrays like in the 
+    example below.
+   </simpara>
+   <para>
+    <informalexample>
+     <programlisting>
+<![CDATA[
+<?php
+/*
+ * This is an array with some data we want to modify 
+ * when running though the for loop.
+ */
+$people = Array(
+        Array('name' => 'Kalle', 'salt' => 856412), 
+        Array('name' => 'Pierre', 'salt' => 215863)
+        );
+
+for($i = 0; $i < sizeof($people); ++$i)
+{
+    $people[$i]['salt'] = rand(000000, 999999);
+}
+?>
+]]>
+     </programlisting>
+    </informalexample>
+   </para>
+   <simpara>
+    The problem lies in the second for expression. This  code can be slow 
+    because it has to calculate the size of the array on each iteration. 
+    Since the size never change, it can be optimized easily using an 
+    intermediate variable to store the size  and use in the loop instead 
+    of sizeof. The example below illustrates this:
+   </simpara>
+   <para>
+    <informalexample>
+     <programlisting>
+<![CDATA[
+<?php
+$people = Array(
+        Array('name' => 'Kalle', 'salt' => 856412), 
+        Array('name' => 'Pierre', 'salt' => 215863)
+        );
+
+for($i = 0, $size = sizeof($people); $i < $size; ++$i)
+{
+    $people[$i]['salt'] = rand(000000, 999999);
+}
+?>
+]]>
+     </programlisting>
+    </informalexample>
    </para>
   </sect1>