<?xml version="1.0" encoding="iso-8859-1"?>
<!-- $Revision: 1.75 $ -->
 <reference id="ref.math">
  <title>Mathematical Functions</title>
  <titleabbrev>Math</titleabbrev>
  
  <partintro>
   <sect1 id="math.intro">
    <title>Introduction</title>
    <para>
     These math functions will only handle values within the range of
     the <type>integer</type> and <type>float</type> types on your computer.  
     (this corresponds currently to the C types long resp. double)
     If you need to handle bigger numbers, take a look at the <link
     linkend="ref.bc">arbitrary precision math functions</link>.
    </para>
    <sect2 id="math.constants">
     <title>Math constants</title>
     <para>
      The following values are defined as constants in PHP by the math
      extension:
      <table>
       <title>Math constants</title>
       <tgroup cols="3">
    <thead>
     <row>
      <entry>Constant</entry>
      <entry>Value</entry>
      <entry>Description</entry>
     </row>
    </thead>
    <tbody>
    <row>
     <entry>M_PI</entry>
     <entry>3.14159265358979323846</entry>
     <entry>Pi</entry>
    </row>
    <row>
     <entry>M_E</entry>
     <entry>2.7182818284590452354</entry>
     <entry>e</entry>
    </row>
    <row>
     <entry>M_LOG2E</entry> 
     <entry>1.4426950408889634074</entry>   
     <entry>log_2 e</entry>
    </row>
    <row>
     <entry>M_LOG10E</entry>
     <entry>0.43429448190325182765</entry>  
     <entry>log_10 e</entry>
    </row>
    <row>
     <entry>M_LN2</entry>   
     <entry>0.69314718055994530942</entry>      
     <entry>log_e 2</entry>
    </row>
    <row>
     <entry>M_LN10</entry>  
     <entry>2.30258509299404568402</entry>  
     <entry>log_e 10</entry>
    </row>
    <row>
     <entry>M_PI_2</entry>
     <entry>1.57079632679489661923</entry>  
     <entry>pi/2</entry>
    </row>
    <row>
     <entry>M_PI_4</entry>
    <entry>0.78539816339744830962</entry>   
     <entry>pi/4</entry>
    </row>
    <row>
     <entry>M_1_PI</entry>
     <entry>0.31830988618379067154</entry>  
     <entry>1/pi</entry>
    </row>
    <row>
     <entry>M_2_PI</entry>
     <entry>0.63661977236758134308</entry>  
     <entry>2/pi</entry>
    </row>
    <row>
     <entry>M_SQRTPI</entry>    
     <entry>1.77245385090551602729</entry>  
     <entry>sqrt(pi) [4.0.2]</entry>
    </row>
    <row>
     <entry>M_2_SQRTPI</entry>  
     <entry>1.12837916709551257390</entry>  
     <entry>2/sqrt(pi)</entry>
    </row>
    <row>
     <entry>M_SQRT2</entry> 
     <entry>1.41421356237309504880</entry>  
     <entry>sqrt(2)</entry>
    </row>
    <row>
     <entry>M_SQRT3</entry> 
     <entry>1.73205080756887729352</entry>  
     <entry>sqrt(3) [4.0.2]</entry>
    </row>
    <row>
     <entry>M_SQRT1_2</entry>   
     <entry>0.70710678118654752440</entry>  
     <entry>1/sqrt(2)</entry>
    </row>
    <row>
     <entry>M_LNPI</entry>  
     <entry>1.14472988584940017414</entry>  
     <entry>log_e(pi) [4.0.2]</entry>
    </row>
    <row>
     <entry>M_EULER</entry> 
     <entry>0.57721566490153286061</entry>  
     <entry>Euler constant [4.0.2]</entry>
    </row>
    </tbody>
       </tgroup>
      </table>
     Only M_PI is available in PHP versions up to and including PHP 4.0.0.
     All other constants are available starting with PHP 4.0.0.  Constants
     labeled [4.0.2] were added in PHP 4.0.2.
     </para>
    </sect2>
   </sect1>
  </partintro>

  <refentry id="function.abs">
   <refnamediv>
    <refname>abs</refname>
    <refpurpose>Absolute value</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>mixed</type><methodname>abs</methodname>
      <methodparam><type>mixed</type><parameter>number</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the absolute value of number.  If the argument number is
     of type <type>float</type>, the return type is also <type>float</type>, 
     otherwise it is <type>integer</type> (as <type>float</type> usually has a
     bigger value range than <type>integer</type>).
    </para>
    <example>
     <title><function>abs</function> example</title>
     <programlisting role="php">
<![CDATA[
$abs = abs(-4.2); // $abs = 4.2; (double/float)
$abs2 = abs(5);   // $abs2 = 5; (integer)
$abs3 = abs(-5);  // $abs3 = 5; (integer)
]]>
     </programlisting>
    </example>
   </refsect1>
  </refentry>

  <refentry id="function.acos">
   <refnamediv>
    <refname>acos</refname>
    <refpurpose>Arc cosine</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>acos</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the arc cosine of <parameter>arg</parameter> in radians.
    </para>
    <para>
     See also <function>acosh</function>, <function>asin</function>
     and <function>atan</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.acosh">
   <refnamediv>
    <refname>acosh</refname>
    <refpurpose>Inverse hyperbolic cosine</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>acosh</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the inverse hyperbolic cosine of
     <parameter>arg</parameter>, i.e. the value whose
     hyperbolic cosine is <parameter>arg</parameter>.
    </para>
    &note.no-windows;
    <para>
     See also <function>acos</function>, <function>asin</function>
     and <function>atan</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.asin">
   <refnamediv>
    <refname>asin</refname>
    <refpurpose>Arc sine</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>asin</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the arc sine of arg in radians.
    </para>
    <para>
     See also <function>asinh</function>, <function>acos</function>
     and <function>atan</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.asinh">
   <refnamediv>
    <refname>asinh</refname>
    <refpurpose>Inverse hyperbolic sine</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>asinh</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the inverse hyperbolic sine of
     <parameter>arg</parameter>, i.e. the value whose
     hyperbolic sine is <parameter>arg</parameter>.
    </para>
    &note.no-windows;
    <para>
     See also <function>asin</function>, <function>acos</function>
     and <function>atan</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.atan">
   <refnamediv>
    <refname>atan</refname>
    <refpurpose>Arc tangent</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>atan</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the arc tangent of arg in radians.
    </para>
    <para>
     See also <function>atanh</function>, <function>asin</function>
     and <function>acos</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.atanh">
   <refnamediv>
    <refname>atanh</refname>
    <refpurpose>Inverse hyperbolic tangent</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>atanh</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the inverse hyperbolic tangent of
     <parameter>arg</parameter>, i.e. the value whose
     hyperbolic tangent is <parameter>arg</parameter>.
    </para>
    &note.no-windows;
    <para>
     See also <function>atan</function>, <function>asin</function>
     and <function>acos</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.atan2">
   <refnamediv>
    <refname>atan2</refname>
    <refpurpose>arc tangent of two variables</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>atan2</methodname>
      <methodparam><type>float</type><parameter>y</parameter></methodparam>
      <methodparam><type>float</type><parameter>x</parameter></methodparam>
     </methodsynopsis>
    <simpara>
     This function calculates the arc tangent of the two variables
     <parameter>x</parameter> and <parameter>y</parameter>.  It is
     similar to calculating the arc tangent of
     <parameter>y</parameter> / <parameter>x</parameter>, except that
     the signs of both arguments are used to determine the quadrant of
     the result.
    </simpara>
    <simpara>
     The function returns the result in radians, which is between -PI
     and PI (inclusive).
    </simpara>
    <para>
     See also <function>acos</function> and <function>atan</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.base-convert">
   <refnamediv>
    <refname>base_convert</refname>
    <refpurpose>Convert a number between arbitrary bases</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>string</type><methodname>base_convert</methodname>
      <methodparam><type>string</type><parameter>number</parameter></methodparam>
      <methodparam><type>int</type><parameter>frombase</parameter></methodparam>
      <methodparam><type>int</type><parameter>tobase</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns a string containing <parameter>number</parameter>
     represented in base <parameter>tobase</parameter>.  The base in
     which <parameter>number</parameter> is given is specified in
     <parameter>frombase</parameter>.  Both
     <parameter>frombase</parameter> and <parameter>tobase</parameter>
     have to be between 2 and 36, inclusive.  Digits in numbers with a
     base higher than 10 will be represented with the letters a-z,
     with a meaning 10, b meaning 11 and z meaning 35.
     <example>
      <title><function>base_convert</function></title>
      <programlisting role="php">
<![CDATA[
$binary = base_convert ($hexadecimal, 16, 2);
]]>
</programlisting>
     </example>
    </para>
   </refsect1>
  </refentry>


  <refentry id="function.bindec">
   <refnamediv>
    <refname>bindec</refname>
    <refpurpose>Binary to decimal</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>int</type><methodname>bindec</methodname>
      <methodparam><type>string</type><parameter>binary_string</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the decimal equivalent of the binary number represented by
     the binary_string argument.
    </para>
    <para>
     <function>bindec</function> converts a binary number to an 
     <type>integer</type>. The largest number that can be converted is 
     31 bits of 1's or 2147483647 in decimal.
    </para>
    <para>
     See also the <function>decbin</function>
     function.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.ceil">
   <refnamediv>
    <refname>ceil</refname>
    <refpurpose>Round fractions up</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>ceil</methodname>
      <methodparam><type>float</type><parameter>value</parameter></methodparam>
     </methodsynopsis>
    <simpara>
     Returns the next highest integer value by rounding up 
     <parameter>value</parameter> if necessary. 
     The return value of <function>ceil</function> is still of type
     <type>float</type> as the value range of <type>float</type> is 
     usually bigger than that of <type>int</type>.
    </simpara>
    <example>
     <title><function>ceil</function> examples</title>
     <programlisting role="php">
<![CDATA[
$five = ceil(4.3);   // $five = 5.0;
$ten = ceil(9.999); // $ten = 10.0;
]]>
     </programlisting>
    </example>
    <simpara>
     See also <function>floor</function> and
     <function>round</function>.
    </simpara>
   </refsect1>
  </refentry>

  <refentry id="function.cos">
   <refnamediv>
    <refname>cos</refname>
    <refpurpose>Cosine</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>cos</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the cosine of arg in radians.
    </para>
    <para>
     See also <function>sin</function> and <function>tan</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.cosh">
   <refnamediv>
    <refname>cosh</refname>
    <refpurpose>Hyperbolic cosine</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>cosh</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the hyperbolic cosine of <parameter>arg</parameter>,
     defined as <literal>(exp(arg) + exp(-arg))/2</literal>.
    </para>
    <para>
     See also <function>cos</function>, <function>acosh</function>,
     <function>sin</function> and <function>tan</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.decbin">
   <refnamediv>
    <refname>decbin</refname>
    <refpurpose>Decimal to binary</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>string</type><methodname>decbin</methodname>
      <methodparam><type>int</type><parameter>number</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns a string containing a binary representation of the given
     number argument.  The largest number that can be converted is
     4294967295 in decimal resulting to a string of 32 1's.
    </para>
    <para>
     See also the <function>bindec</function> function.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.dechex">
   <refnamediv>
    <refname>dechex</refname>
    <refpurpose>Decimal to hexadecimal</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>string</type><methodname>dechex</methodname>
      <methodparam><type>int</type><parameter>number</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns a string containing a hexadecimal representation of the
     given number argument.  The largest number that can
     be converted is 2147483647 in decimal resulting to "7fffffff".
    </para>
    <para>
     See also <function>hexdec</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.decoct">
   <refnamediv>
    <refname>decoct</refname>
    <refpurpose>Decimal to octal</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>string</type><methodname>decoct</methodname>
      <methodparam><type>int</type><parameter>number</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns a string containing an octal representation of the given
     number argument.  The largest number that can be converted is
     2147483647 in decimal resulting to "17777777777". 
    </para>
    <para>
     See also <function>octdec</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.deg2rad">
   <refnamediv>
    <refname>deg2rad</refname>
    <refpurpose>
     Converts the number in degrees to the radian equivalent
    </refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>deg2rad</methodname>
      <methodparam><type>float</type><parameter>number</parameter></methodparam>
     </methodsynopsis>
    <para>
     This function converts <parameter>number</parameter> from degrees
     to the radian equivalent.
    </para>
    <para>
     See also <function>rad2deg</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.exp">
   <refnamediv>
    <refname>exp</refname>
    <refpurpose>e to the power of ...</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>exp</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns e raised to the power of <parameter>arg</parameter>.
    </para>
    <para>
     See also <function>pow</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id='function.expm1'>
   <refnamediv>
    <refname>expm1</refname>
    <refpurpose>
     Returns exp(number) - 1, computed in a way that accurate even
     when the value of number is close to zero
    </refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>expm1</methodname>
      <methodparam><type>float</type><parameter>number</parameter></methodparam>
     </methodsynopsis>
     &warn.experimental.func;
    <para>
     &warn.undocumented.func;
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.floor">
   <refnamediv>
    <refname>floor</refname>
    <refpurpose>Round fractions down</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>floor</methodname>
      <methodparam><type>float</type><parameter>value</parameter></methodparam>
     </methodsynopsis>
    <simpara>
     Returns the next lowest integer value by rounding down
     <parameter>value</parameter> if necessary. 
     The return value of <function>floor</function> is still of type
     <type>float</type> because the value range of <type>float</type> is 
     usually bigger than that of <type>int</type>.
    </simpara>
    <example>
     <title><function>floor</function> examples</title>
     <programlisting role="php">
<![CDATA[
$four = floor(4.3);   // $four = 4.0;
$nine = floor(9.999); // $nine = 9.0;
]]>
     </programlisting>
    </example>
    <simpara>
     See also <function>ceil</function> and
     <function>round</function>.
    </simpara>
   </refsect1>
  </refentry>

  <refentry id="function.getrandmax">
   <refnamediv>
    <refname>getrandmax</refname>
    <refpurpose>Show largest possible random value</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>int</type><methodname>getrandmax</methodname>
      <void/>
     </methodsynopsis>
    <simpara>
     Returns the maximum value that can be returned by a call to
     <function>rand</function>.
    </simpara>
    <simpara>
     See also <function>rand</function>, <function>srand</function>,
     <function>mt_rand</function>, <function>mt_srand</function>, and
     <function>mt_getrandmax</function>.
    </simpara>
   </refsect1>
  </refentry>

  <refentry id="function.hexdec">
   <refnamediv>
    <refname>hexdec</refname>
    <refpurpose>Hexadecimal to decimal</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>int</type><methodname>hexdec</methodname>
      <methodparam><type>string</type><parameter>hex_string</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the decimal equivalent of the hexadecimal number
     represented by the hex_string argument.  <function>hexdec</function> converts a
     hexadecimal string to a decimal number. The largest number that
     can be converted is 7fffffff or 2147483647 in decimal.
    </para>
    <para>
     <function>hexdec</function> will replace of any non-hexadecimal 
     characters it encounters by <literal>0</literal>. This way, all
     left zeros are ignored, but right zeros will be valued.
     <example>
      <title><function>hexdec</function> example</title>
      <programlisting role="php">
<![CDATA[
var_dump(hexdec("See"));
var_dump(hexdec("ee"));
// both prints "int(238)"

var_dump(hexdec("that"));
var_dump(hexdec("a0"));
// both prints int(160)
]]>
      </programlisting>
     </example>
    </para>
    <para>
     See also <function>dechex</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id='function.hypot'>
   <refnamediv>
    <refname>hypot</refname>
    <refpurpose>
     Returns sqrt( num1*num1 + num2*num2)
    </refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>hypot</methodname>
      <methodparam><type>float</type><parameter>num1</parameter></methodparam>
      <methodparam><type>float</type><parameter>num2</parameter></methodparam>
     </methodsynopsis>
    &warn.experimental.func;
    <para>
     &warn.undocumented.func;
    </para>
   </refsect1>
  </refentry>

  <refentry id='function.is-finite'>
   <refnamediv>
    <refname>is_finite</refname>
    <refpurpose>
    </refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>bool</type><methodname>is_finite</methodname>
      <methodparam><type>float</type><parameter>val</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns &true; if <parameter>val</parameter> is a legal finite 
     number within the allowed range for a PHP float on this platform.
    </para>
   </refsect1>
  </refentry>

  <refentry id='function.is-infinite'>
   <refnamediv>
    <refname>is_infinite</refname>
    <refpurpose>
    </refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>bool</type><methodname>is_infinite</methodname>
      <methodparam><type>float</type><parameter>val</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns &true; if <parameter>val</parameter> is infinite (positive or 
     negative), like the result of <literal>log(0)</literal> or any value too 
     big to fit into a float on this platform.
    </para>
   </refsect1>
  </refentry>

  <refentry id='function.is-nan'>
   <refnamediv>
    <refname>is_nan</refname>
    <refpurpose>
    </refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>bool</type><methodname>is_nan</methodname>
      <methodparam><type>float</type><parameter>val</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns &true; if <parameter>val</parameter> is 'not a number', 
     like the result of <literal>acos(1.01)</literal>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.lcg-value">
   <refnamediv>
    <refname>lcg_value</refname>
    <refpurpose>Combined linear congruential generator</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>lcg_value</methodname>
      <void/>
     </methodsynopsis>
    <para>
     <function>lcg_value</function> returns a pseudo random number in
     the range of (0, 1).  The function combines two CGs with periods
     of 2^31 - 85 and 2^31 - 249. The period of this function is equal
     to the product of both primes.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.log">
   <refnamediv>
    <refname>log</refname>
    <refpurpose>Natural logarithm</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>log</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the natural logarithm of arg.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.log10">
   <refnamediv>
    <refname>log10</refname>
    <refpurpose>Base-10 logarithm</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>log10</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the base-10 logarithm of <parameter>arg</parameter>.
    </para>
   </refsect1>
  </refentry>

  <refentry id='function.log1p'>
   <refnamediv>
    <refname>log1p</refname>
    <refpurpose>
     Returns log(1 + number), computed in a way that accurate even when the val
ue of number is close to zero
    </refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>log1p</methodname>
      <methodparam><type>float</type><parameter>number</parameter></methodparam>
     </methodsynopsis>
    &warn.experimental.func;
    <para>
     &warn.undocumented.func;
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.max">
   <refnamediv>
    <refname>max</refname>
    <refpurpose>Find highest value</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>mixed</type><methodname>max</methodname>
      <methodparam><type>mixed</type><parameter>arg1</parameter></methodparam>
      <methodparam><type>mixed</type><parameter>arg2</parameter></methodparam>
      <methodparam><type>mixed</type><parameter>argn</parameter></methodparam>
     </methodsynopsis>
    <para>
     <function>max</function> returns the numerically highest of the
     parameter values.
    </para>
    <para>
     If the first parameter is an array, <function>max</function>
     returns the highest value in that array.  If the first parameter
     is an integer, string or float, you need at least two parameters
     and <function>max</function> returns the biggest of these values.
     You can compare an unlimited number of values.
    </para>
    <para>
     If one or more of the values is a float, all the values will be
     treated as floats, and a float is returned.  If none of the
     values is a float, all of them will be treated as integers, and
     an integer is returned.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.min">
   <refnamediv>
    <refname>min</refname>
    <refpurpose>Find lowest value</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>number</type><methodname>min</methodname>
      <methodparam><type>number</type><parameter>arg1</parameter></methodparam>
      <methodparam><type>number</type><parameter>arg2</parameter></methodparam>
      <methodparam choice="opt"><parameter>...</parameter></methodparam>
     </methodsynopsis>
     <methodsynopsis>
      <type>number</type><methodname>min</methodname>
      <methodparam><type>array</type><parameter>numbers</parameter></methodparam>
     </methodsynopsis>
    <para>
     <function>min</function> returns the numerically lowest of the
     parameter values.
    </para>
    <para>
     In the first variant, you need at least two parameters
     and <function>min</function> returns the lowest of these values.
     You can compare an unlimited number of values.
    </para>
    <para>
     In the second variant, <function>min</function>
     returns the lowest value in <parameter>numbers</parameter>.  
    </para>
    <para>
     If one or more of the values is a <type>float</type>, all the values 
     will be
     treated as floats, and a float is returned.  If none of the
     values is a float, all of them will be treated as <type>integer</type>s, 
     and an integer is returned.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.mt-rand">
   <refnamediv>
    <refname>mt_rand</refname>
    <refpurpose>Generate a better random value</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>int</type><methodname>mt_rand</methodname>
      <void/>
     </methodsynopsis>
     <methodsynopsis>
      <type>int</type><methodname>mt_rand</methodname>
      <methodparam><type>int</type><parameter>min</parameter></methodparam>
      <methodparam><type>int</type><parameter>max</parameter></methodparam>
     </methodsynopsis>
    <simpara>
     Many random number generators of older libcs have dubious or
     unknown characteristics and are slow. By default, PHP uses the
     libc random number generator with the <function>rand</function>
     function.  <function>mt_rand</function> function is a drop-in
     replacement for this.  It uses a random number generator with
     known characteristics, the Mersenne Twister, which will produce
     random numbers that should be suitable for seeding some kinds
     of cryptography (see the home pages for details) and is four
     times faster than what the average libc provides. The Homepage
     of the Mersenne Twister can be found at
     <ulink url="&url.mersenne;">&url.mersenne;</ulink>, and an
     optimized version of the MT source is available from
     <ulink url="&url.mersenne.twister;">&url.mersenne.twister;
     </ulink>.
    </simpara>
    <simpara>
     If called without the optional <parameter>min</parameter>,
     <parameter>max</parameter> arguments <function>mt_rand</function>
     returns a pseudo-random value between 0 and
     <constant>RAND_MAX</constant>.  If you want a random number
     between 5 and 15 (inclusive), for example, use <literal>mt_rand
     (5, 15)</literal>.
    </simpara>
    <simpara>
     Remember to seed the random number generator before use with
     <function>mt_srand</function>.
    </simpara>
    <note>
     <para>
      In versions before 3.0.7 the meaning of
      <parameter>max</parameter> was <parameter>range</parameter>. To
      get the same results in these versions the short example should
      be <literal>mt_rand (5, 11)</literal> to get a random number
      between 5 and 15.
     </para>
    </note>
    <simpara>
     See also <function>mt_srand</function>,
     <function>mt_getrandmax</function>, <function>srand</function>,
     <function>rand</function> and <function>getrandmax</function>.
    </simpara>
   </refsect1>
  </refentry>

  <refentry id="function.mt-srand">
   <refnamediv>
    <refname>mt_srand</refname>
    <refpurpose>Seed the better random number generator</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>void</type><methodname>mt_srand</methodname>
      <methodparam><type>int</type><parameter>seed</parameter></methodparam>
     </methodsynopsis>
    <para>
     Seeds the random number generator with
     <parameter>seed</parameter>.
     <informalexample>
      <programlisting role="php">
<![CDATA[
// seed with microseconds
function make_seed() {
    list($usec, $sec) = explode(' ', microtime());
    return (float) $sec + ((float) $usec * 100000);
}
mt_srand(make_seed());
$randval = mt_rand();
]]>
      </programlisting>
     </informalexample>
    </para>
    <simpara>
     See also <function>mt_rand</function>,
     <function>mt_getrandmax</function>, <function>srand</function>,
     <function>rand</function>, and
     <function>getrandmax</function>.
    </simpara>
   </refsect1>
  </refentry>

  <refentry id="function.mt-getrandmax">
   <refnamediv>
    <refname>mt_getrandmax</refname>
    <refpurpose>Show largest possible random value</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>int</type><methodname>mt_getrandmax</methodname>
      <void/>
     </methodsynopsis>
    <simpara>
     Returns the maximum value that can be returned by a call to
     <function>mt_rand</function>.
    </simpara>
    <simpara>
     See also <function>mt_rand</function>,
     <function>mt_srand</function>, <function>rand</function>,
     <function>srand</function>, and
     <function>getrandmax</function>.
    </simpara>
   </refsect1>
  </refentry>

  <refentry id="function.number-format">
   <refnamediv>
    <refname>number_format</refname>
    <refpurpose>Format a number with grouped thousands</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>string</type><methodname>number_format</methodname>
      <methodparam><type>float</type><parameter>number</parameter></methodparam>
      <methodparam choice="opt"><type>int</type><parameter>decimals</parameter></methodparam>
      <methodparam choice="opt"><type>string</type><parameter>dec_point</parameter></methodparam>
      <methodparam choice="opt"><type>string</type><parameter>thousands_sep</parameter></methodparam>
     </methodsynopsis>
    <para>
     <function>number_format</function> returns a formatted version of
     <parameter>number</parameter>.  This function accepts either one,
     two or four parameters (not three):
    </para>
    <para>If only one parameter is given,
     <parameter>Number</parameter> will be formatted without decimals,
     but with a comma (",") between every group of thousands.
    </para>
    <para>
     If two parameters are given, <parameter>number</parameter> will
     be formatted with <parameter>decimals</parameter> decimals with a
     dot (".") in front, and a comma (",") between every group of
     thousands.
    </para>
    <para>
     If all four parameters are given, <parameter>number</parameter>
     will be formatted with <parameter>decimals</parameter> decimals,
     <parameter>dec_point</parameter> instead of a dot (".") before
     the decimals and <parameter>thousands_sep</parameter> instead of
     a comma (",") between every group of thousands.
    </para>
    <note>
     <para>
      Only the first character of <parameter>thousands_sep</parameter>
      is used.  For example, if you use <literal>foo</literal> as
      <parameter>thousands_sep</parameter> on the number
      <literal>1000</literal>, <function>number_format</function> will
      return <literal>1f000</literal>.
     </para>
    </note>
    <para>
    <example>
     <title><function>number_format</function> Example</title>
     <para>
      For instance, French notation usually use two decimals,
      comma (',') as decimal separator, and space (' ') as
      thousand separator. This is achieved with this line :
     </para>
      <programlisting role="php">
<![CDATA[
<?php

    $number = 1234.56;

    // english notation (default)
    $english_format_number = number_format($number);
    // 1,234.56

    // French notation
    $nombre_format_francais = number_format($number, 2, ',', ' ');
    // 1 234,56


    $number = 1234.5678;

    // english notation without thousands seperator
    $english_format_number = number_format($number, 2, '.', '');
    // 1234.56

?>
]]>
      </programlisting>
     </example>
    </para>
    <note>
     <para>
      See also: <function>sprintf</function>,
      <function>printf</function> and <function>sscanf</function>.
     </para>
    </note>
   </refsect1>
  </refentry>

  <refentry id="function.octdec">
   <refnamediv>
    <refname>octdec</refname>
    <refpurpose>Octal to decimal</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>int</type><methodname>octdec</methodname>
      <methodparam><type>string</type><parameter>octal_string</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the decimal equivalent of the octal number
     represented by the <parameter>octal_string</parameter> argument.
     The largest number that can be converted is 17777777777 or 2147483647 in
     decimal.
    </para>
    <para>
     See also <function>decoct</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.pi">
   <refnamediv>
    <refname>pi</refname>
    <refpurpose>Get value of pi</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>pi</methodname>
      <void/>
     </methodsynopsis>
    <simpara>
     Returns an approximation of pi.  The returned <type>float</type> 
     has a precision based on the 
     <link linkend="ini.precision">precision</link> directive in
     <filename>php.ini</filename>, which defaults to <literal>14</literal>.  
     Also, you can use the <constant>M_PI</constant> constant which yields 
     identical results to <function>pi</function>.
    </simpara>
   <para>
    <informalexample>
     <programlisting>
<![CDATA[
echo pi(); // 3.1415926535898
echo M_PI; // 3.1415926535898
]]>
     </programlisting>
    </informalexample>
   </para>
   </refsect1>
  </refentry>

  <refentry id="function.pow">
   <refnamediv>
    <refname>pow</refname>
    <refpurpose>Exponential expression</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>number</type><methodname>pow</methodname>
      <methodparam><type>number</type><parameter>base</parameter></methodparam>
      <methodparam><type>number</type><parameter>exp</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns <parameter>base</parameter> raised to the power of 
     <parameter>exp</parameter>. If possible, this function will
     return an <type>integer</type>.
    </para>
    <para>
     If the power cannot be computed, a warning will be issued,
     and <function>pow</function> will return &false;.
    </para>
    <example>
     <title>Some examples of <function>pow</function></title>
     <programlisting role="php">
<![CDATA[
<?php

var_dump( pow(2,8) ); // int(256)
echo pow(-1,20); // 1
echo pow(0, 0); // 1

echo pow(-1, 5.5); // error

?>
]]>
     </programlisting>
    </example>
    <warning>
     <para>
      In PHP 4.0.6 and earlier <function>pow</function> always returned
      a <type>float</type>, and did not issue warnings.
     </para>
    </warning>
    <para> 
     See also <function>exp</function> and <function>sqrt</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.rad2deg">
   <refnamediv>
    <refname>rad2deg</refname>
    <refpurpose>
     Converts the radian number to the equivalent number in degrees
    </refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>rad2deg</methodname>
      <methodparam><type>float</type><parameter>number</parameter></methodparam>
     </methodsynopsis>
    <para>
     This function converts <parameter>number</parameter> from radian
     to degrees.
    </para>
    <para>
     See also <function>deg2rad</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.rand">
   <refnamediv>
    <refname>rand</refname>
    <refpurpose>Generate a random value</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>int</type><methodname>rand</methodname>
      <void/>
     </methodsynopsis>
     <methodsynopsis>
      <type>int</type><methodname>rand</methodname>
      <methodparam><type>int</type><parameter>min</parameter></methodparam>
      <methodparam><type>int</type><parameter>max</parameter></methodparam>
     </methodsynopsis>
    <simpara>
     If called without the optional <parameter>min</parameter>,
     <parameter>max</parameter> arguments <function>rand</function>
     returns a pseudo-random value between 0 and
     <constant>RAND_MAX</constant>.  If you want a random number
     between 5 and 15 (inclusive), for example, use <literal>rand (5,
     15)</literal>. 
    </simpara>
    <simpara>
     Remember to seed the random number generator before use with
     <function>srand</function>.
    </simpara>
    <note>
     <para>
      In versions before 3.0.7 the meaning of
      <parameter>max</parameter> was <parameter>range</parameter>. To
      get the same results in these versions the short example should
      be <literal>rand (5, 11)</literal> to get a random number
      between 5 and 15.
     </para>
    </note>
    <simpara>
     See also <function>srand</function>,
     <function>getrandmax</function>, <function>mt_rand</function>,
     <function>mt_srand</function>, and
     <function>mt_getrandmax</function>.
    </simpara>
   </refsect1>
  </refentry>

  <refentry id="function.round">
   <refnamediv>
    <refname>round</refname>
    <refpurpose>Rounds a float</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>round</methodname>
      <methodparam><type>float</type><parameter>val</parameter></methodparam>
      <methodparam choice="opt"><type>int</type><parameter>precision</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the rounded value of <parameter>val</parameter> to
     specified <parameter>precision</parameter> 
     (number of digits after the decimal point).
     <parameter>precision</parameter> can also be negative or zero (default).
    </para>
    <para>
     <caution>
      <simpara>
       PHP doesn't handle strings like <literal>"12,300.2"</literal> correctly
       by default. See <link linkend="language.types.string.conversion"
       >converting from strings</link>.
      </simpara>
     </caution>
    </para>
    <para>
     <informalexample>
      <programlisting role="php">
<![CDATA[
$foo = round(3.4);  // $foo == 3.0
$foo = round(3.5);  // $foo == 4.0
$foo = round(3.6);  // $foo == 4.0
$foo = round(3.6, 0);  // equivalent with above

$foo = round(1.95583, 2);  // $foo == 1.96

$foo = round(1241757, -3);  // $foo == 1242000
]]>
      </programlisting>
     </informalexample>
    </para>
    <note>
     <simpara> 
      The <parameter>precision</parameter> parameter is only
      available in PHP 4. 
     </simpara>
    </note>
    <simpara>
     See also <function>ceil</function> and
     <function>floor</function>.
    </simpara>
   </refsect1>
  </refentry>

  <refentry id="function.sin">
   <refnamediv>
    <refname>sin</refname>
    <refpurpose>Sine</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>sin</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the sine of arg in radians.
    </para>
    <para>
     See also <function>cos</function> and <function>tan</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.sinh">
   <refnamediv>
    <refname>sinh</refname>
    <refpurpose>Hyperbolic sine</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>sinh</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the hyperbolic sine of <parameter>arg</parameter>,
     defined as <literal>(exp(arg) - exp(-arg))/2</literal>.
    </para>
    <para>
     See also <function>sin</function>, <function>asinh</function>,
     <function>cos</function> and <function>tan</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.sqrt">
   <refnamediv>
    <refname>sqrt</refname>
    <refpurpose>Square root</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>sqrt</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the square root of <parameter>arg</parameter>.
    </para>
    <para>
     See also <function>pow</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.srand">
   <refnamediv>
    <refname>srand</refname>
    <refpurpose>Seed the random number generator</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>void</type><methodname>srand</methodname>
      <methodparam><type>int</type><parameter>seed</parameter></methodparam>
     </methodsynopsis>
    <para>
     Seeds the random number generator with
     <parameter>seed</parameter>.
     <informalexample>
      <programlisting role="php">
<![CDATA[
// seed with microseconds
function make_seed() {
    list($usec, $sec) = explode(' ', microtime());
    return (float) $sec + ((float) $usec * 100000);
}
srand(make_seed());
$randval = rand();
]]>
      </programlisting>
     </informalexample>
    </para>
    <simpara>
     See also <function>rand</function>,
     <function>getrandmax</function>, <function>mt_rand</function>,
     <function>mt_srand</function>, and
     <function>mt_getrandmax</function>.
    </simpara>    
   </refsect1>
  </refentry>

  <refentry id="function.tan">
   <refnamediv>
    <refname>tan</refname>
    <refpurpose>Tangent</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>tan</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the tangent of arg in radians.
    </para>
    <para>
     See also <function>sin</function> and <function>cos</function>.
    </para>
   </refsect1>
  </refentry>

  <refentry id="function.tanh">
   <refnamediv>
    <refname>tanh</refname>
    <refpurpose>Hyperbolic tangent</refpurpose>
   </refnamediv>
   <refsect1>
    <title>Description</title>
     <methodsynopsis>
      <type>float</type><methodname>tanh</methodname>
      <methodparam><type>float</type><parameter>arg</parameter></methodparam>
     </methodsynopsis>
    <para>
     Returns the hyperbolic tangent of <parameter>arg</parameter>,
     defined as <literal>sinh(arg)/cosh(arg)</literal>.
    </para>
    <para>
     See also <function>tan</function>, <function>atanh</function>,
     <function>sin</function> and <function>cos</function>.
    </para>
   </refsect1>
  </refentry>

 </reference>

<!-- 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
indent-tabs-mode:nil
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
-->