File: mysqlqb_functions.html

package info (click to toggle)
mysql-gui-tools 5.0r14%2BopenSUSE-2.1
links: PTS, VCS
area: main
in suites: squeeze
size: 116,956 kB
ctags: 48,715
sloc: sql: 341,918; pascal: 276,698; ansic: 91,020; cpp: 90,451; objc: 33,236; sh: 29,481; yacc: 10,756; xml: 10,589; java: 10,079; php: 2,806; python: 2,092; makefile: 1,783; perl: 4
file content (6983 lines) | stat: -rw-r--r-- 571,178 bytes
parent folder | download | duplicates (2)
<?xml version="1.0" encoding="utf-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8" /><title>Functions and Operators</title><link rel="stylesheet" href="main.css" type="text/css" /><meta name="generator" content="DocBook XSL Stylesheets V1.69.1" /><link rel="start" href="index.html" title="Query Browser Help" /><link rel="up" href="index.html" title="Query Browser Help" /><link rel="prev" href="index.html" title="Query Browser Help" /><link rel="next" href="mysqlqb_statements.html" title="SQL Statement Syntax" /></head><body><div class="navheader"><table width="100%" summary="Navigation header"><tr><th colspan="3" align="center">Functions and Operators</th></tr><tr><td width="20%" align="left"><a accesskey="p" href="index.html">Prev</a> </td><th width="60%" align="center"> </th><td width="20%" align="right"> <a accesskey="n" href="mysqlqb_statements.html">Next</a></td></tr></table><hr /></div><div class="chapter" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title"><a id="functions"></a>Functions and Operators</h2></div></div></div><div class="toc"><p><b>Table of Contents</b></p><dl><dt><span class="section"><a href="mysqlqb_functions.html#func-op-summary-ref">Operator and Function Reference</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#non-typed-operators">Operators</a></span></dt><dd><dl><dt><span class="section"><a href="mysqlqb_functions.html#operator-precedence">Operator Precedence</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#type-conversion">Type Conversion in Expression Evaluation</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#comparison-operators">Comparison Functions and Operators</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#logical-operators">Logical Operators</a></span></dt></dl></dd><dt><span class="section"><a href="mysqlqb_functions.html#control-flow-functions">Control Flow Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#string-functions">String Functions</a></span></dt><dd><dl><dt><span class="section"><a href="mysqlqb_functions.html#string-comparison-functions">String Comparison Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#regexp">Regular Expressions</a></span></dt></dl></dd><dt><span class="section"><a href="mysqlqb_functions.html#numeric-functions">Numeric Functions</a></span></dt><dd><dl><dt><span class="section"><a href="mysqlqb_functions.html#arithmetic-functions">Arithmetic Operators</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#mathematical-functions">Mathematical Functions</a></span></dt></dl></dd><dt><span class="section"><a href="mysqlqb_functions.html#date-and-time-functions">Date and Time Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#mysql-calendar">What Calendar Is Used By MySQL?</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-search">Full-Text Search Functions</a></span></dt><dd><dl><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-boolean">Boolean Full-Text Searches</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-query-expansion">Full-Text Searches with Query Expansion</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-stopwords">Full-Text Stopwords</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-restrictions">Full-Text Restrictions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-fine-tuning">Fine-Tuning MySQL Full-Text Search</a></span></dt></dl></dd><dt><span class="section"><a href="mysqlqb_functions.html#cast-functions">Cast Functions and Operators</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#other-functions">Other Functions</a></span></dt><dd><dl><dt><span class="section"><a href="mysqlqb_functions.html#bit-functions">Bit Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#encryption-functions">Encryption and Compression Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#information-functions">Information Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#miscellaneous-functions">Miscellaneous Functions</a></span></dt></dl></dd><dt><span class="section"><a href="mysqlqb_functions.html#group-by-functions-and-modifiers">Functions and Modifiers for Use with <code class="literal">GROUP BY</code> Clauses</a></span></dt><dd><dl><dt><span class="section"><a href="mysqlqb_functions.html#group-by-functions"><code class="literal">GROUP BY</code> (Aggregate) Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#group-by-modifiers"><code class="literal">GROUP BY</code> Modifiers</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#group-by-hidden-fields"><code class="literal">GROUP BY</code> and <code class="literal">HAVING</code> with Hidden
        Fields</a></span></dt></dl></dd></dl></div><a id="id2811215" class="indexterm"></a><a id="id2811227" class="indexterm"></a><a id="id2811234" class="indexterm"></a><p>
    Expressions can be used at several points in SQL statements, such as
    in the <code class="literal">ORDER BY</code> or <code class="literal">HAVING</code>
    clauses of <code class="literal">SELECT</code> statements, in the
    <code class="literal">WHERE</code> clause of a <code class="literal">SELECT</code>,
    <code class="literal">DELETE</code>, or <code class="literal">UPDATE</code> statement,
    or in <code class="literal">SET</code> statements. Expressions can be written
    using literal values, column values, <code class="literal">NULL</code>,
    built-in functions, stored functions, user-defined functions, and
    operators. This chapter describes the functions and operators that
    are allowed for writing expressions in MySQL. Instructions for
    writing stored functions and user-defined functions are given in
    <a href="http://dev.mysql.com/doc/refman/5.0/en/stored-procedures.html" target="_top">Stored Procedures and Functions</a>, and
    <a href="http://dev.mysql.com/doc/refman/5.0/en/adding-functions.html" target="_top">Adding New Functions to MySQL</a>. See
    <a href="http://dev.mysql.com/doc/refman/5.0/en/function-resolution.html" target="_top">Function Name Parsing and Resolution</a>, for the rules describing how
    the server interprets references to different kinds of functions.
  </p><p>
    An expression that contains <code class="literal">NULL</code> always produces
    a <code class="literal">NULL</code> value unless otherwise indicated in the
    documentation for a particular function or operator.
  </p><p>
    <span class="bold"><strong>Note</strong></span>: By default, there must be no
    whitespace between a function name and the parenthesis following it.
    This helps the MySQL parser distinguish between function calls and
    references to tables or columns that happen to have the same name as
    a function. However, spaces around function arguments are permitted.
  </p><p>
    You can tell the MySQL server to accept spaces after function names
    by starting it with the <code class="option">--sql-mode=IGNORE_SPACE</code>
    option. (See <a href="http://dev.mysql.com/doc/refman/5.0/en/server-sql-mode.html" target="_top">SQL Modes</a>.) Individual client
    programs can request this behavior by using the
    <code class="literal">CLIENT_IGNORE_SPACE</code> option for
    <code class="literal">mysql_real_connect()</code>. In either case, all
    function names become reserved words.
  </p><p>
    For the sake of brevity, most examples in this chapter display the
    output from the <span><strong class="command">mysql</strong></span> program in abbreviated
    form. Rather than showing examples in this format:
  </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MOD(29,9);</code></strong>
+-----------+
| mod(29,9) |
+-----------+
|         2 |
+-----------+
1 rows in set (0.00 sec)
</pre><p>
    This format is used instead:
  </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MOD(29,9);</code></strong>
        -&gt; 2
</pre><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="func-op-summary-ref"></a>Operator and Function Reference</h2></div></div></div><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p>
        This table is part of an ongoing process to expand and simplify
        the information provided on these elements. Further improvements
        to the table, and corresponding descriptions will be applied
        over the coming months.
      </p></div><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_abs"><code class="literal">ABS()</code></a></td><td>Return the absolute value</td></tr><tr><td><a href="mysqlqb_functions.html#function_acos"><code class="literal">ACOS()</code></a></td><td>Return the arc cosine</td></tr><tr><td><a href="mysqlqb_functions.html#function_adddate"><code class="literal">ADDDATE()</code></a></td><td>Add dates</td></tr><tr><td><a href="mysqlqb_functions.html#function_addtime"><code class="literal">ADDTIME()</code></a></td><td>Add time</td></tr><tr><td><a href="mysqlqb_functions.html#function_aes-encrypt"><code class="literal">AES_DECRYPT()</code></a></td><td>Decrypt using AES</td></tr><tr><td><a href="mysqlqb_functions.html#function_aes-encrypt"><code class="literal">AES_ENCRYPT()</code></a></td><td>Encrypt using AES</td></tr><tr><td><a href="mysqlqb_functions.html#operator_and"><code class="literal">AND</code>, <code class="literal">&amp;&amp;</code></a></td><td>Logical AND</td></tr><tr><td><a href="mysqlqb_functions.html#function_ascii"><code class="literal">ASCII()</code></a></td><td>Return numeric value of left-most character</td></tr><tr><td><a href="mysqlqb_functions.html#function_asin"><code class="literal">ASIN()</code></a></td><td>Return the arc sine</td></tr><tr><td><a href="mysqlqb_functions.html#function_atan2"><code class="literal">ATAN2()</code>, <code class="literal">ATAN()</code></a></td><td>Return the arc tangent of the two arguments</td></tr><tr><td><a href="mysqlqb_functions.html#function_atan"><code class="literal">ATAN()</code></a></td><td>Return the arc tangent</td></tr><tr><td><a href="mysqlqb_functions.html#function_avg"><code class="literal">AVG()</code></a></td><td>Return the average value of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_benchmark"><code class="literal">BENCHMARK()</code></a></td><td>Repeatedly execute an expression</td></tr><tr><td><a href="mysqlqb_functions.html#function_between"><code class="literal">BETWEEN ... AND ... </code></a></td><td>Check whether a value is within a range of values</td></tr><tr><td><a href="mysqlqb_functions.html#function_bin"><code class="literal">BIN()</code></a></td><td>Return a string representation of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#operator_binary"><code class="literal">BINARY</code></a></td><td>Cast a string to a binary string</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-and"><code class="literal">BIT_AND()</code></a></td><td>Return bitwise and</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-count"><code class="literal">BIT_COUNT()</code></a></td><td>Return the number of bits that are set</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-length"><code class="literal">BIT_LENGTH()</code></a></td><td>Return length of argument in bits</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-or"><code class="literal">BIT_OR()</code></a></td><td>Return bitwise or</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-xor"><code class="literal">BIT_XOR()</code></a></td><td>Return bitwise xor</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-and"><code class="literal">&amp;</code></a></td><td>Bitwise AND</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-or"><code class="literal">|</code></a></td><td>Bitwise OR</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-xor"><code class="literal">^</code></a></td><td>Bitwise XOR</td></tr><tr><td><a href="mysqlqb_functions.html#operator_by"><code class="literal">/</code></a></td><td>Division operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_case"><code class="literal">CASE</code></a></td><td>Case statement</td></tr><tr><td><a href="mysqlqb_functions.html#function_cast"><code class="literal">CAST()</code></a></td><td>Cast a value as a certain type</td></tr><tr><td><a href="mysqlqb_functions.html#function_ceiling"><code class="literal">CEILING()</code>, <code class="literal">CEIL()</code></a></td><td>Return the smallest integer value not less than  the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_char-length"><code class="literal">CHAR_LENGTH()</code></a></td><td>Return number of characters in argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_char"><code class="literal">CHAR()</code></a></td><td>Return the character for each integer passed</td></tr><tr><td><a href="mysqlqb_functions.html#function_character-length"><code class="literal">CHARACTER_LENGTH()</code></a></td><td>A synonym for CHAR_LENGTH()</td></tr><tr><td><a href="mysqlqb_functions.html#function_charset"><code class="literal">CHARSET()</code></a></td><td>Return the character set of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_coalesce"><code class="literal">COALESCE()</code></a></td><td>Return the first non-NULL argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_coercibility"><code class="literal">COERCIBILITY()</code></a></td><td>Return the collation coercibility value of the string  argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_collation"><code class="literal">COLLATION()</code></a></td><td>Return the collation of the string argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_compress"><code class="literal">COMPRESS()</code></a></td><td>Return result as a binary string</td></tr><tr><td><a href="mysqlqb_functions.html#function_concat-ws"><code class="literal">CONCAT_WS()</code></a></td><td>Return concatenate with  separator</td></tr><tr><td><a href="mysqlqb_functions.html#function_concat"><code class="literal">CONCAT()</code></a></td><td>Return concatenated string</td></tr><tr><td><a href="mysqlqb_functions.html#function_connection-id"><code class="literal">CONNECTION_ID()</code></a></td><td>Return the connection ID (thread ID)   for the connection</td></tr><tr><td><a href="mysqlqb_functions.html#function_conv"><code class="literal">CONV()</code></a></td><td>Convert numbers between different number bases</td></tr><tr><td><a href="mysqlqb_functions.html#function_convert-tz"><code class="literal">CONVERT_TZ()</code></a></td><td>Convert from one timezone to another</td></tr><tr><td><a href="mysqlqb_functions.html#function_cos"><code class="literal">COS()</code></a></td><td>Return the cosine</td></tr><tr><td><a href="mysqlqb_functions.html#function_cot"><code class="literal">COT()</code></a></td><td>Return the cotangent</td></tr><tr><td><a href="mysqlqb_functions.html#function_count-distinct"><code class="literal">COUNT(DISTINCT)</code></a></td><td>Return the count of a number of different values</td></tr><tr><td><a href="mysqlqb_functions.html#function_count"><code class="literal">COUNT()</code></a></td><td>Return a count of the number of rows returned</td></tr><tr><td><a href="mysqlqb_functions.html#function_crc32"><code class="literal">CRC32()</code></a></td><td>Compute a cyclic redundancy check value</td></tr><tr><td><a href="mysqlqb_functions.html#function_curdate"><code class="literal">CURDATE()</code></a></td><td>Return the current date</td></tr><tr><td><a href="mysqlqb_functions.html#function_current-date"><code class="literal">CURRENT_DATE()</code>, <code class="literal">CURRENT_DATE</code></a></td><td>Synonyms for CURDATE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_current-time"><code class="literal">CURRENT_TIME()</code>, <code class="literal">CURRENT_TIME</code></a></td><td>Synonyms for CURTIME()</td></tr><tr><td><a href="mysqlqb_functions.html#function_current-timestamp"><code class="literal">CURRENT_TIMESTAMP()</code>, <code class="literal">CURRENT_TIMESTAMP</code></a></td><td>Synonyms for NOW()</td></tr><tr><td><a href="mysqlqb_functions.html#function_current-user"><code class="literal">CURRENT_USER()</code>, <code class="literal">CURRENT_USER</code></a></td><td>Return the username and hostname combination</td></tr><tr><td><a href="mysqlqb_functions.html#function_curtime"><code class="literal">CURTIME()</code></a></td><td>Return the current time</td></tr><tr><td><a href="mysqlqb_functions.html#function_database"><code class="literal">DATABASE()</code></a></td><td>Return the default (current) database name</td></tr><tr><td><a href="mysqlqb_functions.html#function_date-add"><code class="literal">DATE_ADD()</code></a></td><td>Add two dates</td></tr><tr><td><a href="mysqlqb_functions.html#function_date-format"><code class="literal">DATE_FORMAT()</code></a></td><td>Format date as specified</td></tr><tr><td><a href="mysqlqb_functions.html#function_date-sub"><code class="literal">DATE_SUB()</code></a></td><td>Subtract two dates</td></tr><tr><td><a href="mysqlqb_functions.html#function_date"><code class="literal">DATE()</code></a></td><td>Extract the date part of a date or datetime expression</td></tr><tr><td><a href="mysqlqb_functions.html#function_datediff"><code class="literal">DATEDIFF()</code></a></td><td>Subtract two dates</td></tr><tr><td><a href="mysqlqb_functions.html#function_day"><code class="literal">DAY()</code></a></td><td>Synonym for DAYOFMONTH()</td></tr><tr><td><a href="mysqlqb_functions.html#function_dayname"><code class="literal">DAYNAME()</code></a></td><td>Return the name of the weekday</td></tr><tr><td><a href="mysqlqb_functions.html#function_dayofmonth"><code class="literal">DAYOFMONTH()</code></a></td><td>Return the day of the month (1-31)</td></tr><tr><td><a href="mysqlqb_functions.html#function_dayofweek"><code class="literal">DAYOFWEEK()</code></a></td><td>Return the weekday index of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_dayofyear"><code class="literal">DAYOFYEAR()</code></a></td><td>Return the day of the year (1-366)</td></tr><tr><td><a href="mysqlqb_functions.html#function_decode"><code class="literal">DECODE()</code></a></td><td>Decodes a string encrypted using ENCODE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_default"><code class="literal">DEFAULT()</code></a></td><td>Return the default value for a table column</td></tr><tr><td><a href="mysqlqb_functions.html#function_degrees"><code class="literal">DEGREES()</code></a></td><td>Convert radians to degrees</td></tr><tr><td><a href="mysqlqb_functions.html#function_des-decrypt"><code class="literal">DES_DECRYPT()</code></a></td><td>Decrypt a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_des-encrypt"><code class="literal">DES_ENCRYPT()</code></a></td><td>Decrypt a string</td></tr><tr><td><a href="mysqlqb_functions.html#operator_div"><code class="literal">DIV</code></a></td><td>Integer division</td></tr><tr><td><a href="mysqlqb_functions.html#function_elt"><code class="literal">ELT()</code></a></td><td>Return string at index number</td></tr><tr><td><a href="mysqlqb_functions.html#function_encode"><code class="literal">ENCODE()</code></a></td><td>Encode a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_encrypt"><code class="literal">ENCRYPT()</code></a></td><td>Encrypt a string</td></tr><tr><td><a href="mysqlqb_functions.html#operator_equal-to"><code class="literal">&lt;=&gt;</code></a></td><td>NULL-safe equal to operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_equal"><code class="literal">=</code></a></td><td>Equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_exp"><code class="literal">EXP()</code></a></td><td>Raise to the power of</td></tr><tr><td><a href="mysqlqb_functions.html#function_export-set"><code class="literal">EXPORT_SET()</code></a></td><td>Return a string such that for every bit set in the value  bits, you get an on string and for every unset bit, you get an off string</td></tr><tr><td><a href="mysqlqb_functions.html#function_extract"><code class="literal">EXTRACT</code></a></td><td>Extract part of a date</td></tr><tr><td><a href="mysqlqb_functions.html#function_field"><code class="literal">FIELD()</code></a></td><td>Return the index (position) of the first argument  in the subsequent arguments</td></tr><tr><td><a href="mysqlqb_functions.html#function_find-in-set"><code class="literal">FIND_IN_SET()</code></a></td><td>Return the number of times the first argument is found   in the second arguments</td></tr><tr><td><a href="mysqlqb_functions.html#function_floor"><code class="literal">FLOOR()</code></a></td><td>Return the largest integer value not greater than  the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_format"><code class="literal">FORMAT()</code></a></td><td>Return a number formatted to specified number of decimal places</td></tr><tr><td><a href="mysqlqb_functions.html#function_found-rows"><code class="literal">FOUND_ROWS()</code></a></td><td>For a SELECT with a LIMIT clause, the  number of rows that would be returned were there no LIMIT clause</td></tr><tr><td><a href="mysqlqb_functions.html#function_from-days"><code class="literal">FROM_DAYS()</code></a></td><td>Convert a day number to a date</td></tr><tr><td><a href="mysqlqb_functions.html#function_from-unixtime"><code class="literal">FROM_UNIXTIME()</code></a></td><td>Format date as a UNIX timestamp</td></tr><tr><td><a href="mysqlqb_functions.html#function_get-format"><code class="literal">GET_FORMAT()</code></a></td><td>Return a date format string</td></tr><tr><td><a href="mysqlqb_functions.html#function_get-lock"><code class="literal">GET_LOCK()</code></a></td><td>Get a named lock</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than-or-equal"><code class="literal">&gt;=</code></a></td><td>Greater than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than"><code class="literal">&gt;</code></a></td><td>Greater than operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_greatest"><code class="literal">GREATEST()</code></a></td><td>Return the largest argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_group-concat"><code class="literal">GROUP_CONCAT()</code></a></td><td>Return a concatenated string</td></tr><tr><td><a href="mysqlqb_functions.html#function_hex"><code class="literal">HEX()</code></a></td><td>Return a string representation of a hex value</td></tr><tr><td><a href="mysqlqb_functions.html#function_hour"><code class="literal">HOUR()</code></a></td><td>Extract the hour</td></tr><tr><td><a href="mysqlqb_functions.html#function_if"><code class="literal">IF()</code></a></td><td>If/else construct</td></tr><tr><td><a href="mysqlqb_functions.html#function_ifnull"><code class="literal">IFNULL()</code></a></td><td>Null if/else construct</td></tr><tr><td><a href="mysqlqb_functions.html#function_in"><code class="literal">IN</code></a></td><td>Check whether a value is not within a set of values</td></tr><tr><td><a href="mysqlqb_functions.html#function_inet-aton"><code class="literal">INET_ATON()</code></a></td><td>Return the numeric value of an IP address</td></tr><tr><td><a href="mysqlqb_functions.html#function_inet-ntoa"><code class="literal">INET_NTOA()</code></a></td><td>Return the IP address from a numeric value</td></tr><tr><td><a href="mysqlqb_functions.html#function_insert"><code class="literal">INSERT()</code></a></td><td>Insert a substring at the specified position up to  the specified number of characters</td></tr><tr><td><a href="mysqlqb_functions.html#function_instr"><code class="literal">INSTR()</code></a></td><td>Return the index of the first occurrence of substring</td></tr><tr><td><a href="mysqlqb_functions.html#function_interval"><code class="literal">INTERVAL()</code></a></td><td>Return the index of the argument   that is less than the first argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-free-lock"><code class="literal">IS_FREE_LOCK()</code></a></td><td>Checks whether the named lock  is free</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-null"><code class="literal">IS NULL</code></a></td><td>NULL value test</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-used-lock"><code class="literal">IS_USED_LOCK()</code></a></td><td>Checks whether the named lock is in use.  Return connection identifier if true.</td></tr><tr><td><a href="mysqlqb_functions.html#operator_is"><code class="literal">IS</code></a></td><td>Test a value against a boolean</td></tr><tr><td><a href="mysqlqb_functions.html#function_isnull"><code class="literal">ISNULL()</code></a></td><td>Test whether the argument is NULL</td></tr><tr><td><a href="mysqlqb_functions.html#function_last-day"><code class="literal">LAST_DAY</code></a></td><td>Return the last day of the month for the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_last-insert-id"><code class="literal">LAST_INSERT_ID()</code></a></td><td>Value of the AUTOINCREMENT column   for the last INSERT</td></tr><tr><td><a href="mysqlqb_functions.html#function_lcase"><code class="literal">LCASE()</code></a></td><td>Synonym for LOWER() </td></tr><tr><td><a href="mysqlqb_functions.html#function_least"><code class="literal">LEAST()</code></a></td><td>Return the smallest argument</td></tr><tr><td><a href="mysqlqb_functions.html#operator_left-shift"><code class="literal">&lt;&lt;</code></a></td><td>Left shift</td></tr><tr><td><a href="mysqlqb_functions.html#function_left"><code class="literal">LEFT()</code></a></td><td>Return the leftmost number of characters as specified</td></tr><tr><td><a href="mysqlqb_functions.html#function_length"><code class="literal">LENGTH()</code></a></td><td>Return the length of a string in bytes</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than-or-equal"><code class="literal">&lt;=</code></a></td><td>Less than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than"><code class="literal">&lt;</code></a></td><td>Less than operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_like"><code class="literal">LIKE</code></a></td><td>Simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#function_ln"><code class="literal">LN()</code></a></td><td>Return the natural logarithm of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_load-file"><code class="literal">LOAD_FILE()</code></a></td><td>Load the named file</td></tr><tr><td><a href="mysqlqb_functions.html#function_localtime"><code class="literal">LOCALTIME()</code>, <code class="literal">LOCALTIME</code></a></td><td>Synonym for NOW()</td></tr><tr><td><a href="mysqlqb_functions.html#function_localtimestamp"><code class="literal">LOCALTIMESTAMP</code>, <code class="literal">LOCALTIMESTAMP()</code></a></td><td>Synonym for NOW()</td></tr><tr><td><a href="mysqlqb_functions.html#function_locate"><code class="literal">LOCATE()</code></a></td><td>Return the position of the first occurrence  of substring</td></tr><tr><td><a href="mysqlqb_functions.html#function_log10"><code class="literal">LOG10()</code></a></td><td>Return the base-10 logarithm of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_log2"><code class="literal">LOG2()</code></a></td><td>Return the base-2 logarithm of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_log"><code class="literal">LOG()</code></a></td><td>  Return the natural logarithm of the first argument  </td></tr><tr><td><a href="mysqlqb_functions.html#function_lower"><code class="literal">LOWER()</code></a></td><td>Return the argument in lowercase </td></tr><tr><td><a href="mysqlqb_functions.html#function_lpad"><code class="literal">LPAD()</code></a></td><td>Return the string argument, left-padded  with the specified string</td></tr><tr><td><a href="mysqlqb_functions.html#function_ltrim"><code class="literal">LTRIM()</code></a></td><td>Remove leading spaces</td></tr><tr><td><a href="mysqlqb_functions.html#function_make-set"><code class="literal">MAKE_SET()</code></a></td><td>Return a set of comma-separated strings  that have the corresponding bit in bits set</td></tr><tr><td><a href="mysqlqb_functions.html#function_makedate"><code class="literal">MAKEDATE()</code></a></td><td>Create a date from the year and day of year</td></tr><tr><td><a href="mysqlqb_functions.html#function_maketime"><code class="literal">MAKETIME</code></a></td><td>MAKETIME()</td></tr><tr><td><a href="mysqlqb_functions.html#function_master-pos-wait"><code class="literal">MASTER_POS_WAIT()</code></a></td><td>Block until the slave has read and  applied all updates up to the specified position</td></tr><tr><td><a href="mysqlqb_functions.html#function_min"><code class="literal">MAX()</code></a></td><td>Return the maximum value</td></tr><tr><td><a href="mysqlqb_functions.html#function_md5"><code class="literal">MD5()</code></a></td><td>Calculate MD5 checksum</td></tr><tr><td><a href="mysqlqb_functions.html#function_microsecond"><code class="literal">MICROSECOND()</code></a></td><td>Return the microseconds from argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_mid"><code class="literal">MID()</code></a></td><td>Return a substring starting from the specified position</td></tr><tr><td><a href="mysqlqb_functions.html#function_min"><code class="literal">MIN()</code></a></td><td>Return the minimum value</td></tr><tr><td><a href="mysqlqb_functions.html#operator_minus"><code class="literal">-</code></a></td><td>Minus operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_minute"><code class="literal">MINUTE()</code></a></td><td>Return the minute from the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_mod"><code class="literal">MOD()</code></a></td><td>Return the remainder</td></tr><tr><td><a href="mysqlqb_functions.html#operator_mod"><code class="literal">%</code></a></td><td>Modulo operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_month"><code class="literal">MONTH()</code></a></td><td>Return the month from the date passed</td></tr><tr><td><a href="mysqlqb_functions.html#function_monthname"><code class="literal">MONTHNAME()</code></a></td><td>Return the name of the month</td></tr><tr><td><a href="mysqlqb_functions.html#function_name-const"><code class="literal">NAME_CONST()</code></a>(v5.0.12)</td><td>Causes the column to have the given name</td></tr><tr><td><a href="mysqlqb_functions.html#function_not-between"><code class="literal">NOT BETWEEN ... AND ...</code></a></td><td>Check whether a value is not within a range of values</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-equal"><code class="literal">!=</code>, <code class="literal">&lt;&gt;</code></a></td><td>Not equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_not-in"><code class="literal">NOT IN</code></a></td><td>Check whether a value is not within a set of values</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-like"><code class="literal">NOT LIKE</code></a></td><td>Negation of simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-regexp"><code class="literal">NOT RGEXP</code></a></td><td>Negation of REGEXP</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not"><code class="literal">NOT</code>, <code class="literal">!</code></a></td><td>Negates value</td></tr><tr><td><a href="mysqlqb_functions.html#function_now"><code class="literal">NOW()</code></a></td><td>Return the current date and time</td></tr><tr><td><a href="mysqlqb_functions.html#function_nullif"><code class="literal">NULLIF()</code></a></td><td>Return NULL if expr1 = expr2</td></tr><tr><td><a href="mysqlqb_functions.html#function_oct"><code class="literal">OCT()</code></a></td><td>Return a string representation of the octal argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_octet-length"><code class="literal">OCTET_LENGTH()</code></a></td><td>A synonym for LENGTH()</td></tr><tr><td><a href="mysqlqb_functions.html#function_old-password"><code class="literal">OLD_PASSWORD()</code></a></td><td>Return the value of the  old (pre-4.1) implementation of PASSWORD</td></tr><tr><td><a href="mysqlqb_functions.html#operator_or"><code class="literal">||</code>, <code class="literal">OR</code></a></td><td>Logical OR</td></tr><tr><td><a href="mysqlqb_functions.html#function_ord"><code class="literal">ORD()</code></a></td><td>If the leftmost character of the  argument is a multi-byte character,  returns the code for that character</td></tr><tr><td><a href="mysqlqb_functions.html#function_password"><code class="literal">PASSWORD()</code></a></td><td>Calculate and return a password string</td></tr><tr><td><a href="mysqlqb_functions.html#function_period-add"><code class="literal">PERIOD_ADD()</code></a></td><td>Add a period to a year-month</td></tr><tr><td><a href="mysqlqb_functions.html#function_period-diff"><code class="literal">PERIOD_DIFF()</code></a></td><td>Return the number of months between periods</td></tr><tr><td><a href="mysqlqb_functions.html#function_pi"><code class="literal">PI()</code></a></td><td>Return the value of pi</td></tr><tr><td><a href="mysqlqb_functions.html#operator_plus"><code class="literal">+</code></a></td><td>Addition operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_position"><code class="literal">POSITION()</code></a></td><td>A synonym for LOCATE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_pow"><code class="literal">POW()</code>, <code class="literal">POWER()</code></a></td><td>Return the argument raised to the specified power</td></tr><tr><td><a href="mysqlqb_functions.html#function_quarter"><code class="literal">QUARTER()</code></a></td><td>Return the quarter from a date argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_quote"><code class="literal">QUOTE()</code></a></td><td>Escape the argument for use in an SQL statement</td></tr><tr><td><a href="mysqlqb_functions.html#function_radians"><code class="literal">RADIANS()</code></a></td><td>Return argument converted to radians</td></tr><tr><td><a href="mysqlqb_functions.html#function_rand"><code class="literal">RAND()</code></a></td><td>Return a random floating-point value</td></tr><tr><td><a href="mysqlqb_functions.html#operator_regexp"><code class="literal">REGEXP</code></a></td><td>Pattern matching using regular expressions</td></tr><tr><td><a href="mysqlqb_functions.html#function_release-lock"><code class="literal">RELEASE_LOCK()</code></a></td><td>Releases the named lock</td></tr><tr><td><a href="mysqlqb_functions.html#function_repeat"><code class="literal">REPEAT()</code></a></td><td>Repeat a string the specified number of times</td></tr><tr><td><a href="mysqlqb_functions.html#function_replace"><code class="literal">REPLACE()</code></a></td><td>Replace occurrences of a specified string</td></tr><tr><td><a href="mysqlqb_functions.html#function_reverse"><code class="literal">REVERSE()</code></a></td><td>Reverse the characters in a string</td></tr><tr><td><a href="mysqlqb_functions.html#operator_right-shift"><code class="literal">&gt;&gt;</code></a></td><td>Right shift</td></tr><tr><td><a href="mysqlqb_functions.html#function_right"><code class="literal">RIGHT()</code></a></td><td>Return the specified rightmost number of characters</td></tr><tr><td><a href="mysqlqb_functions.html#operator_regexp"><code class="literal">RLIKE</code></a></td><td>Synonym for REGEXP</td></tr><tr><td><a href="mysqlqb_functions.html#function_round"><code class="literal">ROUND()</code></a></td><td>Round the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_row-count"><code class="literal">ROW_COUNT()</code></a></td><td>The number of rows  updated</td></tr><tr><td><a href="mysqlqb_functions.html#function_rpad"><code class="literal">RPAD()</code></a></td><td>Append string the specified number of times</td></tr><tr><td><a href="mysqlqb_functions.html#function_rtrim"><code class="literal">RTRIM()</code></a></td><td>Remove trailing spaces</td></tr><tr><td><a href="mysqlqb_functions.html#function_schema"><code class="literal">SCHEMA()</code></a></td><td>A synonym for DATABASE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_sec-to-time"><code class="literal">SEC_TO_TIME()</code></a></td><td>Converts seconds to 'HH:MM:SS' format</td></tr><tr><td><a href="mysqlqb_functions.html#function_second"><code class="literal">SECOND()</code></a></td><td>Return the second (0-59)</td></tr><tr><td><a href="mysqlqb_functions.html#function_session-user"><code class="literal">SESSION_USER()</code></a></td><td>Synonym for USER()</td></tr><tr><td><a href="mysqlqb_functions.html#function_sha1"><code class="literal">SHA1()</code>, <code class="literal">SHA()</code></a></td><td>Calculate an SHA-1 160-bit checksum</td></tr><tr><td><a href="mysqlqb_functions.html#function_sign"><code class="literal">SIGN()</code></a></td><td>Return the sign of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_sin"><code class="literal">SIN()</code></a></td><td>Return the sine of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_sleep"><code class="literal">SLEEP()</code></a></td><td>Sleep for a number of seconds</td></tr><tr><td><a href="mysqlqb_functions.html#function_soundex"><code class="literal">SOUNDEX()</code></a></td><td>Return a soundex string</td></tr><tr><td><a href="mysqlqb_functions.html#operator_sounds-like"><code class="literal">SOUNDS LIKE</code></a></td><td>Compare sounds</td></tr><tr><td><a href="mysqlqb_functions.html#function_space"><code class="literal">SPACE()</code></a></td><td>Return a string of the specified number of spaces</td></tr><tr><td><a href="mysqlqb_functions.html#function_sqrt"><code class="literal">SQRT()</code></a></td><td>Return the square root of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_std"><code class="literal">STD()</code>, <code class="literal">STDDEV()</code></a></td><td>Return the population standard deviation</td></tr><tr><td><a href="mysqlqb_functions.html#function_stddev-pop"><code class="literal">STDDEV_POP()</code></a></td><td>Return the population standard deviation</td></tr><tr><td><a href="mysqlqb_functions.html#function_stddev-samp"><code class="literal">STDDEV_SAMP()</code></a></td><td>Return the sample standard deviation</td></tr><tr><td><a href="mysqlqb_functions.html#function_str-to-date"><code class="literal">STR_TO_DATE()</code></a></td><td>Convert a string to a date</td></tr><tr><td><a href="mysqlqb_functions.html#function_strcmp"><code class="literal">STRCMP()</code></a></td><td>Compare two strings</td></tr><tr><td><a href="mysqlqb_functions.html#function_subdate"><code class="literal">SUBDATE()</code></a></td><td>When invoked with three arguments a synonym for DATE_SUB()</td></tr><tr><td><a href="mysqlqb_functions.html#function_substring-index"><code class="literal">SUBSTRING_INDEX()</code></a></td><td>Return a substring from a  string before the specified number of occurrences of the delimiter</td></tr><tr><td><a href="mysqlqb_functions.html#function_substring"><code class="literal">SUBSTRING()</code></a></td><td>Return the substring as specified</td></tr><tr><td><a href="mysqlqb_functions.html#function_subtime"><code class="literal">SUBTIME()</code></a></td><td>Subtract times</td></tr><tr><td><a href="mysqlqb_functions.html#function_sum"><code class="literal">SUM()</code></a></td><td>Return the sum</td></tr><tr><td><a href="mysqlqb_functions.html#function_sysdate"><code class="literal">SYSDATE()</code></a></td><td>Return the time at which the function executes</td></tr><tr><td><a href="mysqlqb_functions.html#function_system-user"><code class="literal">SYSTEM_USER()</code></a></td><td>Synonym for USER()</td></tr><tr><td><a href="mysqlqb_functions.html#function_tan"><code class="literal">TAN()</code></a></td><td>Return the tangent of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#operator_tilde"><code class="literal">~</code></a></td><td>Invert bits</td></tr><tr><td><a href="mysqlqb_functions.html#function_time-format"><code class="literal">TIME_FORMAT()</code></a></td><td>Format as time</td></tr><tr><td><a href="mysqlqb_functions.html#function_time-to-sec"><code class="literal">TIME_TO_SEC()</code></a></td><td>Return the argument converted to seconds</td></tr><tr><td><a href="mysqlqb_functions.html#function_time"><code class="literal">TIME()</code></a></td><td>Extract the time portion of the expression passed</td></tr><tr><td><a href="mysqlqb_functions.html#function_timediff"><code class="literal">TIMEDIFF()</code></a></td><td>Subtract time</td></tr><tr><td><a href="mysqlqb_functions.html#operator_times"><code class="literal">*</code></a></td><td>Times operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_timestamp"><code class="literal">TIMESTAMP()</code></a></td><td>With a single argument, this function returns the date or  datetime expression. With two arguments, the sum of the arguments</td></tr><tr><td><a href="mysqlqb_functions.html#function_timestampadd"><code class="literal">TIMESTAMPADD()</code></a></td><td>Add an interval to a datetime expression</td></tr><tr><td><a href="mysqlqb_functions.html#function_timestampdiff"><code class="literal">TIMESTAMPDIFF()</code></a></td><td>Subtract an interval from a datetime expression</td></tr><tr><td><a href="mysqlqb_functions.html#function_to-days"><code class="literal">TO_DAYS()</code></a></td><td>Return the date argument converted to days</td></tr><tr><td><a href="mysqlqb_functions.html#function_trim"><code class="literal">TRIM()</code></a></td><td>Return </td></tr><tr><td><a href="mysqlqb_functions.html#function_truncate"><code class="literal">TRUNCATE()</code></a></td><td>Truncate to specified number of decimal places</td></tr><tr><td><a href="mysqlqb_functions.html#function_ucase"><code class="literal">UCASE()</code></a></td><td>Synonym for UPPER()</td></tr><tr><td><a href="mysqlqb_functions.html#operator_unary-minus"><code class="literal">-</code></a></td><td>Change the sign of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_uncompress"><code class="literal">UNCOMPRESS()</code></a></td><td>Uncompress a string compressed</td></tr><tr><td><a href="mysqlqb_functions.html#function_uncompressed-length"><code class="literal">UNCOMPRESSED_LENGTH()</code></a></td><td>Return the length of a string before compression</td></tr><tr><td><a href="mysqlqb_functions.html#function_unhex"><code class="literal">UNHEX()</code></a></td><td>Convert each pair of hexadecimal digits  to a character</td></tr><tr><td><a href="mysqlqb_functions.html#function_unix-timestamp"><code class="literal">UNIX_TIMESTAMP()</code></a></td><td>Return a UNIX timestamp</td></tr><tr><td><a href="mysqlqb_functions.html#function_upper"><code class="literal">UPPER()</code></a></td><td>Convert to uppercase</td></tr><tr><td><a href="mysqlqb_functions.html#function_user"><code class="literal">USER()</code></a></td><td>Return the current username and hostname</td></tr><tr><td><a href="mysqlqb_functions.html#function_utc-date"><code class="literal">UTC_DATE()</code></a></td><td>Return the current UTC date</td></tr><tr><td><a href="mysqlqb_functions.html#function_utc-time"><code class="literal">UTC_TIME()</code></a></td><td>Return the current UTC time</td></tr><tr><td><a href="mysqlqb_functions.html#function_utc-timestamp"><code class="literal">UTC_TIMESTAMP()</code></a></td><td>Return the current UTC date and time</td></tr><tr><td><a href="mysqlqb_functions.html#function_uuid"><code class="literal">UUID()</code></a></td><td>Return a Universal Unique Identifier (UUID)</td></tr><tr><td><a href="mysqlqb_functions.html#function_values"><code class="literal">VALUES()</code></a></td><td>Defines the values to be used during an INSERT</td></tr><tr><td><a href="mysqlqb_functions.html#function_var-pop"><code class="literal">VAR_POP()</code></a></td><td>Return the population standard variance</td></tr><tr><td><a href="mysqlqb_functions.html#function_var-samp"><code class="literal">VAR_SAMP()</code></a></td><td>Return the sample variance</td></tr><tr><td><a href="mysqlqb_functions.html#function_variance"><code class="literal">VARIANCE()</code></a></td><td>Return the population standard variance</td></tr><tr><td><a href="mysqlqb_functions.html#function_week"><code class="literal">WEEK()</code></a></td><td>Return the week number</td></tr><tr><td><a href="mysqlqb_functions.html#function_weekday"><code class="literal">WEEKDAY()</code></a></td><td>Return the weekday index</td></tr><tr><td><a href="mysqlqb_functions.html#function_weekofyear"><code class="literal">WEEKOFYEAR()</code></a></td><td>Return the calendar week of the date (1-53)</td></tr><tr><td><a href="mysqlqb_functions.html#operator_xor"><code class="literal">XOR</code></a></td><td>Logical XOR</td></tr><tr><td><a href="mysqlqb_functions.html#function_year"><code class="literal">YEAR()</code></a></td><td>Return the year</td></tr><tr><td><a href="mysqlqb_functions.html#function_yearweek"><code class="literal">YEARWEEK()</code></a></td><td>Return the year and week</td></tr></tbody></table></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="non-typed-operators"></a>Operators</h2></div></div></div><div class="toc"><dl><dt><span class="section"><a href="mysqlqb_functions.html#operator-precedence">Operator Precedence</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#type-conversion">Type Conversion in Expression Evaluation</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#comparison-operators">Comparison Functions and Operators</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#logical-operators">Logical Operators</a></span></dt></dl></div><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#operator_and"><code class="literal">AND</code>, <code class="literal">&amp;&amp;</code></a></td><td>Logical AND</td></tr><tr><td><a href="mysqlqb_functions.html#operator_binary"><code class="literal">BINARY</code></a></td><td>Cast a string to a binary string</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-and"><code class="literal">&amp;</code></a></td><td>Bitwise AND</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-or"><code class="literal">|</code></a></td><td>Bitwise OR</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-xor"><code class="literal">^</code></a></td><td>Bitwise XOR</td></tr><tr><td><a href="mysqlqb_functions.html#operator_by"><code class="literal">/</code></a></td><td>Division operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_div"><code class="literal">DIV</code></a></td><td>Integer division</td></tr><tr><td><a href="mysqlqb_functions.html#operator_equal-to"><code class="literal">&lt;=&gt;</code></a></td><td>NULL-safe equal to operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_equal"><code class="literal">=</code></a></td><td>Equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than-or-equal"><code class="literal">&gt;=</code></a></td><td>Greater than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than"><code class="literal">&gt;</code></a></td><td>Greater than operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_is"><code class="literal">IS</code></a></td><td>Test a value against a boolean</td></tr><tr><td><a href="mysqlqb_functions.html#operator_left-shift"><code class="literal">&lt;&lt;</code></a></td><td>Left shift</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than-or-equal"><code class="literal">&lt;=</code></a></td><td>Less than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than"><code class="literal">&lt;</code></a></td><td>Less than operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_like"><code class="literal">LIKE</code></a></td><td>Simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#operator_minus"><code class="literal">-</code></a></td><td>Minus operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_mod"><code class="literal">%</code></a></td><td>Modulo operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-equal"><code class="literal">!=</code>, <code class="literal">&lt;&gt;</code></a></td><td>Not equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-like"><code class="literal">NOT LIKE</code></a></td><td>Negation of simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-regexp"><code class="literal">NOT RGEXP</code></a></td><td>Negation of REGEXP</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not"><code class="literal">NOT</code>, <code class="literal">!</code></a></td><td>Negates value</td></tr><tr><td><a href="mysqlqb_functions.html#operator_or"><code class="literal">||</code>, <code class="literal">OR</code></a></td><td>Logical OR</td></tr><tr><td><a href="mysqlqb_functions.html#operator_plus"><code class="literal">+</code></a></td><td>Addition operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_regexp"><code class="literal">REGEXP</code></a></td><td>Pattern matching using regular expressions</td></tr><tr><td><a href="mysqlqb_functions.html#operator_right-shift"><code class="literal">&gt;&gt;</code></a></td><td>Right shift</td></tr><tr><td><a href="mysqlqb_functions.html#operator_regexp"><code class="literal">RLIKE</code></a></td><td>Synonym for REGEXP</td></tr><tr><td><a href="mysqlqb_functions.html#operator_sounds-like"><code class="literal">SOUNDS LIKE</code></a></td><td>Compare sounds</td></tr><tr><td><a href="mysqlqb_functions.html#operator_tilde"><code class="literal">~</code></a></td><td>Invert bits</td></tr><tr><td><a href="mysqlqb_functions.html#operator_times"><code class="literal">*</code></a></td><td>Times operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_unary-minus"><code class="literal">-</code></a></td><td>Change the sign of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#operator_xor"><code class="literal">XOR</code></a></td><td>Logical XOR</td></tr></tbody></table></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="operator-precedence"></a>Operator Precedence</h3></div></div></div><a id="id2814446" class="indexterm"></a><a id="id2810577" class="indexterm"></a><p>
        Operator precedences are shown in the following list, from
        lowest precedence to the highest. Operators that are shown
        together on a line have the same precedence.
      </p><pre class="programlisting">:=
||, OR, XOR
&amp;&amp;, AND
NOT
BETWEEN, CASE, WHEN, THEN, ELSE
=, &lt;=&gt;, &gt;=, &gt;, &lt;=, &lt;, &lt;&gt;, !=, IS, LIKE, REGEXP, IN
|
&amp;
&lt;&lt;, &gt;&gt;
-, +
*, /, DIV, %, MOD
^
- (unary minus), ~ (unary bit inversion)
!
BINARY, COLLATE
</pre><p>
        The precedence shown for <code class="literal">NOT</code> is as of MySQL
        5.0.2. For earlier versions, or from 5.0.2 on if the
        <code class="literal">HIGH_NOT_PRECEDENCE</code> SQL mode is enabled, the
        precedence of <code class="literal">NOT</code> is the same as that of the
        <code class="literal">!</code> operator. See
        <a href="http://dev.mysql.com/doc/refman/5.0/en/server-sql-mode.html" target="_top">SQL Modes</a>.
      </p><a id="id2814750" class="indexterm"></a><a id="id2814757" class="indexterm"></a><a id="id2814763" class="indexterm"></a><a id="id2814773" class="indexterm"></a><p>
        The precedence of operators determines the order of evaluation
        of terms in an expression. To override this order and group
        terms explicitly, use parentheses. For example:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1+2*3;</code></strong>
        -&gt; 7
mysql&gt; <strong class="userinput"><code>SELECT (1+2)*3;</code></strong>
        -&gt; 9
</pre></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="type-conversion"></a>Type Conversion in Expression Evaluation</h3></div></div></div><a id="id2814810" class="indexterm"></a><a id="id2814817" class="indexterm"></a><p>
        When an operator is used with operands of different types, type
        conversion occurs to make the operands compatible. Some
        conversions occur implicitly. For example, MySQL automatically
        converts numbers to strings as necessary, and vice versa.
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1+'1';</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT CONCAT(2,' test');</code></strong>
        -&gt; '2 test'
</pre><p>
        It is also possible to perform explicit conversions. If you want
        to convert a number to a string explicitly, use the
        <code class="literal">CAST()</code> or <code class="literal">CONCAT()</code>
        function (<code class="literal">CAST()</code> is preferable):
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 38.8, CAST(38.8 AS CHAR);</code></strong>
        -&gt; 38.8, '38.8'
mysql&gt; <strong class="userinput"><code>SELECT 38.8, CONCAT(38.8);</code></strong>
        -&gt; 38.8, '38.8'
</pre><p>
        The following rules describe how conversion occurs for
        comparison operations:
      </p><div class="itemizedlist"><ul type="disc"><li><p>
            If one or both arguments are <code class="literal">NULL</code>, the
            result of the comparison is <code class="literal">NULL</code>, except
            for the <code class="literal">NULL</code>-safe
            <code class="literal">&lt;=&gt;</code> equality comparison operator.
            For <code class="literal">NULL &lt;=&gt; NULL</code>, the result is
            true.
          </p></li><li><p>
            If both arguments in a comparison operation are strings,
            they are compared as strings.
          </p></li><li><p>
            If both arguments are integers, they are compared as
            integers.
          </p></li><li><p>
            Hexadecimal values are treated as binary strings if not
            compared to a number.
          </p></li><li><p>
            <a id="id2814937" class="indexterm"></a>

            <a id="id2814944" class="indexterm"></a>

            If one of the arguments is a <code class="literal">TIMESTAMP</code> or
            <code class="literal">DATETIME</code> column and the other argument is
            a constant, the constant is converted to a timestamp before
            the comparison is performed. This is done to be more
            ODBC-friendly. Note that this is not done for the arguments
            to <code class="literal">IN()</code>! To be safe, always use complete
            datetime, date, or time strings when doing comparisons.
          </p></li><li><p>
            In all other cases, the arguments are compared as
            floating-point (real) numbers.
          </p></li></ul></div><p>
        The following examples illustrate conversion of strings to
        numbers for comparison operations:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 &gt; '6x';</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 7 &gt; '6x';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 0 &gt; 'x6';</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 0 = 'x6';</code></strong>
        -&gt; 1
</pre><p>
        Note that when you are comparing a string column with a number,
        MySQL cannot use an index on the column to look up the value
        quickly. If <em class="replaceable"><code>str_col</code></em> is an indexed
        string column, the index cannot be used when performing the
        lookup in the following statement:
      </p><pre class="programlisting">SELECT * FROM <em class="replaceable"><code>tbl_name</code></em> WHERE <em class="replaceable"><code>str_col</code></em>=1;
</pre><p>
        The reason for this is that there are many different strings
        that may convert to the value <code class="literal">1</code>, such as
        <code class="literal">'1'</code>, <code class="literal">' 1'</code>, or
        <code class="literal">'1a'</code>.
      </p><p>
        Comparisons that use floating-point numbers (or values that are
        converted to floating-point numbers) are approximate because
        such numbers are inexact. This might lead to results that appear
        inconsistent:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT '18015376320243458' = 18015376320243458;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT '18015376320243459' = 18015376320243459;</code></strong>
        -&gt; 0
</pre><p>
        Such results can occur because the values are converted to
        floating-point numbers, which have only 53 bits of precision and
        are subject to rounding:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT '18015376320243459'+0.0;</code></strong>
        -&gt; 1.8015376320243e+16
</pre><p>
        Furthermore, the conversion from string to floating-point and
        from integer to floating-point do not necessarily occur the same
        way. The integer may be converted to floating-point by the CPU,
        whereas the string is converted digit by digit in an operation
        that involves floating-point multiplications.
      </p><p>
        The results shown will vary on different systems, and can be
        affected by factors such as computer architecture or the
        compiler version or optimization level. One way to avoid such
        problems is to use <code class="literal">CAST()</code> so that a value
        will not be converted implicitly to a float-point number:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CAST('18015376320243459' AS UNSIGNED) = 18015376320243459;</code></strong>
        -&gt; 1
</pre><p>
        For more information about floating-point comparisons, see
        <a href="http://dev.mysql.com/doc/refman/5.0/en/problems-with-float.html" target="_top">Problems with Floating-Point Comparisons</a>.
      </p></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="comparison-operators"></a>Comparison Functions and Operators</h3></div></div></div><a id="id2815139" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_between"><code class="literal">BETWEEN ... AND ... </code></a></td><td>Check whether a value is within a range of values</td></tr><tr><td><a href="mysqlqb_functions.html#function_coalesce"><code class="literal">COALESCE()</code></a></td><td>Return the first non-NULL argument</td></tr><tr><td><a href="mysqlqb_functions.html#operator_equal-to"><code class="literal">&lt;=&gt;</code></a></td><td>NULL-safe equal to operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_equal"><code class="literal">=</code></a></td><td>Equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than-or-equal"><code class="literal">&gt;=</code></a></td><td>Greater than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than"><code class="literal">&gt;</code></a></td><td>Greater than operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_greatest"><code class="literal">GREATEST()</code></a></td><td>Return the largest argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_in"><code class="literal">IN</code></a></td><td>Check whether a value is not within a set of values</td></tr><tr><td><a href="mysqlqb_functions.html#function_interval"><code class="literal">INTERVAL()</code></a></td><td>Return the index of the argument   that is less than the first argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-null"><code class="literal">IS NULL</code></a></td><td>NULL value test</td></tr><tr><td><a href="mysqlqb_functions.html#operator_is"><code class="literal">IS</code></a></td><td>Test a value against a boolean</td></tr><tr><td><a href="mysqlqb_functions.html#function_isnull"><code class="literal">ISNULL()</code></a></td><td>Test whether the argument is NULL</td></tr><tr><td><a href="mysqlqb_functions.html#function_least"><code class="literal">LEAST()</code></a></td><td>Return the smallest argument</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than-or-equal"><code class="literal">&lt;=</code></a></td><td>Less than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than"><code class="literal">&lt;</code></a></td><td>Less than operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_like"><code class="literal">LIKE</code></a></td><td>Simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#function_not-between"><code class="literal">NOT BETWEEN ... AND ...</code></a></td><td>Check whether a value is not within a range of values</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-equal"><code class="literal">!=</code>, <code class="literal">&lt;&gt;</code></a></td><td>Not equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_not-in"><code class="literal">NOT IN</code></a></td><td>Check whether a value is not within a set of values</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-like"><code class="literal">NOT LIKE</code></a></td><td>Negation of simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#operator_sounds-like"><code class="literal">SOUNDS LIKE</code></a></td><td>Compare sounds</td></tr></tbody></table></div><a id="id2815425" class="indexterm"></a><a id="id2815432" class="indexterm"></a><p>
        Comparison operations result in a value of <code class="literal">1</code>
        (<code class="literal">TRUE</code>), <code class="literal">0</code>
        (<code class="literal">FALSE</code>), or <code class="literal">NULL</code>. These
        operations work for both numbers and strings. Strings are
        automatically converted to numbers and numbers to strings as
        necessary.
      </p><p>
        Some of the functions in this section (such as
        <code class="literal">LEAST()</code> and <code class="literal">GREATEST()</code>)
        return values other than <code class="literal">1</code>
        (<code class="literal">TRUE</code>), <code class="literal">0</code>
        (<code class="literal">FALSE</code>), or <code class="literal">NULL</code>. However,
        the value they return is based on comparison operations
        performed according to the rules described in
        <a href="mysqlqb_functions.html#type-conversion" title="Type Conversion in Expression Evaluation">Type Conversion in Expression Evaluation</a>.
      </p><p>
        To convert a value to a specific type for comparison purposes,
        you can use the <code class="literal">CAST()</code> function. String
        values can be converted to a different character set using
        <code class="literal">CONVERT()</code>. See
        <a href="mysqlqb_functions.html#cast-functions" title="Cast Functions and Operators">Cast Functions and Operators</a>.
      </p><p>
        By default, string comparisons are not case sensitive and use
        the current character set. The default is
        <code class="literal">latin1</code> (cp1252 West European), which also
        works well for English.
      </p><div class="itemizedlist"><ul type="disc"><li><p><a id="operator_equal"></a>
            <a id="id2815550" class="indexterm"></a>

            <a id="id2815557" class="indexterm"></a>

            <code class="literal">=</code>
          </p><p>
            Equal:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 = 0;</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT '0' = 0;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT '0.0' = 0;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT '0.01' = 0;</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT '.01' = 0.01;</code></strong>
        -&gt; 1
</pre></li><li><p><a id="operator_equal-to"></a>
            <a id="id2815625" class="indexterm"></a>

            <a id="id2815636" class="indexterm"></a>

            <code class="literal">&lt;=&gt;</code>
          </p><p>
            <code class="literal">NULL</code>-safe equal. This operator performs
            an equality comparison like the <code class="literal">=</code>
            operator, but returns <code class="literal">1</code> rather than
            <code class="literal">NULL</code> if both operands are
            <code class="literal">NULL</code>, and <code class="literal">0</code> rather
            than <code class="literal">NULL</code> if one operand is
            <code class="literal">NULL</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 &lt;=&gt; 1, NULL &lt;=&gt; NULL, 1 &lt;=&gt; NULL;</code></strong>
        -&gt; 1, 1, 0
mysql&gt; <strong class="userinput"><code>SELECT 1 = 1, NULL = NULL, 1 = NULL;</code></strong>
        -&gt; 1, NULL, NULL
</pre></li><li><p><a id="operator_not-equal"></a>
            <a id="id2815740" class="indexterm"></a>

            <a id="id2815747" class="indexterm"></a>

            <a id="id2815754" class="indexterm"></a>

            <a id="id2815761" class="indexterm"></a>

            <code class="literal">&lt;&gt;</code>, <code class="literal">!=</code>
          </p><p>
            Not equal:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT '.01' &lt;&gt; '0.01';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT .01 &lt;&gt; '0.01';</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'zapp' &lt;&gt; 'zappp';</code></strong>
        -&gt; 1
</pre></li><li><p><a id="operator_less-than-or-equal"></a>
            <a id="id2815833" class="indexterm"></a>

            <a id="id2815841" class="indexterm"></a>

            <code class="literal">&lt;=</code>
          </p><p>
            Less than or equal:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 0.1 &lt;= 2;</code></strong>
        -&gt; 1
</pre></li><li><p><a id="operator_less-than"></a>
            <a id="id2815901" class="indexterm"></a>

            <a id="id2815908" class="indexterm"></a>

            <code class="literal">&lt;</code>
          </p><p>
            Less than:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 2 &lt; 2;</code></strong>
        -&gt; 0
</pre></li><li><p><a id="operator_greater-than-or-equal"></a>
            <a id="id2815969" class="indexterm"></a>

            <a id="id2815976" class="indexterm"></a>

            <code class="literal">&gt;=</code>
          </p><p>
            Greater than or equal:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 2 &gt;= 2;</code></strong>
        -&gt; 1
</pre></li><li><p><a id="operator_greater-than"></a>
            <a id="id2816037" class="indexterm"></a>

            <a id="id2816044" class="indexterm"></a>

            <code class="literal">&gt;</code>
          </p><p>
            Greater than:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 2 &gt; 2;</code></strong>
        -&gt; 0
</pre></li><li><p><a id="operator_is"></a>
            <a id="id2816105" class="indexterm"></a>

            <a id="id2816116" class="indexterm"></a>

            <a id="id2816127" class="indexterm"></a>

            <a id="id2816138" class="indexterm"></a>

            <a id="id2816145" class="indexterm"></a>

            <code class="literal">IS
            <em class="replaceable"><code>boolean_value</code></em></code>,
            <code class="literal">IS NOT
            <em class="replaceable"><code>boolean_value</code></em></code>
          </p><p>
            Tests a value against a boolean value, where
            <em class="replaceable"><code>boolean_value</code></em> can be
            <code class="literal">TRUE</code>, <code class="literal">FALSE</code>, or
            <code class="literal">UNKNOWN</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 IS TRUE, 0 IS FALSE, NULL IS UNKNOWN;</code></strong>
        -&gt; 1, 1, 1
mysql&gt; <strong class="userinput"><code>SELECT 1 IS NOT UNKNOWN, 0 IS NOT UNKNOWN, NULL IS NOT UNKNOWN;</code></strong>
        -&gt; 1, 1, 0
</pre><p>
            <code class="literal">IS [NOT]
            <em class="replaceable"><code>boolean_value</code></em></code> syntax
            was added in MySQL 5.0.2.
          </p></li><li><p><a id="function_is-null"></a>
            <a id="id2816254" class="indexterm"></a>

            <a id="id2816265" class="indexterm"></a>

            <a id="id2816272" class="indexterm"></a>

            <code class="literal">IS NULL</code>, <code class="literal">IS NOT NULL</code>
          </p><p>
            Tests whether a value is or is not <code class="literal">NULL</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 IS NULL, 0 IS NULL, NULL IS NULL;</code></strong>
        -&gt; 0, 0, 1
mysql&gt; <strong class="userinput"><code>SELECT 1 IS NOT NULL, 0 IS NOT NULL, NULL IS NOT NULL;</code></strong>
        -&gt; 1, 1, 0
</pre><p>
            <a id="id2816329" class="indexterm"></a>

            <a id="id2816335" class="indexterm"></a>

            To work well with ODBC programs, MySQL supports the
            following extra features when using <code class="literal">IS
            NULL</code>:
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                You can find the row that contains the most recent
                <code class="literal">AUTO_INCREMENT</code> value by issuing a
                statement of the following form immediately after
                generating the value:
              </p><pre class="programlisting">SELECT * FROM <em class="replaceable"><code>tbl_name</code></em> WHERE <em class="replaceable"><code>auto_col</code></em> IS NULL
</pre><p>
                This behavior can be disabled by setting
                <code class="literal">SQL_AUTO_IS_NULL=0</code>. See
                <a href="mysqlqb_statements.html#set-option" title="SET Syntax"><code class="literal">SET</code> Syntax</a>.
              </p></li><li><p>
                For <code class="literal">DATE</code> and
                <code class="literal">DATETIME</code> columns that are declared as
                <code class="literal">NOT NULL</code>, you can find the special
                date <code class="literal">'0000-00-00'</code> by using a
                statement like this:
              </p><pre class="programlisting">SELECT * FROM <em class="replaceable"><code>tbl_name</code></em> WHERE <em class="replaceable"><code>date_column</code></em> IS NULL
</pre><p>
                This is needed to get some ODBC applications to work
                because ODBC does not support a
                <code class="literal">'0000-00-00'</code> date value.
              </p></li></ul></div></li><li><p><a id="function_between"></a>
            <a id="id2816465" class="indexterm"></a>

            <code class="literal"><em class="replaceable"><code>expr</code></em> BETWEEN
            <em class="replaceable"><code>min</code></em> AND
            <em class="replaceable"><code>max</code></em></code>
          </p><p>
            If <em class="replaceable"><code>expr</code></em> is greater than or equal
            to <em class="replaceable"><code>min</code></em> and
            <em class="replaceable"><code>expr</code></em> is less than or equal to
            <em class="replaceable"><code>max</code></em>, <code class="literal">BETWEEN</code>
            returns <code class="literal">1</code>, otherwise it returns
            <code class="literal">0</code>. This is equivalent to the expression
            <code class="literal">(<em class="replaceable"><code>min</code></em> &lt;=
            <em class="replaceable"><code>expr</code></em> AND
            <em class="replaceable"><code>expr</code></em> &lt;=
            <em class="replaceable"><code>max</code></em>)</code> if all the
            arguments are of the same type. Otherwise type conversion
            takes place according to the rules described in
            <a href="mysqlqb_functions.html#type-conversion" title="Type Conversion in Expression Evaluation">Type Conversion in Expression Evaluation</a>, but applied to all the
            three arguments.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 BETWEEN 2 AND 3;</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'b' BETWEEN 'a' AND 'c';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 2 BETWEEN 2 AND '3';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 2 BETWEEN 2 AND 'x-3';</code></strong>
        -&gt; 0
</pre><p>
            For best results when using <code class="literal">BETWEEN</code> with
            date or time values, you should use
            <code class="literal">CAST()</code> to explicitly convert the values
            to the desired data type. Examples: If you compare a
            <code class="literal">DATETIME</code> to two <code class="literal">DATE</code>
            values, convert the <code class="literal">DATE</code> values to
            <code class="literal">DATETIME</code> values. If you use a string
            constant such as <code class="literal">'2001-1-1'</code> in a
            comparison to a <code class="literal">DATE</code>, cast the string to
            a <code class="literal">DATE</code>.
          </p></li><li><p><a id="function_not-between"></a>
            <a id="id2816644" class="indexterm"></a>

            <code class="literal"><em class="replaceable"><code>expr</code></em> NOT BETWEEN
            <em class="replaceable"><code>min</code></em> AND
            <em class="replaceable"><code>max</code></em></code>
          </p><p>
            This is the same as <code class="literal">NOT
            (<em class="replaceable"><code>expr</code></em> BETWEEN
            <em class="replaceable"><code>min</code></em> AND
            <em class="replaceable"><code>max</code></em>)</code>.
          </p></li><li><p><a id="function_coalesce"></a>
            <a id="id2816715" class="indexterm"></a>

            <a id="id2816726" class="indexterm"></a>

            <code class="literal">COALESCE(<em class="replaceable"><code>value</code></em>,...)</code>
          </p><p>
            Returns the first non-<code class="literal">NULL</code> value in the
            list, or <code class="literal">NULL</code> if there are no
            non-<code class="literal">NULL</code> values.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT COALESCE(NULL,1);</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT COALESCE(NULL,NULL,NULL);</code></strong>
        -&gt; NULL
</pre></li><li><p><a id="function_greatest"></a>
            <a id="id2816807" class="indexterm"></a>

            <code class="literal">GREATEST(<em class="replaceable"><code>value1</code></em>,<em class="replaceable"><code>value2</code></em>,...)</code>
          </p><p>
            With two or more arguments, returns the largest
            (maximum-valued) argument. The arguments are compared using
            the same rules as for <code class="literal">LEAST()</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT GREATEST(2,0);</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT GREATEST(34.0,3.0,5.0,767.0);</code></strong>
        -&gt; 767.0
mysql&gt; <strong class="userinput"><code>SELECT GREATEST('B','A','C');</code></strong>
        -&gt; 'C'
</pre><p>
            Before MySQL 5.0.13, <code class="literal">GREATEST()</code> returns
            <code class="literal">NULL</code> only if all arguments are
            <code class="literal">NULL</code>. As of 5.0.13, it returns
            <code class="literal">NULL</code> if any argument is
            <code class="literal">NULL</code>.
          </p></li><li><p><a id="function_in"></a>
            <a id="id2816914" class="indexterm"></a>

            <code class="literal"><em class="replaceable"><code>expr</code></em> IN
            (<em class="replaceable"><code>value</code></em>,...)</code>
          </p><p>
            Returns <code class="literal">1</code> if
            <em class="replaceable"><code>expr</code></em> is equal to any of the
            values in the <code class="literal">IN</code> list, else returns
            <code class="literal">0</code>. If all values are constants, they are
            evaluated according to the type of
            <em class="replaceable"><code>expr</code></em> and sorted. The search for
            the item then is done using a binary search. This means
            <code class="literal">IN</code> is very quick if the
            <code class="literal">IN</code> value list consists entirely of
            constants. Otherwise, type conversion takes place according
            to the rules described in <a href="mysqlqb_functions.html#type-conversion" title="Type Conversion in Expression Evaluation">Type Conversion in Expression Evaluation</a>,
            but applied to all the arguments.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 2 IN (0,3,5,7);</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'wefwf' IN ('wee','wefwf','weg');</code></strong>
        -&gt; 1
</pre><p>
            You should never mix quoted and unquoted values in an
            <code class="literal">IN</code> list because the comparison rules for
            quoted values (such as strings) and unquoted values (such as
            numbers) differ. Mixing types may therefore lead to
            inconsistent results. For example, do not write an
            <code class="literal">IN</code> expression like this:
          </p><pre class="programlisting">SELECT val1 FROM tbl1 WHERE val1 IN (1,2,'a');
</pre><p>
            Instead, write it like this:
          </p><pre class="programlisting">SELECT val1 FROM tbl1 WHERE val1 IN ('1','2','a');
</pre><p>
            The number of values in the <code class="literal">IN</code> list is
            only limited by the <code class="literal">max_allowed_packet</code>
            value.
          </p><p>
            To comply with the SQL standard, <code class="literal">IN</code>
            returns <code class="literal">NULL</code> not only if the expression
            on the left hand side is <code class="literal">NULL</code>, but also
            if no match is found in the list and one of the expressions
            in the list is <code class="literal">NULL</code>.
          </p><p>
            <code class="literal">IN()</code> syntax can also be used to write
            certain types of subqueries. See
            <a href="mysqlqb_statements.html#any-in-some-subqueries" title="Subqueries with ANY, IN, and&#10;          SOME">Subqueries with <code class="literal">ANY</code>, <code class="literal">IN</code>, and
          <code class="literal">SOME</code></a>.
          </p></li><li><p><a id="function_not-in"></a>
            <a id="id2817098" class="indexterm"></a>

            <code class="literal"><em class="replaceable"><code>expr</code></em> NOT IN
            (<em class="replaceable"><code>value</code></em>,...)</code>
          </p><p>
            This is the same as <code class="literal">NOT
            (<em class="replaceable"><code>expr</code></em> IN
            (<em class="replaceable"><code>value</code></em>,...))</code>.
          </p></li><li><p><a id="function_isnull"></a>
            <a id="id2817164" class="indexterm"></a>

            <code class="literal">ISNULL(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            If <em class="replaceable"><code>expr</code></em> is
            <code class="literal">NULL</code>, <code class="literal">ISNULL()</code> returns
            <code class="literal">1</code>, otherwise it returns
            <code class="literal">0</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ISNULL(1+1);</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT ISNULL(1/0);</code></strong>
        -&gt; 1
</pre><p>
            <code class="literal">ISNULL()</code> can be used instead of
            <code class="literal">=</code> to test whether a value is
            <code class="literal">NULL</code>. (Comparing a value to
            <code class="literal">NULL</code> using <code class="literal">=</code> always
            yields false.)
          </p><p>
            The <code class="literal">ISNULL()</code> function shares some special
            behaviors with the <code class="literal">IS NULL</code> comparison
            operator. See the description of <code class="literal">IS NULL</code>.
          </p></li><li><p><a id="function_interval"></a>
            <a id="id2817291" class="indexterm"></a>

            <code class="literal">INTERVAL(<em class="replaceable"><code>N</code></em>,<em class="replaceable"><code>N1</code></em>,<em class="replaceable"><code>N2</code></em>,<em class="replaceable"><code>N3</code></em>,...)</code>
          </p><p>
            Returns <code class="literal">0</code> if <em class="replaceable"><code>N</code></em>
            &lt; <em class="replaceable"><code>N1</code></em>, <code class="literal">1</code> if
            <em class="replaceable"><code>N</code></em> &lt;
            <em class="replaceable"><code>N2</code></em> and so on or
            <code class="literal">-1</code> if <em class="replaceable"><code>N</code></em> is
            <code class="literal">NULL</code>. All arguments are treated as
            integers. It is required that <em class="replaceable"><code>N1</code></em>
            &lt; <em class="replaceable"><code>N2</code></em> &lt;
            <em class="replaceable"><code>N3</code></em> &lt; <code class="literal">...</code>
            &lt; <em class="replaceable"><code>Nn</code></em> for this function to work
            correctly. This is because a binary search is used (very
            fast).
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT INTERVAL(23, 1, 15, 17, 30, 44, 200);</code></strong>
        -&gt; 3
mysql&gt; <strong class="userinput"><code>SELECT INTERVAL(10, 1, 10, 100, 1000);</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT INTERVAL(22, 23, 30, 44, 200);</code></strong>
        -&gt; 0
</pre></li><li><p><a id="function_least"></a>
            <a id="id2817428" class="indexterm"></a>

            <code class="literal">LEAST(<em class="replaceable"><code>value1</code></em>,<em class="replaceable"><code>value2</code></em>,...)</code>
          </p><p>
            With two or more arguments, returns the smallest
            (minimum-valued) argument. The arguments are compared using
            the following rules:
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                If the return value is used in an
                <code class="literal">INTEGER</code> context or all arguments are
                integer-valued, they are compared as integers.
              </p></li><li><p>
                If the return value is used in a <code class="literal">REAL</code>
                context or all arguments are real-valued, they are
                compared as reals.
              </p></li><li><p>
                If any argument is a case-sensitive string, the
                arguments are compared as case-sensitive strings.
              </p></li><li><p>
                In all other cases, the arguments are compared as
                case-insensitive strings.
              </p></li></ul></div><p>
            Before MySQL 5.0.13, <code class="literal">LEAST()</code> returns
            <code class="literal">NULL</code> only if all arguments are
            <code class="literal">NULL</code>. As of 5.0.13, it returns
            <code class="literal">NULL</code> if any argument is
            <code class="literal">NULL</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LEAST(2,0);</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT LEAST(34.0,3.0,5.0,767.0);</code></strong>
        -&gt; 3.0
mysql&gt; <strong class="userinput"><code>SELECT LEAST('B','A','C');</code></strong>
        -&gt; 'A'
</pre><p>
            Note that the preceding conversion rules can produce strange
            results in some borderline cases:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CAST(LEAST(3600, 9223372036854775808.0) as SIGNED);</code></strong>
        -&gt; -9223372036854775808
</pre><p>
            This happens because MySQL reads
            <code class="literal">9223372036854775808.0</code> in an integer
            context. The integer representation is not good enough to
            hold the value, so it wraps to a signed integer.
          </p></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="logical-operators"></a>Logical Operators</h3></div></div></div><a id="id2817596" class="indexterm"></a><a id="id2817603" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#operator_and"><code class="literal">AND</code>, <code class="literal">&amp;&amp;</code></a></td><td>Logical AND</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not"><code class="literal">NOT</code>, <code class="literal">!</code></a></td><td>Negates value</td></tr><tr><td><a href="mysqlqb_functions.html#operator_or"><code class="literal">||</code>, <code class="literal">OR</code></a></td><td>Logical OR</td></tr><tr><td><a href="mysqlqb_functions.html#operator_xor"><code class="literal">XOR</code></a></td><td>Logical XOR</td></tr></tbody></table></div><p>
        In SQL, all logical operators evaluate to
        <code class="literal">TRUE</code>, <code class="literal">FALSE</code>, or
        <code class="literal">NULL</code> (<code class="literal">UNKNOWN</code>). In MySQL,
        these are implemented as 1 (<code class="literal">TRUE</code>), 0
        (<code class="literal">FALSE</code>), and <code class="literal">NULL</code>. Most of
        this is common to different SQL database servers, although some
        servers may return any non-zero value for
        <code class="literal">TRUE</code>.
      </p><div class="itemizedlist"><ul type="disc"><li><p><a id="operator_not"></a>
            <a id="id2817773" class="indexterm"></a>

            <a id="id2817783" class="indexterm"></a>

            <code class="literal">NOT</code>, <code class="literal">!</code>
          </p><p>
            Logical NOT. Evaluates to <code class="literal">1</code> if the
            operand is <code class="literal">0</code>, to <code class="literal">0</code> if
            the operand is non-zero, and <code class="literal">NOT NULL</code>
            returns <code class="literal">NULL</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT NOT 10;</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT NOT 0;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT NOT NULL;</code></strong>
        -&gt; NULL
mysql&gt; <strong class="userinput"><code>SELECT ! (1+1);</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT ! 1+1;</code></strong>
        -&gt; 1
</pre><p>
            The last example produces <code class="literal">1</code> because the
            expression evaluates the same way as
            <code class="literal">(!1)+1</code>.
          </p><p>
            Note that the precedence of the <code class="literal">NOT</code>
            operator changed in MySQL 5.0.2. See
            <a href="mysqlqb_functions.html#operator-precedence" title="Operator Precedence">Operator Precedence</a>.
          </p></li><li><p><a id="operator_and"></a>
            <a id="id2817914" class="indexterm"></a>

            <a id="id2817924" class="indexterm"></a>

            <code class="literal">AND</code>, <code class="literal">&amp;&amp;</code>
          </p><p>
            Logical AND. Evaluates to <code class="literal">1</code> if all
            operands are non-zero and not <code class="literal">NULL</code>, to
            <code class="literal">0</code> if one or more operands are
            <code class="literal">0</code>, otherwise <code class="literal">NULL</code> is
            returned.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 &amp;&amp; 1;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 1 &amp;&amp; 0;</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 1 &amp;&amp; NULL;</code></strong>
        -&gt; NULL
mysql&gt; <strong class="userinput"><code>SELECT 0 &amp;&amp; NULL;</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT NULL &amp;&amp; 0;</code></strong>
        -&gt; 0
</pre></li><li><p><a id="operator_or"></a>
            <a id="id2818031" class="indexterm"></a>

            <a id="id2818041" class="indexterm"></a>

            <code class="literal">OR</code>, <code class="literal">||</code>
          </p><p>
            Logical OR. When both operands are
            non-<code class="literal">NULL</code>, the result is
            <code class="literal">1</code> if any operand is non-zero, and
            <code class="literal">0</code> otherwise. With a
            <code class="literal">NULL</code> operand, the result is
            <code class="literal">1</code> if the other operand is non-zero, and
            <code class="literal">NULL</code> otherwise. If both operands are
            <code class="literal">NULL</code>, the result is
            <code class="literal">NULL</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 || 1;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 1 || 0;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 0 || 0;</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 0 || NULL;</code></strong>
        -&gt; NULL
mysql&gt; <strong class="userinput"><code>SELECT 1 || NULL;</code></strong>
        -&gt; 1
</pre></li><li><p><a id="operator_xor"></a>
            <a id="id2818154" class="indexterm"></a>

            <code class="literal">XOR</code>
          </p><p>
            Logical XOR. Returns <code class="literal">NULL</code> if either
            operand is <code class="literal">NULL</code>. For
            non-<code class="literal">NULL</code> operands, evaluates to
            <code class="literal">1</code> if an odd number of operands is
            non-zero, otherwise <code class="literal">0</code> is returned.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 XOR 1;</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 1 XOR 0;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 1 XOR NULL;</code></strong>
        -&gt; NULL
mysql&gt; <strong class="userinput"><code>SELECT 1 XOR 1 XOR 1;</code></strong>
        -&gt; 1
</pre><p>
            <code class="literal">a XOR b</code> is mathematically equal to
            <code class="literal">(a AND (NOT b)) OR ((NOT a) and b)</code>.
          </p></li></ul></div></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="control-flow-functions"></a>Control Flow Functions</h2></div></div></div><a id="id2818256" class="indexterm"></a><a id="id2818263" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_case"><code class="literal">CASE</code></a></td><td>Case statement</td></tr><tr><td><a href="mysqlqb_functions.html#function_if"><code class="literal">IF()</code></a></td><td>If/else construct</td></tr><tr><td><a href="mysqlqb_functions.html#function_ifnull"><code class="literal">IFNULL()</code></a></td><td>Null if/else construct</td></tr><tr><td><a href="mysqlqb_functions.html#function_nullif"><code class="literal">NULLIF()</code></a></td><td>Return NULL if expr1 = expr2</td></tr></tbody></table></div><div class="itemizedlist"><ul type="disc"><li><p><a id="function_case"></a>
          <a id="id2818390" class="indexterm"></a>

          <code class="literal">CASE <em class="replaceable"><code>value</code></em> WHEN
          [<em class="replaceable"><code>compare_value</code></em>] THEN
          <em class="replaceable"><code>result</code></em> [WHEN
          [<em class="replaceable"><code>compare_value</code></em>] THEN
          <em class="replaceable"><code>result</code></em> ...] [ELSE
          <em class="replaceable"><code>result</code></em>] END</code>
        </p><p>
          <code class="literal">CASE WHEN [<em class="replaceable"><code>condition</code></em>] THEN
          <em class="replaceable"><code>result</code></em> [WHEN
          [<em class="replaceable"><code>condition</code></em>] THEN
          <em class="replaceable"><code>result</code></em> ...] [ELSE
          <em class="replaceable"><code>result</code></em>] END</code>
        </p><p>
          The first version returns the
          <em class="replaceable"><code>result</code></em> where
          <code class="literal"><em class="replaceable"><code>value</code></em>=<em class="replaceable"><code>compare_value</code></em></code>.
          The second version returns the result for the first condition
          that is true. If there was no matching result value, the
          result after <code class="literal">ELSE</code> is returned, or
          <code class="literal">NULL</code> if there is no <code class="literal">ELSE</code>
          part.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CASE 1 WHEN 1 THEN 'one'</code></strong>
    -&gt;     <strong class="userinput"><code>WHEN 2 THEN 'two' ELSE 'more' END;</code></strong>
        -&gt; 'one'
mysql&gt; <strong class="userinput"><code>SELECT CASE WHEN 1&gt;0 THEN 'true' ELSE 'false' END;</code></strong>
        -&gt; 'true'
mysql&gt; <strong class="userinput"><code>SELECT CASE BINARY 'B'</code></strong>
    -&gt;     <strong class="userinput"><code>WHEN 'a' THEN 1 WHEN 'b' THEN 2 END;</code></strong>
        -&gt; NULL
</pre><p>
          The default return type of a <code class="literal">CASE</code>
          expression is the compatible aggregated type of all return
          values, but also depends on the context in which it is used.
          If used in a string context, the result is returned as a
          string. If used in a numeric context, then the result is
          returned as a decimal, real, or integer value.
        </p><p>
          <span class="bold"><strong>Note</strong></span>: The syntax of the
          <code class="literal">CASE</code> <span class="emphasis"><em>expression</em></span> shown
          here differs slightly from that of the SQL
          <code class="literal">CASE</code> <span class="emphasis"><em>statement</em></span>
          described in <a href="http://dev.mysql.com/doc/refman/5.0/en/case-statement.html" target="_top"><code class="literal">CASE</code> Statement</a>, for use inside
          stored routines. The <code class="literal">CASE</code> statement cannot
          have an <code class="literal">ELSE NULL</code> clause, and it is
          terminated with <code class="literal">END CASE</code> instead of
          <code class="literal">END</code>.
        </p></li><li><p><a id="function_if"></a>
          <a id="id2818618" class="indexterm"></a>

          <code class="literal">IF(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>,<em class="replaceable"><code>expr3</code></em>)</code>
        </p><p>
          If <em class="replaceable"><code>expr1</code></em> is <code class="literal">TRUE</code>
          (<code class="literal"><em class="replaceable"><code>expr1</code></em> &lt;&gt;
          0</code> and <code class="literal"><em class="replaceable"><code>expr1</code></em>
          &lt;&gt; NULL</code>) then <code class="literal">IF()</code> returns
          <em class="replaceable"><code>expr2</code></em>; otherwise it returns
          <em class="replaceable"><code>expr3</code></em>. <code class="literal">IF()</code>
          returns a numeric or string value, depending on the context in
          which it is used.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT IF(1&gt;2,2,3);</code></strong>
        -&gt; 3
mysql&gt; <strong class="userinput"><code>SELECT IF(1&lt;2,'yes','no');</code></strong>
        -&gt; 'yes'
mysql&gt; <strong class="userinput"><code>SELECT IF(STRCMP('test','test1'),'no','yes');</code></strong>
        -&gt; 'no'
</pre><p>
          If only one of <em class="replaceable"><code>expr2</code></em> or
          <em class="replaceable"><code>expr3</code></em> is explicitly
          <code class="literal">NULL</code>, the result type of the
          <code class="literal">IF()</code> function is the type of the
          non-<code class="literal">NULL</code> expression.
        </p><p>
          <em class="replaceable"><code>expr1</code></em> is evaluated as an integer
          value, which means that if you are testing floating-point or
          string values, you should do so using a comparison operation.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT IF(0.1,1,0);</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT IF(0.1&lt;&gt;0,1,0);</code></strong>
        -&gt; 1
</pre><p>
          In the first case shown, <code class="literal">IF(0.1)</code> returns
          <code class="literal">0</code> because <code class="literal">0.1</code> is
          converted to an integer value, resulting in a test of
          <code class="literal">IF(0)</code>. This may not be what you expect. In
          the second case, the comparison tests the original
          floating-point value to see whether it is non-zero. The result
          of the comparison is used as an integer.
        </p><p>
          The default return type of <code class="literal">IF()</code> (which may
          matter when it is stored into a temporary table) is calculated
          as follows:
        </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>Expression</strong></span></td><td><span class="bold"><strong>Return Value</strong></span></td></tr><tr><td><em class="replaceable"><code>expr2</code></em> or <em class="replaceable"><code>expr3</code></em>
                  returns a string</td><td>string</td></tr><tr><td><em class="replaceable"><code>expr2</code></em> or <em class="replaceable"><code>expr3</code></em>
                  returns a floating-point value</td><td>floating-point</td></tr><tr><td><em class="replaceable"><code>expr2</code></em> or <em class="replaceable"><code>expr3</code></em>
                  returns an integer</td><td>integer</td></tr></tbody></table></div><p>
          If <em class="replaceable"><code>expr2</code></em> and
          <em class="replaceable"><code>expr3</code></em> are both strings, the result
          is case sensitive if either string is case sensitive.
        </p><p>
          <span class="bold"><strong>Note</strong></span>: There is also an
          <code class="literal">IF</code> <span class="emphasis"><em>statement</em></span>, which
          differs from the <code class="literal">IF()</code>
          <span class="emphasis"><em>function</em></span> described here. See
          <a href="http://dev.mysql.com/doc/refman/5.0/en/if-statement.html" target="_top"><code class="literal">IF</code> Statement</a>.
        </p></li><li><p><a id="function_ifnull"></a>
          <a id="id2818949" class="indexterm"></a>

          <a id="id2818960" class="indexterm"></a>

          <code class="literal">IFNULL(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>)</code>
        </p><p>
          If <em class="replaceable"><code>expr1</code></em> is not
          <code class="literal">NULL</code>, <code class="literal">IFNULL()</code> returns
          <em class="replaceable"><code>expr1</code></em>; otherwise it returns
          <em class="replaceable"><code>expr2</code></em>. <code class="literal">IFNULL()</code>
          returns a numeric or string value, depending on the context in
          which it is used.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT IFNULL(1,0);</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT IFNULL(NULL,10);</code></strong>
        -&gt; 10
mysql&gt; <strong class="userinput"><code>SELECT IFNULL(1/0,10);</code></strong>
        -&gt; 10
mysql&gt; <strong class="userinput"><code>SELECT IFNULL(1/0,'yes');</code></strong>
        -&gt; 'yes'
</pre><p>
          The default result value of
          <code class="literal">IFNULL(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>)</code>
          is the more “<span class="quote">general</span>” of the two expressions, in
          the order <code class="literal">STRING</code>, <code class="literal">REAL</code>,
          or <code class="literal">INTEGER</code>. Consider the case of a table
          based on expressions or where MySQL must internally store a
          value returned by <code class="literal">IFNULL()</code> in a temporary
          table:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>CREATE TABLE tmp SELECT IFNULL(1,'test') AS test;</code></strong>
mysql&gt; <strong class="userinput"><code>DESCRIBE tmp;</code></strong>
+-------+---------+------+-----+---------+-------+
| Field | Type    | Null | Key | Default | Extra |
+-------+---------+------+-----+---------+-------+
| test  | char(4) |      |     |         |       |
+-------+---------+------+-----+---------+-------+
</pre><p>
          In this example, the type of the <code class="literal">test</code>
          column is <code class="literal">CHAR(4)</code>.
        </p></li><li><p><a id="function_nullif"></a>
          <a id="id2819129" class="indexterm"></a>

          <code class="literal">NULLIF(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>)</code>
        </p><p>
          Returns <code class="literal">NULL</code> if
          <code class="literal"><em class="replaceable"><code>expr1</code></em> =
          <em class="replaceable"><code>expr2</code></em></code> is true, otherwise
          returns <em class="replaceable"><code>expr1</code></em>. This is the same as
          <code class="literal">CASE WHEN <em class="replaceable"><code>expr1</code></em> =
          <em class="replaceable"><code>expr2</code></em> THEN NULL ELSE
          <em class="replaceable"><code>expr1</code></em> END</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT NULLIF(1,1);</code></strong>
        -&gt; NULL
mysql&gt; <strong class="userinput"><code>SELECT NULLIF(1,2);</code></strong>
        -&gt; 1
</pre><p>
          Note that MySQL evaluates <em class="replaceable"><code>expr1</code></em>
          twice if the arguments are not equal.
        </p></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="string-functions"></a>String Functions</h2></div></div></div><div class="toc"><dl><dt><span class="section"><a href="mysqlqb_functions.html#string-comparison-functions">String Comparison Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#regexp">Regular Expressions</a></span></dt></dl></div><a id="id2819240" class="indexterm"></a><a id="id2819247" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_ascii"><code class="literal">ASCII()</code></a></td><td>Return numeric value of left-most character</td></tr><tr><td><a href="mysqlqb_functions.html#function_bin"><code class="literal">BIN()</code></a></td><td>Return a string representation of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-length"><code class="literal">BIT_LENGTH()</code></a></td><td>Return length of argument in bits</td></tr><tr><td><a href="mysqlqb_functions.html#function_char-length"><code class="literal">CHAR_LENGTH()</code></a></td><td>Return number of characters in argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_char"><code class="literal">CHAR()</code></a></td><td>Return the character for each integer passed</td></tr><tr><td><a href="mysqlqb_functions.html#function_character-length"><code class="literal">CHARACTER_LENGTH()</code></a></td><td>A synonym for CHAR_LENGTH()</td></tr><tr><td><a href="mysqlqb_functions.html#function_concat-ws"><code class="literal">CONCAT_WS()</code></a></td><td>Return concatenate with  separator</td></tr><tr><td><a href="mysqlqb_functions.html#function_concat"><code class="literal">CONCAT()</code></a></td><td>Return concatenated string</td></tr><tr><td><a href="mysqlqb_functions.html#function_conv"><code class="literal">CONV()</code></a></td><td>Convert numbers between different number bases</td></tr><tr><td><a href="mysqlqb_functions.html#function_elt"><code class="literal">ELT()</code></a></td><td>Return string at index number</td></tr><tr><td><a href="mysqlqb_functions.html#operator_equal-to"><code class="literal">&lt;=&gt;</code></a></td><td>NULL-safe equal to operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_equal"><code class="literal">=</code></a></td><td>Equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_export-set"><code class="literal">EXPORT_SET()</code></a></td><td>Return a string such that for every bit set in the value  bits, you get an on string and for every unset bit, you get an off string</td></tr><tr><td><a href="mysqlqb_functions.html#function_field"><code class="literal">FIELD()</code></a></td><td>Return the index (position) of the first argument  in the subsequent arguments</td></tr><tr><td><a href="mysqlqb_functions.html#function_find-in-set"><code class="literal">FIND_IN_SET()</code></a></td><td>Return the number of times the first argument is found   in the second arguments</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than-or-equal"><code class="literal">&gt;=</code></a></td><td>Greater than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than"><code class="literal">&gt;</code></a></td><td>Greater than operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_hex"><code class="literal">HEX()</code></a></td><td>Return a string representation of a hex value</td></tr><tr><td><a href="mysqlqb_functions.html#function_insert"><code class="literal">INSERT()</code></a></td><td>Insert a substring at the specified position up to  the specified number of characters</td></tr><tr><td><a href="mysqlqb_functions.html#function_instr"><code class="literal">INSTR()</code></a></td><td>Return the index of the first occurrence of substring</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-null"><code class="literal">IS NULL</code></a></td><td>NULL value test</td></tr><tr><td><a href="mysqlqb_functions.html#operator_is"><code class="literal">IS</code></a></td><td>Test a value against a boolean</td></tr><tr><td><a href="mysqlqb_functions.html#function_lcase"><code class="literal">LCASE()</code></a></td><td>Synonym for LOWER() </td></tr><tr><td><a href="mysqlqb_functions.html#function_left"><code class="literal">LEFT()</code></a></td><td>Return the leftmost number of characters as specified</td></tr><tr><td><a href="mysqlqb_functions.html#function_length"><code class="literal">LENGTH()</code></a></td><td>Return the length of a string in bytes</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than-or-equal"><code class="literal">&lt;=</code></a></td><td>Less than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than"><code class="literal">&lt;</code></a></td><td>Less than operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_like"><code class="literal">LIKE</code></a></td><td>Simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#function_load-file"><code class="literal">LOAD_FILE()</code></a></td><td>Load the named file</td></tr><tr><td><a href="mysqlqb_functions.html#function_locate"><code class="literal">LOCATE()</code></a></td><td>Return the position of the first occurrence  of substring</td></tr><tr><td><a href="mysqlqb_functions.html#function_lower"><code class="literal">LOWER()</code></a></td><td>Return the argument in lowercase </td></tr><tr><td><a href="mysqlqb_functions.html#function_lpad"><code class="literal">LPAD()</code></a></td><td>Return the string argument, left-padded  with the specified string</td></tr><tr><td><a href="mysqlqb_functions.html#function_ltrim"><code class="literal">LTRIM()</code></a></td><td>Remove leading spaces</td></tr><tr><td><a href="mysqlqb_functions.html#function_make-set"><code class="literal">MAKE_SET()</code></a></td><td>Return a set of comma-separated strings  that have the corresponding bit in bits set</td></tr><tr><td><a href="mysqlqb_functions.html#function_mid"><code class="literal">MID()</code></a></td><td>Return a substring starting from the specified position</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-equal"><code class="literal">!=</code>, <code class="literal">&lt;&gt;</code></a></td><td>Not equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-like"><code class="literal">NOT LIKE</code></a></td><td>Negation of simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-regexp"><code class="literal">NOT RGEXP</code></a></td><td>Negation of REGEXP</td></tr><tr><td><a href="mysqlqb_functions.html#function_oct"><code class="literal">OCT()</code></a></td><td>Return a string representation of the octal argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_octet-length"><code class="literal">OCTET_LENGTH()</code></a></td><td>A synonym for LENGTH()</td></tr><tr><td><a href="mysqlqb_functions.html#function_ord"><code class="literal">ORD()</code></a></td><td>If the leftmost character of the  argument is a multi-byte character,  returns the code for that character</td></tr><tr><td><a href="mysqlqb_functions.html#function_position"><code class="literal">POSITION()</code></a></td><td>A synonym for LOCATE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_quote"><code class="literal">QUOTE()</code></a></td><td>Escape the argument for use in an SQL statement</td></tr><tr><td><a href="mysqlqb_functions.html#operator_regexp"><code class="literal">REGEXP</code></a></td><td>Pattern matching using regular expressions</td></tr><tr><td><a href="mysqlqb_functions.html#function_repeat"><code class="literal">REPEAT()</code></a></td><td>Repeat a string the specified number of times</td></tr><tr><td><a href="mysqlqb_functions.html#function_replace"><code class="literal">REPLACE()</code></a></td><td>Replace occurrences of a specified string</td></tr><tr><td><a href="mysqlqb_functions.html#function_reverse"><code class="literal">REVERSE()</code></a></td><td>Reverse the characters in a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_right"><code class="literal">RIGHT()</code></a></td><td>Return the specified rightmost number of characters</td></tr><tr><td><a href="mysqlqb_functions.html#operator_regexp"><code class="literal">RLIKE</code></a></td><td>Synonym for REGEXP</td></tr><tr><td><a href="mysqlqb_functions.html#function_rpad"><code class="literal">RPAD()</code></a></td><td>Append string the specified number of times</td></tr><tr><td><a href="mysqlqb_functions.html#function_rtrim"><code class="literal">RTRIM()</code></a></td><td>Remove trailing spaces</td></tr><tr><td><a href="mysqlqb_functions.html#function_soundex"><code class="literal">SOUNDEX()</code></a></td><td>Return a soundex string</td></tr><tr><td><a href="mysqlqb_functions.html#operator_sounds-like"><code class="literal">SOUNDS LIKE</code></a></td><td>Compare sounds</td></tr><tr><td><a href="mysqlqb_functions.html#function_space"><code class="literal">SPACE()</code></a></td><td>Return a string of the specified number of spaces</td></tr><tr><td><a href="mysqlqb_functions.html#function_strcmp"><code class="literal">STRCMP()</code></a></td><td>Compare two strings</td></tr><tr><td><a href="mysqlqb_functions.html#function_substring-index"><code class="literal">SUBSTRING_INDEX()</code></a></td><td>Return a substring from a  string before the specified number of occurrences of the delimiter</td></tr><tr><td><a href="mysqlqb_functions.html#function_substring"><code class="literal">SUBSTRING()</code></a></td><td>Return the substring as specified</td></tr><tr><td><a href="mysqlqb_functions.html#function_trim"><code class="literal">TRIM()</code></a></td><td>Return </td></tr><tr><td><a href="mysqlqb_functions.html#function_ucase"><code class="literal">UCASE()</code></a></td><td>Synonym for UPPER()</td></tr><tr><td><a href="mysqlqb_functions.html#function_unhex"><code class="literal">UNHEX()</code></a></td><td>Convert each pair of hexadecimal digits  to a character</td></tr><tr><td><a href="mysqlqb_functions.html#function_upper"><code class="literal">UPPER()</code></a></td><td>Convert to uppercase</td></tr></tbody></table></div><p>
      String-valued functions return <code class="literal">NULL</code> if the
      length of the result would be greater than the value of the
      <code class="literal">max_allowed_packet</code> system variable. See
      <a href="http://dev.mysql.com/doc/refman/5.0/en/server-parameters.html" target="_top">Tuning Server Parameters</a>.
    </p><p>
      For functions that operate on string positions, the first position
      is numbered 1.
    </p><p>
      For functions that take length arguments, non-integer arguments
      are rounded to the nearest integer.
    </p><div class="itemizedlist"><ul type="disc"><li><p><a id="function_ascii"></a>
          <a id="id2820027" class="indexterm"></a>

          <code class="literal">ASCII(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the numeric value of the leftmost character of the
          string <em class="replaceable"><code>str</code></em>. Returns
          <code class="literal">0</code> if <em class="replaceable"><code>str</code></em> is the
          empty string. Returns <code class="literal">NULL</code> if
          <em class="replaceable"><code>str</code></em> is <code class="literal">NULL</code>.
          <code class="literal">ASCII()</code> works for characters with numeric
          values from <code class="literal">0</code> to <code class="literal">255</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ASCII('2');</code></strong>
        -&gt; 50
mysql&gt; <strong class="userinput"><code>SELECT ASCII(2);</code></strong>
        -&gt; 50
mysql&gt; <strong class="userinput"><code>SELECT ASCII('dx');</code></strong>
        -&gt; 100
</pre><p>
          See also the <code class="literal">ORD()</code> function.
        </p></li><li><p><a id="function_bin"></a>
          <a id="id2820141" class="indexterm"></a>

          <code class="literal">BIN(<em class="replaceable"><code>N</code></em>)</code>
        </p><p>
          Returns a string representation of the binary value of
          <em class="replaceable"><code>N</code></em>, where
          <em class="replaceable"><code>N</code></em> is a longlong
          (<code class="literal">BIGINT</code>) number. This is equivalent to
          <code class="literal">CONV(<em class="replaceable"><code>N</code></em>,10,2)</code>.
          Returns <code class="literal">NULL</code> if
          <em class="replaceable"><code>N</code></em> is <code class="literal">NULL</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT BIN(12);</code></strong>
        -&gt; '1100'
</pre></li><li><p><a id="function_bit-length"></a>
          <a id="id2820235" class="indexterm"></a>

          <code class="literal">BIT_LENGTH(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the length of the string
          <em class="replaceable"><code>str</code></em> in bits.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT BIT_LENGTH('text');</code></strong>
        -&gt; 32
</pre></li><li><p><a id="function_char"></a>
          <a id="id2820303" class="indexterm"></a>

          <code class="literal">CHAR(<em class="replaceable"><code>N</code></em>,... [USING
          <em class="replaceable"><code>charset_name</code></em>])</code>
        </p><p>
          <code class="literal">CHAR()</code> interprets each argument
          <em class="replaceable"><code>N</code></em> as an integer and returns a
          string consisting of the characters given by the code values
          of those integers. <code class="literal">NULL</code> values are skipped.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CHAR(77,121,83,81,'76');</code></strong>
        -&gt; 'MySQL'
mysql&gt; <strong class="userinput"><code>SELECT CHAR(77,77.3,'77.3');</code></strong>
        -&gt; 'MMM'
</pre><p>
          As of MySQL 5.0.15, <code class="literal">CHAR()</code> arguments larger
          than 255 are converted into multiple result bytes. For
          example, <code class="literal">CHAR(256)</code> is equivalent to
          <code class="literal">CHAR(1,0)</code>, and
          <code class="literal">CHAR(256*256)</code> is equivalent to
          <code class="literal">CHAR(1,0,0)</code>:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT HEX(CHAR(1,0)), HEX(CHAR(256));</code></strong>
+----------------+----------------+
| HEX(CHAR(1,0)) | HEX(CHAR(256)) |
+----------------+----------------+
| 0100           | 0100           |
+----------------+----------------+
mysql&gt; <strong class="userinput"><code>SELECT HEX(CHAR(1,0,0)), HEX(CHAR(256*256));</code></strong>
+------------------+--------------------+
| HEX(CHAR(1,0,0)) | HEX(CHAR(256*256)) |
+------------------+--------------------+
| 010000           | 010000             |
+------------------+--------------------+
</pre><p>
          By default, <code class="literal">CHAR()</code> returns a binary string.
          To produce a string in a given character set, use the optional
          <code class="literal">USING</code> clause:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CHARSET(CHAR(0x65)), CHARSET(CHAR(0x65 USING utf8));</code></strong>
+---------------------+--------------------------------+
| CHARSET(CHAR(0x65)) | CHARSET(CHAR(0x65 USING utf8)) |
+---------------------+--------------------------------+
| binary              | utf8                           |
+---------------------+--------------------------------+
</pre><p>
          If <code class="literal">USING</code> is given and the result string is
          illegal for the given character set, a warning is issued.
          Also, if strict SQL mode is enabled, the result from
          <code class="literal">CHAR()</code> becomes <code class="literal">NULL</code>.
        </p><p>
          Before MySQL 5.0.15, <code class="literal">CHAR()</code> returns a
          string in the connection character set and the
          <code class="literal">USING</code> clause is unavailable. In addition,
          each argument is interpreted modulo 256, so
          <code class="literal">CHAR(256)</code> and
          <code class="literal">CHAR(256*256)</code> both are equivalent to
          <code class="literal">CHAR(0)</code>.
        </p></li><li><p><a id="function_char-length"></a>
          <a id="id2820515" class="indexterm"></a>

          <code class="literal">CHAR_LENGTH(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the length of the string
          <em class="replaceable"><code>str</code></em>, measured in characters. A
          multi-byte character counts as a single character. This means
          that for a string containing five two-byte characters,
          <code class="literal">LENGTH()</code> returns <code class="literal">10</code>,
          whereas <code class="literal">CHAR_LENGTH()</code> returns
          <code class="literal">5</code>.
        </p></li><li><p><a id="function_character-length"></a>
          <a id="id2820588" class="indexterm"></a>

          <code class="literal">CHARACTER_LENGTH(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          <code class="literal">CHARACTER_LENGTH()</code> is a synonym for
          <code class="literal">CHAR_LENGTH()</code>.
        </p></li><li><p><a id="function_concat"></a>
          <a id="id2820647" class="indexterm"></a>

          <code class="literal">CONCAT(<em class="replaceable"><code>str1</code></em>,<em class="replaceable"><code>str2</code></em>,...)</code>
        </p><p>
          Returns the string that results from concatenating the
          arguments. May have one or more arguments. If all arguments
          are non-binary strings, the result is a non-binary string. If
          the arguments include any binary strings, the result is a
          binary string. A numeric argument is converted to its
          equivalent binary string form; if you want to avoid that, you
          can use an explicit type cast, as in this example:
        </p><pre class="programlisting">SELECT CONCAT(CAST(<em class="replaceable"><code>int_col</code></em> AS CHAR), <em class="replaceable"><code>char_col</code></em>);
</pre><p>
          <code class="literal">CONCAT()</code> returns <code class="literal">NULL</code> if
          any argument is <code class="literal">NULL</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CONCAT('My', 'S', 'QL');</code></strong>
        -&gt; 'MySQL'
mysql&gt; <strong class="userinput"><code>SELECT CONCAT('My', NULL, 'QL');</code></strong>
        -&gt; NULL
mysql&gt; <strong class="userinput"><code>SELECT CONCAT(14.3);</code></strong>
        -&gt; '14.3'
</pre></li><li><p><a id="function_concat-ws"></a>
          <a id="id2820757" class="indexterm"></a>

          <code class="literal">CONCAT_WS(<em class="replaceable"><code>separator</code></em>,<em class="replaceable"><code>str1</code></em>,<em class="replaceable"><code>str2</code></em>,...)</code>
        </p><p>
          <code class="literal">CONCAT_WS()</code> stands for Concatenate With
          Separator and is a special form of
          <code class="literal">CONCAT()</code>. The first argument is the
          separator for the rest of the arguments. The separator is
          added between the strings to be concatenated. The separator
          can be a string, as can the rest of the arguments. If the
          separator is <code class="literal">NULL</code>, the result is
          <code class="literal">NULL</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CONCAT_WS(',','First name','Second name','Last Name');</code></strong>
        -&gt; 'First name,Second name,Last Name'
mysql&gt; <strong class="userinput"><code>SELECT CONCAT_WS(',','First name',NULL,'Last Name');</code></strong>
        -&gt; 'First name,Last Name'
</pre><p>
          <code class="literal">CONCAT_WS()</code> does not skip empty strings.
          However, it does skip any <code class="literal">NULL</code> values after
          the separator argument.
        </p></li><li><p><a id="function_conv"></a>
          <a id="id2820869" class="indexterm"></a>

          <code class="literal">CONV(<em class="replaceable"><code>N</code></em>,<em class="replaceable"><code>from_base</code></em>,<em class="replaceable"><code>to_base</code></em>)</code>
        </p><p>
          Converts numbers between different number bases. Returns a
          string representation of the number
          <em class="replaceable"><code>N</code></em>, converted from base
          <em class="replaceable"><code>from_base</code></em> to base
          <em class="replaceable"><code>to_base</code></em>. Returns
          <code class="literal">NULL</code> if any argument is
          <code class="literal">NULL</code>. The argument
          <em class="replaceable"><code>N</code></em> is interpreted as an integer, but
          may be specified as an integer or a string. The minimum base
          is <code class="literal">2</code> and the maximum base is
          <code class="literal">36</code>. If <em class="replaceable"><code>to_base</code></em>
          is a negative number, <em class="replaceable"><code>N</code></em> is regarded
          as a signed number. Otherwise, <em class="replaceable"><code>N</code></em> is
          treated as unsigned. <code class="literal">CONV()</code> works with
          64-bit precision.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CONV('a',16,2);</code></strong>
        -&gt; '1010'
mysql&gt; <strong class="userinput"><code>SELECT CONV('6E',18,8);</code></strong>
        -&gt; '172'
mysql&gt; <strong class="userinput"><code>SELECT CONV(-17,10,-18);</code></strong>
        -&gt; '-H'
mysql&gt; <strong class="userinput"><code>SELECT CONV(10+'10'+'10'+0xa,10,10);</code></strong>
        -&gt; '40'
</pre></li><li><p><a id="function_elt"></a>
          <a id="id2821005" class="indexterm"></a>

          <code class="literal">ELT(<em class="replaceable"><code>N</code></em>,<em class="replaceable"><code>str1</code></em>,<em class="replaceable"><code>str2</code></em>,<em class="replaceable"><code>str3</code></em>,...)</code>
        </p><p>
          Returns <em class="replaceable"><code>str1</code></em> if
          <em class="replaceable"><code>N</code></em> = <code class="literal">1</code>,
          <em class="replaceable"><code>str2</code></em> if
          <em class="replaceable"><code>N</code></em> = <code class="literal">2</code>, and so
          on. Returns <code class="literal">NULL</code> if
          <em class="replaceable"><code>N</code></em> is less than <code class="literal">1</code>
          or greater than the number of arguments.
          <code class="literal">ELT()</code> is the complement of
          <code class="literal">FIELD()</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ELT(1, 'ej', 'Heja', 'hej', 'foo');</code></strong>
        -&gt; 'ej'
mysql&gt; <strong class="userinput"><code>SELECT ELT(4, 'ej', 'Heja', 'hej', 'foo');</code></strong>
        -&gt; 'foo'
</pre></li><li><p><a id="function_export-set"></a>
          <a id="id2821126" class="indexterm"></a>

          <code class="literal">EXPORT_SET(<em class="replaceable"><code>bits</code></em>,<em class="replaceable"><code>on</code></em>,<em class="replaceable"><code>off</code></em>[,<em class="replaceable"><code>separator</code></em>[,<em class="replaceable"><code>number_of_bits</code></em>]])</code>
        </p><p>
          Returns a string such that for every bit set in the value
          <em class="replaceable"><code>bits</code></em>, you get an
          <em class="replaceable"><code>on</code></em> string and for every bit not set
          in the value, you get an <em class="replaceable"><code>off</code></em>
          string. Bits in <em class="replaceable"><code>bits</code></em> are examined
          from right to left (from low-order to high-order bits).
          Strings are added to the result from left to right, separated
          by the <em class="replaceable"><code>separator</code></em> string (the
          default being the comma character
          ‘<code class="literal">,</code>’). The number of bits
          examined is given by <em class="replaceable"><code>number_of_bits</code></em>
          (defaults to 64).
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT EXPORT_SET(5,'Y','N',',',4);</code></strong>
        -&gt; 'Y,N,Y,N'
mysql&gt; <strong class="userinput"><code>SELECT EXPORT_SET(6,'1','0',',',10);</code></strong>
        -&gt; '0,1,1,0,0,0,0,0,0,0'
</pre></li><li><p><a id="function_field"></a>
          <a id="id2821244" class="indexterm"></a>

          <code class="literal">FIELD(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>str1</code></em>,<em class="replaceable"><code>str2</code></em>,<em class="replaceable"><code>str3</code></em>,...)</code>
        </p><p>
          Returns the index (position) of <em class="replaceable"><code>str</code></em>
          in the <em class="replaceable"><code>str1</code></em>,
          <em class="replaceable"><code>str2</code></em>,
          <em class="replaceable"><code>str3</code></em>, <code class="literal">...</code> list.
          Returns <code class="literal">0</code> if <em class="replaceable"><code>str</code></em>
          is not found.
        </p><p>
          If all arguments to <code class="literal">FIELD()</code> are strings,
          all arguments are compared as strings. If all arguments are
          numbers, they are compared as numbers. Otherwise, the
          arguments are compared as double.
        </p><p>
          If <em class="replaceable"><code>str</code></em> is <code class="literal">NULL</code>,
          the return value is <code class="literal">0</code> because
          <code class="literal">NULL</code> fails equality comparison with any
          value. <code class="literal">FIELD()</code> is the complement of
          <code class="literal">ELT()</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT FIELD('ej', 'Hej', 'ej', 'Heja', 'hej', 'foo');</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT FIELD('fo', 'Hej', 'ej', 'Heja', 'hej', 'foo');</code></strong>
        -&gt; 0
</pre></li><li><p><a id="function_find-in-set"></a>
          <a id="id2821388" class="indexterm"></a>

          <code class="literal">FIND_IN_SET(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>strlist</code></em>)</code>
        </p><p>
          Returns a value in the range of 1 to
          <em class="replaceable"><code>N</code></em> if the string
          <em class="replaceable"><code>str</code></em> is in the string list
          <em class="replaceable"><code>strlist</code></em> consisting of
          <em class="replaceable"><code>N</code></em> substrings. A string list is a
          string composed of substrings separated by
          ‘<code class="literal">,</code>’ characters. If the first
          argument is a constant string and the second is a column of
          type <code class="literal">SET</code>, the
          <code class="literal">FIND_IN_SET()</code> function is optimized to use
          bit arithmetic. Returns <code class="literal">0</code> if
          <em class="replaceable"><code>str</code></em> is not in
          <em class="replaceable"><code>strlist</code></em> or if
          <em class="replaceable"><code>strlist</code></em> is the empty string.
          Returns <code class="literal">NULL</code> if either argument is
          <code class="literal">NULL</code>. This function does not work properly
          if the first argument contains a comma
          (‘<code class="literal">,</code>’) character.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT FIND_IN_SET('b','a,b,c,d');</code></strong>
        -&gt; 2
</pre></li><li><p><a id="function_format"></a>
          <a id="id2821518" class="indexterm"></a>

          <code class="literal">FORMAT(<em class="replaceable"><code>X</code></em>,<em class="replaceable"><code>D</code></em>)</code>
        </p><p>
          Formats the number <em class="replaceable"><code>X</code></em> to a format
          like <code class="literal">'#,###,###.##'</code>, rounded to
          <em class="replaceable"><code>D</code></em> decimal places, and returns the
          result as a string. If <em class="replaceable"><code>D</code></em> is
          <code class="literal">0</code>, the result has no decimal point or
          fractional part.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT FORMAT(12332.123456, 4);</code></strong>
        -&gt; '12,332.1235'
mysql&gt; <strong class="userinput"><code>SELECT FORMAT(12332.1,4);</code></strong>
        -&gt; '12,332.1000'
mysql&gt; <strong class="userinput"><code>SELECT FORMAT(12332.2,0);</code></strong>
        -&gt; '12,332'
</pre></li><li><p><a id="function_hex"></a>
          <a id="id2821615" class="indexterm"></a>

          <code class="literal">HEX(<em class="replaceable"><code>N_or_S</code></em>)</code>
        </p><p>
          If <em class="replaceable"><code>N_or_S</code></em> is a number, returns a
          string representation of the hexadecimal value of
          <em class="replaceable"><code>N</code></em>, where
          <em class="replaceable"><code>N</code></em> is a longlong
          (<code class="literal">BIGINT</code>) number. This is equivalent to
          <code class="literal">CONV(<em class="replaceable"><code>N</code></em>,10,16)</code>.
        </p><p>
          If <em class="replaceable"><code>N_or_S</code></em> is a string, returns a
          hexadecimal string representation of
          <em class="replaceable"><code>N_or_S</code></em> where each character in
          <em class="replaceable"><code>N_or_S</code></em> is converted to two
          hexadecimal digits.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT HEX(255);</code></strong>
        -&gt; 'FF'
mysql&gt; <strong class="userinput"><code>SELECT 0x616263;</code></strong>
        -&gt; 'abc'
mysql&gt; <strong class="userinput"><code>SELECT HEX('abc');</code></strong>
        -&gt; 616263
</pre></li><li><p><a id="function_insert"></a>
          <a id="id2821735" class="indexterm"></a>

          <code class="literal">INSERT(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>pos</code></em>,<em class="replaceable"><code>len</code></em>,<em class="replaceable"><code>newstr</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em>, with the
          substring beginning at position <em class="replaceable"><code>pos</code></em>
          and <em class="replaceable"><code>len</code></em> characters long replaced by
          the string <em class="replaceable"><code>newstr</code></em>. Returns the
          original string if <em class="replaceable"><code>pos</code></em> is not
          within the length of the string. Replaces the rest of the
          string from position <em class="replaceable"><code>pos</code></em> is
          <em class="replaceable"><code>len</code></em> is not within the length of the
          rest of the string. Returns <code class="literal">NULL</code> if any
          argument is <code class="literal">NULL</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT INSERT('Quadratic', 3, 4, 'What');</code></strong>
        -&gt; 'QuWhattic'
mysql&gt; <strong class="userinput"><code>SELECT INSERT('Quadratic', -1, 4, 'What');</code></strong>
        -&gt; 'Quadratic'
mysql&gt; <strong class="userinput"><code>SELECT INSERT('Quadratic', 3, 100, 'What');</code></strong>
        -&gt; 'QuWhat'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_instr"></a>
          <a id="id2821861" class="indexterm"></a>

          <code class="literal">INSTR(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>substr</code></em>)</code>
        </p><p>
          Returns the position of the first occurrence of substring
          <em class="replaceable"><code>substr</code></em> in string
          <em class="replaceable"><code>str</code></em>. This is the same as the
          two-argument form of <code class="literal">LOCATE()</code>, except that
          the order of the arguments is reversed.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT INSTR('foobarbar', 'bar');</code></strong>
        -&gt; 4
mysql&gt; <strong class="userinput"><code>SELECT INSTR('xbar', 'foobar');</code></strong>
        -&gt; 0
</pre><p>
          This function is multi-byte safe, and is case sensitive only
          if at least one argument is a binary string.
        </p></li><li><p><a id="function_lcase"></a>
          <a id="id2821952" class="indexterm"></a>

          <code class="literal">LCASE(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          <code class="literal">LCASE()</code> is a synonym for
          <code class="literal">LOWER()</code>.
        </p></li><li><p><a id="function_left"></a>
          <a id="id2822010" class="indexterm"></a>

          <code class="literal">LEFT(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>len</code></em>)</code>
        </p><p>
          Returns the leftmost <em class="replaceable"><code>len</code></em> characters
          from the string <em class="replaceable"><code>str</code></em>, or
          <code class="literal">NULL</code> if any argument is
          <code class="literal">NULL</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LEFT('foobarbar', 5);</code></strong>
        -&gt; 'fooba'
</pre></li><li><p><a id="function_length"></a>
          <a id="id2822093" class="indexterm"></a>

          <code class="literal">LENGTH(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the length of the string
          <em class="replaceable"><code>str</code></em>, measured in bytes. A
          multi-byte character counts as multiple bytes. This means that
          for a string containing five two-byte characters,
          <code class="literal">LENGTH()</code> returns <code class="literal">10</code>,
          whereas <code class="literal">CHAR_LENGTH()</code> returns
          <code class="literal">5</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LENGTH('text');</code></strong>
        -&gt; 4
</pre></li><li><p><a id="function_load-file"></a>
          <a id="id2822178" class="indexterm"></a>

          <a id="id2822184" class="indexterm"></a>

          <code class="literal">LOAD_FILE(<em class="replaceable"><code>file_name</code></em>)</code>
        </p><p>
          Reads the file and returns the file contents as a string. To
          use this function, the file must be located on the server
          host, you must specify the full pathname to the file, and you
          must have the <code class="literal">FILE</code> privilege. The file must
          be readable by all and its size less than
          <code class="literal">max_allowed_packet</code> bytes.
        </p><p>
          If the file does not exist or cannot be read because one of
          the preceding conditions is not satisfied, the function
          returns <code class="literal">NULL</code>.
        </p><p>
          As of MySQL 5.0.19, the
          <code class="literal">character_set_filesystem</code> system variable
          controls interpretation of filenames that are given as literal
          strings.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>UPDATE t</code></strong>
          <strong class="userinput"><code>  SET blob_col=LOAD_FILE('/tmp/picture')</code></strong>
          <strong class="userinput"><code>  WHERE id=1;</code></strong>
</pre></li><li><p><a id="function_locate"></a>
          <a id="id2822287" class="indexterm"></a>

          <code class="literal">LOCATE(<em class="replaceable"><code>substr</code></em>,<em class="replaceable"><code>str</code></em>)</code>,
          <code class="literal">LOCATE(<em class="replaceable"><code>substr</code></em>,<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>pos</code></em>)</code>
        </p><p>
          The first syntax returns the position of the first occurrence
          of substring <em class="replaceable"><code>substr</code></em> in string
          <em class="replaceable"><code>str</code></em>. The second syntax returns the
          position of the first occurrence of substring
          <em class="replaceable"><code>substr</code></em> in string
          <em class="replaceable"><code>str</code></em>, starting at position
          <em class="replaceable"><code>pos</code></em>. Returns <code class="literal">0</code>
          if <em class="replaceable"><code>substr</code></em> is not in
          <em class="replaceable"><code>str</code></em>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LOCATE('bar', 'foobarbar');</code></strong>
        -&gt; 4
mysql&gt; <strong class="userinput"><code>SELECT LOCATE('xbar', 'foobar');</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT LOCATE('bar', 'foobarbar', 5);</code></strong>
        -&gt; 7
</pre><p>
          This function is multi-byte safe, and is case-sensitive only
          if at least one argument is a binary string.
        </p></li><li><p><a id="function_lower"></a>
          <a id="id2822415" class="indexterm"></a>

          <code class="literal">LOWER(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em> with all
          characters changed to lowercase according to the current
          character set mapping. The default is
          <code class="literal">latin1</code> (cp1252 West European).
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LOWER('QUADRATICALLY');</code></strong>
        -&gt; 'quadratically'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_lpad"></a>
          <a id="id2822493" class="indexterm"></a>

          <code class="literal">LPAD(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>len</code></em>,<em class="replaceable"><code>padstr</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em>, left-padded
          with the string <em class="replaceable"><code>padstr</code></em> to a length
          of <em class="replaceable"><code>len</code></em> characters. If
          <em class="replaceable"><code>str</code></em> is longer than
          <em class="replaceable"><code>len</code></em>, the return value is shortened
          to <em class="replaceable"><code>len</code></em> characters.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LPAD('hi',4,'??');</code></strong>
        -&gt; '??hi'
mysql&gt; <strong class="userinput"><code>SELECT LPAD('hi',1,'??');</code></strong>
        -&gt; 'h'
</pre></li><li><p><a id="function_ltrim"></a>
          <a id="id2822592" class="indexterm"></a>

          <code class="literal">LTRIM(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em> with leading
          space characters removed.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LTRIM('  barbar');</code></strong>
        -&gt; 'barbar'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_make-set"></a>
          <a id="id2822664" class="indexterm"></a>

          <code class="literal">MAKE_SET(<em class="replaceable"><code>bits</code></em>,<em class="replaceable"><code>str1</code></em>,<em class="replaceable"><code>str2</code></em>,...)</code>
        </p><p>
          Returns a set value (a string containing substrings separated
          by ‘<code class="literal">,</code>’ characters) consisting
          of the strings that have the corresponding bit in
          <em class="replaceable"><code>bits</code></em> set.
          <em class="replaceable"><code>str1</code></em> corresponds to bit 0,
          <em class="replaceable"><code>str2</code></em> to bit 1, and so on.
          <code class="literal">NULL</code> values in
          <em class="replaceable"><code>str1</code></em>,
          <em class="replaceable"><code>str2</code></em>, <code class="literal">...</code> are
          not appended to the result.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MAKE_SET(1,'a','b','c');</code></strong>
        -&gt; 'a'
mysql&gt; <strong class="userinput"><code>SELECT MAKE_SET(1 | 4,'hello','nice','world');</code></strong>
        -&gt; 'hello,world'
mysql&gt; <strong class="userinput"><code>SELECT MAKE_SET(1 | 4,'hello','nice',NULL,'world');</code></strong>
        -&gt; 'hello'
mysql&gt; <strong class="userinput"><code>SELECT MAKE_SET(0,'a','b','c');</code></strong>
        -&gt; ''
</pre></li><li><p><a id="function_mid"></a>
          <a id="id2822785" class="indexterm"></a>

          <code class="literal">MID(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>pos</code></em>,<em class="replaceable"><code>len</code></em>)</code>
        </p><p>
          <code class="literal">MID(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>pos</code></em>,<em class="replaceable"><code>len</code></em>)</code>
          is a synonym for
          <code class="literal">SUBSTRING(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>pos</code></em>,<em class="replaceable"><code>len</code></em>)</code>.
        </p></li><li><p><a id="function_oct"></a>
          <a id="id2822869" class="indexterm"></a>

          <code class="literal">OCT(<em class="replaceable"><code>N</code></em>)</code>
        </p><p>
          Returns a string representation of the octal value of
          <em class="replaceable"><code>N</code></em>, where
          <em class="replaceable"><code>N</code></em> is a longlong
          (<code class="literal">BIGINT</code>) number. This is equivalent to
          <code class="literal">CONV(<em class="replaceable"><code>N</code></em>,10,8)</code>.
          Returns <code class="literal">NULL</code> if
          <em class="replaceable"><code>N</code></em> is <code class="literal">NULL</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT OCT(12);</code></strong>
        -&gt; '14'
</pre></li><li><p><a id="function_octet-length"></a>
          <a id="id2822964" class="indexterm"></a>

          <code class="literal">OCTET_LENGTH(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          <code class="literal">OCTET_LENGTH()</code> is a synonym for
          <code class="literal">LENGTH()</code>.
        </p></li><li><p><a id="function_ord"></a>
          <a id="id2823022" class="indexterm"></a>

          <code class="literal">ORD(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          If the leftmost character of the string
          <em class="replaceable"><code>str</code></em> is a multi-byte character,
          returns the code for that character, calculated from the
          numeric values of its constituent bytes using this formula:
        </p><pre class="programlisting">  (1st byte code)
+ (2nd byte code × 256)
+ (3rd byte code × 256<sup>2</sup>) ...
</pre><p>
          If the leftmost character is not a multi-byte character,
          <code class="literal">ORD()</code> returns the same value as the
          <code class="literal">ASCII()</code> function.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ORD('2');</code></strong>
        -&gt; 50
</pre></li><li><p><a id="function_position"></a>
          <a id="id2823114" class="indexterm"></a>

          <code class="literal">POSITION(<em class="replaceable"><code>substr</code></em> IN
          <em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          <code class="literal">POSITION(<em class="replaceable"><code>substr</code></em> IN
          <em class="replaceable"><code>str</code></em>)</code> is a synonym for
          <code class="literal">LOCATE(<em class="replaceable"><code>substr</code></em>,<em class="replaceable"><code>str</code></em>)</code>.
        </p></li><li><p><a id="function_quote"></a>
          <a id="id2823190" class="indexterm"></a>

          <code class="literal">QUOTE(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Quotes a string to produce a result that can be used as a
          properly escaped data value in an SQL statement. The string is
          returned enclosed by single quotes and with each instance of
          single quote (‘<code class="literal">'</code>’), backslash
          (‘<code class="literal">\</code>’), ASCII
          <code class="literal">NUL</code>, and Control-Z preceded by a backslash.
          If the argument is <code class="literal">NULL</code>, the return value
          is the word “<span class="quote">NULL</span>” without enclosing single
          quotes.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT QUOTE('Don\'t!');</code></strong>
        -&gt; 'Don\'t!'
mysql&gt; <strong class="userinput"><code>SELECT QUOTE(NULL);</code></strong>
        -&gt; NULL
</pre></li><li><p><a id="function_repeat"></a>
          <a id="id2823287" class="indexterm"></a>

          <code class="literal">REPEAT(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>count</code></em>)</code>
        </p><p>
          Returns a string consisting of the string
          <em class="replaceable"><code>str</code></em> repeated
          <em class="replaceable"><code>count</code></em> times. If
          <em class="replaceable"><code>count</code></em> is less than 1, returns an
          empty string. Returns <code class="literal">NULL</code> if
          <em class="replaceable"><code>str</code></em> or
          <em class="replaceable"><code>count</code></em> are <code class="literal">NULL</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT REPEAT('MySQL', 3);</code></strong>
        -&gt; 'MySQLMySQLMySQL'
</pre></li><li><p><a id="function_replace"></a>
          <a id="id2823383" class="indexterm"></a>

          <code class="literal">REPLACE(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>from_str</code></em>,<em class="replaceable"><code>to_str</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em> with all
          occurrences of the string <em class="replaceable"><code>from_str</code></em>
          replaced by the string <em class="replaceable"><code>to_str</code></em>.
          <code class="literal">REPLACE()</code> performs a case-sensitive match
          when searching for <em class="replaceable"><code>from_str</code></em>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT REPLACE('www.mysql.com', 'w', 'Ww');</code></strong>
        -&gt; 'WwWwWw.mysql.com'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_reverse"></a>
          <a id="id2823480" class="indexterm"></a>

          <code class="literal">REVERSE(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em> with the
          order of the characters reversed.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT REVERSE('abc');</code></strong>
        -&gt; 'cba'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_right"></a>
          <a id="id2823552" class="indexterm"></a>

          <code class="literal">RIGHT(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>len</code></em>)</code>
        </p><p>
          Returns the rightmost <em class="replaceable"><code>len</code></em>
          characters from the string <em class="replaceable"><code>str</code></em>, or
          <code class="literal">NULL</code> if any argument is
          <code class="literal">NULL</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT RIGHT('foobarbar', 4);</code></strong>
        -&gt; 'rbar'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_rpad"></a>
          <a id="id2823638" class="indexterm"></a>

          <code class="literal">RPAD(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>len</code></em>,<em class="replaceable"><code>padstr</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em>,
          right-padded with the string <em class="replaceable"><code>padstr</code></em>
          to a length of <em class="replaceable"><code>len</code></em> characters. If
          <em class="replaceable"><code>str</code></em> is longer than
          <em class="replaceable"><code>len</code></em>, the return value is shortened
          to <em class="replaceable"><code>len</code></em> characters.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT RPAD('hi',5,'?');</code></strong>
        -&gt; 'hi???'
mysql&gt; <strong class="userinput"><code>SELECT RPAD('hi',1,'?');</code></strong>
        -&gt; 'h'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_rtrim"></a>
          <a id="id2823741" class="indexterm"></a>

          <code class="literal">RTRIM(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em> with
          trailing space characters removed.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT RTRIM('barbar   ');</code></strong>
        -&gt; 'barbar'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_soundex"></a>
          <a id="id2823813" class="indexterm"></a>

          <code class="literal">SOUNDEX(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns a soundex string from <em class="replaceable"><code>str</code></em>.
          Two strings that sound almost the same should have identical
          soundex strings. A standard soundex string is four characters
          long, but the <code class="literal">SOUNDEX()</code> function returns an
          arbitrarily long string. You can use
          <code class="literal">SUBSTRING()</code> on the result to get a standard
          soundex string. All non-alphabetic characters in
          <em class="replaceable"><code>str</code></em> are ignored. All international
          alphabetic characters outside the A-Z range are treated as
          vowels.
        </p><p>
          <span class="bold"><strong>Important</strong></span>: When using
          <code class="literal">SOUNDEX()</code>, you should be aware of the
          following limitations:
        </p><div class="itemizedlist"><ul type="circle"><li><p>
              This function, as currently implemented, is intended to
              work well with strings that are in the English language
              only. Strings in other languages may not produce reliable
              results.
            </p></li><li><p>
              This function is not guaranteed to provide consistent
              results with strings that use multi-byte character sets,
              including <code class="literal">utf-8</code>.
            </p><p>
              We hope to remove these limitations in a future release.
              See <a href="http://bugs.mysql.com/22638" target="_top">Bug#22638</a> for more information.
            </p></li></ul></div><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SOUNDEX('Hello');</code></strong>
        -&gt; 'H400'
mysql&gt; <strong class="userinput"><code>SELECT SOUNDEX('Quadratically');</code></strong>
        -&gt; 'Q36324'
</pre><p>
          <span class="bold"><strong>Note</strong></span>: This function
          implements the original Soundex algorithm, not the more
          popular enhanced version (also described by D. Knuth). The
          difference is that original version discards vowels first and
          duplicates second, whereas the enhanced version discards
          duplicates first and vowels second.
        </p></li><li><p><a id="operator_sounds-like"></a>
          <a id="id2823978" class="indexterm"></a>

          <code class="literal"><em class="replaceable"><code>expr1</code></em> SOUNDS LIKE
          <em class="replaceable"><code>expr2</code></em></code>
        </p><p>
          This is the same as
          <code class="literal">SOUNDEX(<em class="replaceable"><code>expr1</code></em>) =
          SOUNDEX(<em class="replaceable"><code>expr2</code></em>)</code>.
        </p></li><li><p><a id="function_space"></a>
          <a id="id2824042" class="indexterm"></a>

          <code class="literal">SPACE(<em class="replaceable"><code>N</code></em>)</code>
        </p><p>
          Returns a string consisting of <em class="replaceable"><code>N</code></em>
          space characters.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SPACE(6);</code></strong>
        -&gt; '      '
</pre></li><li><p><a id="function_substring"></a>
          <a id="id2824110" class="indexterm"></a>

          <a id="id2824116" class="indexterm"></a>

          <code class="literal">SUBSTRING(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>pos</code></em>)</code>,
          <code class="literal">SUBSTRING(<em class="replaceable"><code>str</code></em> FROM
          <em class="replaceable"><code>pos</code></em>)</code>,
          <code class="literal">SUBSTRING(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>pos</code></em>,<em class="replaceable"><code>len</code></em>)</code>,
          <code class="literal">SUBSTRING(<em class="replaceable"><code>str</code></em> FROM
          <em class="replaceable"><code>pos</code></em> FOR
          <em class="replaceable"><code>len</code></em>)</code>
        </p><p>
          The forms without a <em class="replaceable"><code>len</code></em> argument
          return a substring from string <em class="replaceable"><code>str</code></em>
          starting at position <em class="replaceable"><code>pos</code></em>. The forms
          with a <em class="replaceable"><code>len</code></em> argument return a
          substring <em class="replaceable"><code>len</code></em> characters long from
          string <em class="replaceable"><code>str</code></em>, starting at position
          <em class="replaceable"><code>pos</code></em>. The forms that use
          <code class="literal">FROM</code> are standard SQL syntax. It is also
          possible to use a negative value for
          <em class="replaceable"><code>pos</code></em>. In this case, the beginning of
          the substring is <em class="replaceable"><code>pos</code></em> characters
          from the end of the string, rather than the beginning. A
          negative value may be used for <em class="replaceable"><code>pos</code></em>
          in any of the forms of this function.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SUBSTRING('Quadratically',5);</code></strong>
        -&gt; 'ratically'
mysql&gt; <strong class="userinput"><code>SELECT SUBSTRING('foobarbar' FROM 4);</code></strong>
        -&gt; 'barbar'
mysql&gt; <strong class="userinput"><code>SELECT SUBSTRING('Quadratically',5,6);</code></strong>
        -&gt; 'ratica'        
mysql&gt; <strong class="userinput"><code>SELECT SUBSTRING('Sakila', -3);</code></strong>
        -&gt; 'ila'        
mysql&gt; <strong class="userinput"><code>SELECT SUBSTRING('Sakila', -5, 3);</code></strong>
        -&gt; 'aki'
mysql&gt; <strong class="userinput"><code>SELECT SUBSTRING('Sakila' FROM -4 FOR 2);</code></strong>
        -&gt; 'ki'
</pre><p>
          This function is multi-byte safe.
        </p><p>
          If <em class="replaceable"><code>len</code></em> is less than 1, the result
          is the empty string.
        </p><p>
          <code class="literal">SUBSTR()</code> is a synonym for
          <code class="literal">SUBSTRING()</code>.
        </p></li><li><p><a id="function_substring-index"></a>
          <a id="id2824319" class="indexterm"></a>

          <code class="literal">SUBSTRING_INDEX(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>delim</code></em>,<em class="replaceable"><code>count</code></em>)</code>
        </p><p>
          Returns the substring from string
          <em class="replaceable"><code>str</code></em> before
          <em class="replaceable"><code>count</code></em> occurrences of the delimiter
          <em class="replaceable"><code>delim</code></em>. If
          <em class="replaceable"><code>count</code></em> is positive, everything to
          the left of the final delimiter (counting from the left) is
          returned. If <em class="replaceable"><code>count</code></em> is negative,
          everything to the right of the final delimiter (counting from
          the right) is returned. <code class="literal">SUBSTRING_INDEX()</code>
          performs a case-sensitive match when searching for
          <em class="replaceable"><code>delim</code></em>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SUBSTRING_INDEX('www.mysql.com', '.', 2);</code></strong>
        -&gt; 'www.mysql'
mysql&gt; <strong class="userinput"><code>SELECT SUBSTRING_INDEX('www.mysql.com', '.', -2);</code></strong>
        -&gt; 'mysql.com'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_trim"></a>
          <a id="id2824439" class="indexterm"></a>

          <code class="literal">TRIM([{BOTH | LEADING | TRAILING}
          [<em class="replaceable"><code>remstr</code></em>] FROM]
          <em class="replaceable"><code>str</code></em>)</code>,
          <code class="literal">TRIM([<em class="replaceable"><code>remstr</code></em> FROM]
          <em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em> with all
          <em class="replaceable"><code>remstr</code></em> prefixes or suffixes
          removed. If none of the specifiers <code class="literal">BOTH</code>,
          <code class="literal">LEADING</code>, or <code class="literal">TRAILING</code> is
          given, <code class="literal">BOTH</code> is assumed.
          <em class="replaceable"><code>remstr</code></em> is optional and, if not
          specified, spaces are removed.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TRIM('  bar   ');</code></strong>
        -&gt; 'bar'
mysql&gt; <strong class="userinput"><code>SELECT TRIM(LEADING 'x' FROM 'xxxbarxxx');</code></strong>
        -&gt; 'barxxx'
mysql&gt; <strong class="userinput"><code>SELECT TRIM(BOTH 'x' FROM 'xxxbarxxx');</code></strong>
        -&gt; 'bar'
mysql&gt; <strong class="userinput"><code>SELECT TRIM(TRAILING 'xyz' FROM 'barxxyz');</code></strong>
        -&gt; 'barx'
</pre><p>
          This function is multi-byte safe.
        </p></li><li><p><a id="function_ucase"></a>
          <a id="id2824566" class="indexterm"></a>

          <code class="literal">UCASE(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          <code class="literal">UCASE()</code> is a synonym for
          <code class="literal">UPPER()</code>.
        </p></li><li><p><a id="function_unhex"></a>
          <a id="id2824634" class="indexterm"></a>

          <code class="literal">UNHEX(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Performs the inverse operation of
          <code class="literal">HEX(<em class="replaceable"><code>str</code></em>)</code>. That
          is, it interprets each pair of hexadecimal digits in the
          argument as a number and converts it to the character
          represented by the number. The resulting characters are
          returned as a binary string.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UNHEX('4D7953514C');</code></strong>
        -&gt; 'MySQL'
mysql&gt; <strong class="userinput"><code>SELECT 0x4D7953514C;</code></strong>
        -&gt; 'MySQL'
mysql&gt; <strong class="userinput"><code>SELECT UNHEX(HEX('string'));</code></strong>
        -&gt; 'string'
mysql&gt; <strong class="userinput"><code>SELECT HEX(UNHEX('1267'));</code></strong>
        -&gt; '1267'
</pre><p>
          The characters in the argument string must be legal
          hexadecimal digits: <code class="literal">'0'</code> ..
          <code class="literal">'9'</code>, <code class="literal">'A'</code> ..
          <code class="literal">'F'</code>, <code class="literal">'a'</code> ..
          <code class="literal">'f'</code>. If <code class="literal">UNHEX()</code>
          encounters any non-hexadecimal digits in the argument, it
          returns <code class="literal">NULL</code>:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UNHEX('GG');</code></strong>
+-------------+
| UNHEX('GG') |
+-------------+
| NULL        | 
+-------------+
</pre><p>
          A <code class="literal">NULL</code> result can occur if the argument to
          <code class="literal">UNHEX()</code> is a <code class="literal">BINARY</code>
          column, because values are padded with 0x00 bytes when stored
          but those bytes are not stripped on retrieval. For example
          <code class="literal">'aa'</code> is stored into a
          <code class="literal">CHAR(3)</code> column as
          <code class="literal">'aa '</code> and retrieved as
          <code class="literal">'aa'</code> (with the trailing pad space
          stripped), so <code class="literal">UNHEX()</code> for the column value
          returns <code class="literal">'A'</code>. By contrast
          <code class="literal">'aa'</code> is stored into a
          <code class="literal">BINARY(3)</code> column as
          <code class="literal">'aa\0'</code> and retrieved as
          <code class="literal">'aa\0'</code> (with the trailing pad
          <code class="literal">0x00</code> byte not stripped).
          <code class="literal">'\0'</code> is not a legal hexadecimal digit, so
          <code class="literal">UNHEX()</code> for the column value returns
          <code class="literal">NULL</code>.
        </p></li><li><p><a id="function_upper"></a>
          <a id="id2824834" class="indexterm"></a>

          <code class="literal">UPPER(<em class="replaceable"><code>str</code></em>)</code>
        </p><p>
          Returns the string <em class="replaceable"><code>str</code></em> with all
          characters changed to uppercase according to the current
          character set mapping. The default is
          <code class="literal">latin1</code> (cp1252 West European).
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UPPER('Hej');</code></strong>
        -&gt; 'HEJ'
</pre><p>
          This function is multi-byte safe.
        </p></li></ul></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="string-comparison-functions"></a>String Comparison Functions</h3></div></div></div><a id="id2824903" class="indexterm"></a><a id="id2824910" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#operator_equal-to"><code class="literal">&lt;=&gt;</code></a></td><td>NULL-safe equal to operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_equal"><code class="literal">=</code></a></td><td>Equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than-or-equal"><code class="literal">&gt;=</code></a></td><td>Greater than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_greater-than"><code class="literal">&gt;</code></a></td><td>Greater than operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-null"><code class="literal">IS NULL</code></a></td><td>NULL value test</td></tr><tr><td><a href="mysqlqb_functions.html#operator_is"><code class="literal">IS</code></a></td><td>Test a value against a boolean</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than-or-equal"><code class="literal">&lt;=</code></a></td><td>Less than or equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_less-than"><code class="literal">&lt;</code></a></td><td>Less than operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_like"><code class="literal">LIKE</code></a></td><td>Simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-equal"><code class="literal">!=</code>, <code class="literal">&lt;&gt;</code></a></td><td>Not equal operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_not-like"><code class="literal">NOT LIKE</code></a></td><td>Negation of simple pattern matching</td></tr><tr><td><a href="mysqlqb_functions.html#operator_sounds-like"><code class="literal">SOUNDS LIKE</code></a></td><td>Compare sounds</td></tr></tbody></table></div><p>
        If a string function is given a binary string as an argument,
        the resulting string is also a binary string. A number converted
        to a string is treated as a binary string. This affects only
        comparisons.
      </p><a id="id2825106" class="indexterm"></a><a id="id2825116" class="indexterm"></a><p>
        Normally, if any expression in a string comparison is case
        sensitive, the comparison is performed in case-sensitive
        fashion.
      </p><div class="itemizedlist"><ul type="disc"><li><p><a id="operator_like"></a>
            <a id="id2825159" class="indexterm"></a>

            <code class="literal"><em class="replaceable"><code>expr</code></em> LIKE
            <em class="replaceable"><code>pat</code></em> [ESCAPE
            '<em class="replaceable"><code>escape_char</code></em>']</code>
          </p><p>
            Pattern matching using SQL simple regular expression
            comparison. Returns <code class="literal">1</code>
            (<code class="literal">TRUE</code>) or <code class="literal">0</code>
            (<code class="literal">FALSE</code>). If either
            <em class="replaceable"><code>expr</code></em> or
            <em class="replaceable"><code>pat</code></em> is <code class="literal">NULL</code>,
            the result is <code class="literal">NULL</code>.
          </p><p>
            The pattern need not be a literal string. For example, it
            can be specified as a string expression or table column.
          </p><p>
            Per the SQL standard, <code class="literal">LIKE</code> performs
            matching on a per-character basis, thus it can produce
            results different from the <code class="literal">=</code> comparison
            operator:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'ä' LIKE 'ae' COLLATE latin1_german2_ci;</code></strong>
+-----------------------------------------+
| 'ä' LIKE 'ae' COLLATE latin1_german2_ci |
+-----------------------------------------+
|                                       0 |
+-----------------------------------------+
mysql&gt; <strong class="userinput"><code>SELECT 'ä' = 'ae' COLLATE latin1_german2_ci;</code></strong>
+--------------------------------------+
| 'ä' = 'ae' COLLATE latin1_german2_ci |
+--------------------------------------+
|                                    1 |
+--------------------------------------+
</pre><p>
            With <code class="literal">LIKE</code> you can use the following two
            wildcard characters in the pattern:
          </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>Character</strong></span></td><td><span class="bold"><strong>Description</strong></span></td></tr><tr><td><code class="literal">%</code></td><td>Matches any number of characters, even zero characters</td></tr><tr><td><code class="literal">_</code></td><td>Matches exactly one character</td></tr></tbody></table></div><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'David!' LIKE 'David_';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'David!' LIKE '%D%v%';</code></strong>
        -&gt; 1
</pre><p>
            To test for literal instances of a wildcard character,
            precede it by the escape character. If you do not specify
            the <code class="literal">ESCAPE</code> character,
            ‘<code class="literal">\</code>’ is assumed.
          </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>String</strong></span></td><td><span class="bold"><strong>Description</strong></span></td></tr><tr><td><code class="literal">\%</code></td><td>Matches one ‘<code class="literal">%</code>’ character</td></tr><tr><td><code class="literal">\_</code></td><td>Matches one ‘<code class="literal">_</code>’ character</td></tr></tbody></table></div><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'David!' LIKE 'David\_';</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'David_' LIKE 'David\_';</code></strong>
        -&gt; 1
</pre><p>
            To specify a different escape character, use the
            <code class="literal">ESCAPE</code> clause:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'David_' LIKE 'David|_' ESCAPE '|';</code></strong>
        -&gt; 1
</pre><p>
            The escape sequence should be empty or one character long.
            As of MySQL 5.0.16, if the
            <code class="literal">NO_BACKSLASH_ESCAPES</code> SQL mode is enabled,
            the sequence cannot be empty.
          </p><p>
            The following two statements illustrate that string
            comparisons are not case sensitive unless one of the
            operands is a binary string:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'abc' LIKE 'ABC';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'abc' LIKE BINARY 'ABC';</code></strong>
        -&gt; 0
</pre><p>
            In MySQL, <code class="literal">LIKE</code> is allowed on numeric
            expressions. (This is an extension to the standard SQL
            <code class="literal">LIKE</code>.)
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 10 LIKE '1%';</code></strong>
        -&gt; 1
</pre><p>
            <span class="bold"><strong>Note</strong></span>: Because MySQL uses C
            escape syntax in strings (for example,
            ‘<code class="literal">\n</code>’ to represent a newline
            character), you must double any
            ‘<code class="literal">\</code>’ that you use in
            <code class="literal">LIKE</code> strings. For example, to search for
            ‘<code class="literal">\n</code>’, specify it as
            ‘<code class="literal">\\n</code>’. To search for
            ‘<code class="literal">\</code>’, specify it as
            ‘<code class="literal">\\\\</code>’; this is because the
            backslashes are stripped once by the parser and again when
            the pattern match is made, leaving a single backslash to be
            matched against. (Exception: At the end of the pattern
            string, backslash can be specified as
            ‘<code class="literal">\\</code>’. At the end of the
            string, backslash stands for itself because there is nothing
            following to escape.)
          </p></li><li><p><a id="operator_not-like"></a>
            <a id="id2825593" class="indexterm"></a>

            <code class="literal"><em class="replaceable"><code>expr</code></em> NOT LIKE
            <em class="replaceable"><code>pat</code></em> [ESCAPE
            '<em class="replaceable"><code>escape_char</code></em>']</code>
          </p><p>
            This is the same as <code class="literal">NOT
            (<em class="replaceable"><code>expr</code></em> LIKE
            <em class="replaceable"><code>pat</code></em> [ESCAPE
            '<em class="replaceable"><code>escape_char</code></em>'])</code>.
          </p><div class="note" style="margin-left: 0.5in; margin-right: 0.5in;"><h3 class="title">Note</h3><p>
              Aggegate queries involving <code class="literal">NOT LIKE</code>
              comparisons with columns containing
              <code class="literal">NULL</code> may yield unexpected results. For
              example, consider the following table and data:

</p><pre class="programlisting">CREATE TABLE foo (bar VARCHAR(10));

INSERT INTO foo VALUES (NULL), (NULL);
</pre><p>

              The query <code class="literal">SELECT COUNT(*) FROM foo WHERE bar LIKE
              '%baz%';</code> returns <code class="literal">0</code>. You might
              assume that <code class="literal">SELECT COUNT(*) FROM foo WHERE bar
              NOT LIKE '%baz%';</code> would return
              <code class="literal">2</code>. However, this is not the case: The
              second query returns <code class="literal">0</code>. This is because
              <code class="literal">NULL NOT LIKE
              <em class="replaceable"><code>expr</code></em></code> always returns
              <code class="literal">NULL</code>, regardless of the value of
              <em class="replaceable"><code>expr</code></em>. The same is true for
              aggregate queries involving <code class="literal">NULL</code> and
              comparisons using <code class="literal">NOT RLIKE</code> or
              <code class="literal">NOT REGEXP</code>. In such cases, you must
              test explicitly for <code class="literal">NOT NULL</code> using
              <code class="literal">OR</code> (and not <code class="literal">AND</code>), as
              shown here:

</p><pre class="programlisting">SELECT COUNT(*) FROM foo WHERE bar NOT LIKE '%baz%' OR bar IS NULL;
</pre><p>
            </p></div></li><li><p><a id="operator_not-regexp"></a>
            <a id="id2825755" class="indexterm"></a>

            <code class="literal"><em class="replaceable"><code>expr</code></em> NOT REGEXP
            <em class="replaceable"><code>pat</code></em></code>,
            <code class="literal"><em class="replaceable"><code>expr</code></em> NOT RLIKE
            <em class="replaceable"><code>pat</code></em></code>
          </p><p>
            This is the same as <code class="literal">NOT
            (<em class="replaceable"><code>expr</code></em> REGEXP
            <em class="replaceable"><code>pat</code></em>)</code>.
          </p></li><li><p><a id="operator_regexp"></a>
            <a id="id2825834" class="indexterm"></a>

            <a id="id2825840" class="indexterm"></a>

            <a id="id2825851" class="indexterm"></a>

            <a id="id2825857" class="indexterm"></a>

            <code class="literal"><em class="replaceable"><code>expr</code></em> REGEXP
            <em class="replaceable"><code>pat</code></em></code>,
            <code class="literal"><em class="replaceable"><code>expr</code></em> RLIKE
            <em class="replaceable"><code>pat</code></em></code>
          </p><p>
            Performs a pattern match of a string expression
            <em class="replaceable"><code>expr</code></em> against a pattern
            <em class="replaceable"><code>pat</code></em>. The pattern can be an
            extended regular expression. The syntax for regular
            expressions is discussed in <a href="mysqlqb_functions.html#regexp" title="Regular Expressions">Regular Expressions</a>.
            Returns <code class="literal">1</code> if
            <em class="replaceable"><code>expr</code></em> matches
            <em class="replaceable"><code>pat</code></em>; otherwise it returns
            <code class="literal">0</code>. If either
            <em class="replaceable"><code>expr</code></em> or
            <em class="replaceable"><code>pat</code></em> is <code class="literal">NULL</code>,
            the result is <code class="literal">NULL</code>.
            <code class="literal">RLIKE</code> is a synonym for
            <code class="literal">REGEXP</code>, provided for
            <code class="literal">mSQL</code> compatibility.
          </p><p>
            The pattern need not be a literal string. For example, it
            can be specified as a string expression or table column.
          </p><p>
            <span class="bold"><strong>Note</strong></span>: Because MySQL uses
            the C escape syntax in strings (for example,
            ‘<code class="literal">\n</code>’ to represent the newline
            character), you must double any
            ‘<code class="literal">\</code>’ that you use in your
            <code class="literal">REGEXP</code> strings.
          </p><p>
            <code class="literal">REGEXP</code> is not case sensitive, except when
            used with binary strings.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'Monty!' REGEXP 'm%y%%';</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'Monty!' REGEXP '.*';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'new*\n*line' REGEXP 'new\\*.\\*line';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'a' REGEXP 'A', 'a' REGEXP BINARY 'A';</code></strong>
        -&gt; 1  0
mysql&gt; <strong class="userinput"><code>SELECT 'a' REGEXP '^[a-d]';</code></strong>
        -&gt; 1
</pre><p>
            <code class="literal">REGEXP</code> and <code class="literal">RLIKE</code> use
            the current character set when deciding the type of a
            character. The default is <code class="literal">latin1</code> (cp1252
            West European). <span class="bold"><strong>Warning</strong></span>:
            These operators are not multi-byte safe.
          </p></li><li><p><a id="function_strcmp"></a>
            <a id="id2826070" class="indexterm"></a>

            <code class="literal">STRCMP(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>)</code>
          </p><p>
            <code class="literal">STRCMP()</code> returns <code class="literal">0</code> if
            the strings are the same, <code class="literal">-1</code> if the first
            argument is smaller than the second according to the current
            sort order, and <code class="literal">1</code> otherwise.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT STRCMP('text', 'text2');</code></strong>
        -&gt; -1
mysql&gt; <strong class="userinput"><code>SELECT STRCMP('text2', 'text');</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT STRCMP('text', 'text');</code></strong>
        -&gt; 0
</pre><p>
            <code class="literal">STRCMP()</code> uses the current character set
            when performing comparisons. This makes the default
            comparison behavior case insensitive unless one or both of
            the operands are binary strings.
          </p></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="regexp"></a>Regular Expressions</h3></div></div></div><a id="id2826165" class="indexterm"></a><a id="id2826172" class="indexterm"></a><a id="id2826178" class="indexterm"></a><a id="id2826185" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#operator_not-regexp"><code class="literal">NOT RGEXP</code></a></td><td>Negation of REGEXP</td></tr><tr><td><a href="mysqlqb_functions.html#operator_regexp"><code class="literal">REGEXP</code></a></td><td>Pattern matching using regular expressions</td></tr><tr><td><a href="mysqlqb_functions.html#operator_regexp"><code class="literal">RLIKE</code></a></td><td>Synonym for REGEXP</td></tr></tbody></table></div><p>
        A regular expression is a powerful way of specifying a pattern
        for a complex search.
      </p><p>
        MySQL uses Henry Spencer's implementation of regular
        expressions, which is aimed at conformance with POSIX 1003.2.
        See <a href="http://dev.mysql.com/doc/refman/5.1/en/credits.html" target="_top">Credits</a>. MySQL uses the extended version
        to support pattern-matching operations performed with the
        <code class="literal">REGEXP</code> operator in SQL statements. See
        <a href="http://dev.mysql.com/doc/refman/5.0/en/pattern-matching.html" target="_top">Pattern Matching</a>, and
        <a href="mysqlqb_functions.html#string-comparison-functions" title="String Comparison Functions">String Comparison Functions</a>.
      </p><p>
        This section is a summary, with examples, of the special
        characters and constructs that can be used in MySQL for
        <code class="literal">REGEXP</code> operations. It does not contain all
        the details that can be found in Henry Spencer's
        <code class="literal">regex(7)</code> manual page. That manual page is
        included in MySQL source distributions, in the
        <code class="filename">regex.7</code> file under the
        <code class="filename">regex</code> directory.
      </p><p>
        A regular expression describes a set of strings. The simplest
        regular expression is one that has no special characters in it.
        For example, the regular expression <code class="literal">hello</code>
        matches <code class="literal">hello</code> and nothing else.
      </p><p>
        Non-trivial regular expressions use certain special constructs
        so that they can match more than one string. For example, the
        regular expression <code class="literal">hello|word</code> matches either
        the string <code class="literal">hello</code> or the string
        <code class="literal">word</code>.
      </p><p>
        As a more complex example, the regular expression
        <code class="literal">B[an]*s</code> matches any of the strings
        <code class="literal">Bananas</code>, <code class="literal">Baaaaas</code>,
        <code class="literal">Bs</code>, and any other string starting with a
        <code class="literal">B</code>, ending with an <code class="literal">s</code>, and
        containing any number of <code class="literal">a</code> or
        <code class="literal">n</code> characters in between.
      </p><p>
        A regular expression for the <code class="literal">REGEXP</code> operator
        may use any of the following special characters and constructs:
      </p><div class="itemizedlist"><ul type="disc"><li><p>
            <code class="literal">^</code>
          </p><p>
            Match the beginning of a string.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'fo\nfo' REGEXP '^fo$';</code></strong>                   -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'fofo' REGEXP '^fo';</code></strong>                      -&gt; 1
</pre></li><li><p>
            <code class="literal">$</code>
          </p><p>
            Match the end of a string.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'fo\no' REGEXP '^fo\no$';</code></strong>                 -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'fo\no' REGEXP '^fo$';</code></strong>                    -&gt; 0
</pre></li><li><p>
            <code class="literal">.</code>
          </p><p>
            Match any character (including carriage return and newline).
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'fofo' REGEXP '^f.*$';</code></strong>                    -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'fo\r\nfo' REGEXP '^f.*$';</code></strong>                -&gt; 1
</pre></li><li><p>
            <code class="literal">a*</code>
          </p><p>
            Match any sequence of zero or more <code class="literal">a</code>
            characters.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'Ban' REGEXP '^Ba*n';</code></strong>                     -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'Baaan' REGEXP '^Ba*n';</code></strong>                   -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'Bn' REGEXP '^Ba*n';</code></strong>                      -&gt; 1
</pre></li><li><p>
            <code class="literal">a+</code>
          </p><p>
            Match any sequence of one or more <code class="literal">a</code>
            characters.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'Ban' REGEXP '^Ba+n';</code></strong>                     -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'Bn' REGEXP '^Ba+n';</code></strong>                      -&gt; 0
</pre></li><li><p>
            <code class="literal">a?</code>
          </p><p>
            Match either zero or one <code class="literal">a</code> character.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'Bn' REGEXP '^Ba?n';</code></strong>                      -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'Ban' REGEXP '^Ba?n';</code></strong>                     -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'Baan' REGEXP '^Ba?n';</code></strong>                    -&gt; 0
</pre></li><li><p>
            <code class="literal">de|abc</code>
          </p><p>
            Match either of the sequences <code class="literal">de</code> or
            <code class="literal">abc</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'pi' REGEXP 'pi|apa';</code></strong>                     -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'axe' REGEXP 'pi|apa';</code></strong>                    -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'apa' REGEXP 'pi|apa';</code></strong>                    -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'apa' REGEXP '^(pi|apa)$';</code></strong>                -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'pi' REGEXP '^(pi|apa)$';</code></strong>                 -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'pix' REGEXP '^(pi|apa)$';</code></strong>                -&gt; 0
</pre></li><li><p>
            <code class="literal">(abc)*</code>
          </p><p>
            Match zero or more instances of the sequence
            <code class="literal">abc</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'pi' REGEXP '^(pi)*$';</code></strong>                    -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'pip' REGEXP '^(pi)*$';</code></strong>                   -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'pipi' REGEXP '^(pi)*$';</code></strong>                  -&gt; 1
</pre></li><li><p>
            <code class="literal">{1}</code>, <code class="literal">{2,3}</code>
          </p><p>
            <code class="literal">{n}</code> or <code class="literal">{m,n}</code> notation
            provides a more general way of writing regular expressions
            that match many occurrences of the previous atom (or
            “<span class="quote">piece</span>”) of the pattern. <code class="literal">m</code>
            and <code class="literal">n</code> are integers.
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                <code class="literal">a*</code>
              </p><p>
                Can be written as <code class="literal">a{0,}</code>.
              </p></li><li><p>
                <code class="literal">a+</code>
              </p><p>
                Can be written as <code class="literal">a{1,}</code>.
              </p></li><li><p>
                <code class="literal">a?</code>
              </p><p>
                Can be written as <code class="literal">a{0,1}</code>.
              </p></li></ul></div><p>
            To be more precise, <code class="literal">a{n}</code> matches exactly
            <code class="literal">n</code> instances of <code class="literal">a</code>.
            <code class="literal">a{n,}</code> matches <code class="literal">n</code> or
            more instances of <code class="literal">a</code>.
            <code class="literal">a{m,n}</code> matches <code class="literal">m</code>
            through <code class="literal">n</code> instances of
            <code class="literal">a</code>, inclusive.
          </p><p>
            <code class="literal">m</code> and <code class="literal">n</code> must be in the
            range from <code class="literal">0</code> to
            <code class="literal">RE_DUP_MAX</code> (default 255), inclusive. If
            both <code class="literal">m</code> and <code class="literal">n</code> are
            given, <code class="literal">m</code> must be less than or equal to
            <code class="literal">n</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'abcde' REGEXP 'a[bcd]{2}e';</code></strong>              -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'abcde' REGEXP 'a[bcd]{3}e';</code></strong>              -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'abcde' REGEXP 'a[bcd]{1,10}e';</code></strong>           -&gt; 1
</pre></li><li><p>
            <code class="literal">[a-dX]</code>, <code class="literal">[^a-dX]</code>
          </p><p>
            Matches any character that is (or is not, if ^ is used)
            either <code class="literal">a</code>, <code class="literal">b</code>,
            <code class="literal">c</code>, <code class="literal">d</code> or
            <code class="literal">X</code>. A <code class="literal">-</code> character
            between two other characters forms a range that matches all
            characters from the first character to the second. For
            example, <code class="literal">[0-9]</code> matches any decimal digit.
            To include a literal <code class="literal">]</code> character, it must
            immediately follow the opening bracket <code class="literal">[</code>.
            To include a literal <code class="literal">-</code> character, it must
            be written first or last. Any character that does not have a
            defined special meaning inside a <code class="literal">[]</code> pair
            matches only itself.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'aXbc' REGEXP '[a-dXYZ]';</code></strong>                 -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'aXbc' REGEXP '^[a-dXYZ]$';</code></strong>               -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'aXbc' REGEXP '^[a-dXYZ]+$';</code></strong>              -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'aXbc' REGEXP '^[^a-dXYZ]+$';</code></strong>             -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'gheis' REGEXP '^[^a-dXYZ]+$';</code></strong>            -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'gheisa' REGEXP '^[^a-dXYZ]+$';</code></strong>           -&gt; 0
</pre></li><li><p>
            <code class="literal">[.characters.]</code>
          </p><p>
            Within a bracket expression (written using
            <code class="literal">[</code> and <code class="literal">]</code>), matches the
            sequence of characters of that collating element.
            <code class="literal">characters</code> is either a single character
            or a character name like <code class="literal">newline</code>. The
            following table lists the allowable character names.
          </p><p>
            The following table shows the allowable character names and
            the characters that they match. For characters given as
            numeric values, the values are represented in octal.
          </p><div class="informaltable"><table border="1"><colgroup><col /><col /><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>Name</strong></span></td><td><span class="bold"><strong>Character</strong></span></td><td><span class="bold"><strong>Name</strong></span></td><td><span class="bold"><strong>Character</strong></span></td></tr><tr><td><code class="literal">NUL</code></td><td><code class="literal">0</code></td><td><code class="literal">SOH</code></td><td><code class="literal">001</code></td></tr><tr><td><code class="literal">STX</code></td><td><code class="literal">002</code></td><td><code class="literal">ETX</code></td><td><code class="literal">003</code></td></tr><tr><td><code class="literal">EOT</code></td><td><code class="literal">004</code></td><td><code class="literal">ENQ</code></td><td><code class="literal">005</code></td></tr><tr><td><code class="literal">ACK</code></td><td><code class="literal">006</code></td><td><code class="literal">BEL</code></td><td><code class="literal">007</code></td></tr><tr><td><code class="literal">alert</code></td><td><code class="literal">007</code></td><td><code class="literal">BS</code></td><td><code class="literal">010</code></td></tr><tr><td><code class="literal">backspace</code></td><td><code class="literal">'\b'</code></td><td><code class="literal">HT</code></td><td><code class="literal">011</code></td></tr><tr><td><code class="literal">tab</code></td><td><code class="literal">'\t'</code></td><td><code class="literal">LF</code></td><td><code class="literal">012</code></td></tr><tr><td><code class="literal">newline</code></td><td><code class="literal">'\n'</code></td><td><code class="literal">VT</code></td><td><code class="literal">013</code></td></tr><tr><td><code class="literal">vertical-tab</code></td><td><code class="literal">'\v'</code></td><td><code class="literal">FF</code></td><td><code class="literal">014</code></td></tr><tr><td><code class="literal">form-feed</code></td><td><code class="literal">'\f'</code></td><td><code class="literal">CR</code></td><td><code class="literal">015</code></td></tr><tr><td><code class="literal">carriage-return</code></td><td><code class="literal">'\r'</code></td><td><code class="literal">SO</code></td><td><code class="literal">016</code></td></tr><tr><td><code class="literal">SI</code></td><td><code class="literal">017</code></td><td><code class="literal">DLE</code></td><td><code class="literal">020</code></td></tr><tr><td><code class="literal">DC1</code></td><td><code class="literal">021</code></td><td><code class="literal">DC2</code></td><td><code class="literal">022</code></td></tr><tr><td><code class="literal">DC3</code></td><td><code class="literal">023</code></td><td><code class="literal">DC4</code></td><td><code class="literal">024</code></td></tr><tr><td><code class="literal">NAK</code></td><td><code class="literal">025</code></td><td><code class="literal">SYN</code></td><td><code class="literal">026</code></td></tr><tr><td><code class="literal">ETB</code></td><td><code class="literal">027</code></td><td><code class="literal">CAN</code></td><td><code class="literal">030</code></td></tr><tr><td><code class="literal">EM</code></td><td><code class="literal">031</code></td><td><code class="literal">SUB</code></td><td><code class="literal">032</code></td></tr><tr><td><code class="literal">ESC</code></td><td><code class="literal">033</code></td><td><code class="literal">IS4</code></td><td><code class="literal">034</code></td></tr><tr><td><code class="literal">FS</code></td><td><code class="literal">034</code></td><td><code class="literal">IS3</code></td><td><code class="literal">035</code></td></tr><tr><td><code class="literal">GS</code></td><td><code class="literal">035</code></td><td><code class="literal">IS2</code></td><td><code class="literal">036</code></td></tr><tr><td><code class="literal">RS</code></td><td><code class="literal">036</code></td><td><code class="literal">IS1</code></td><td><code class="literal">037</code></td></tr><tr><td><code class="literal">US</code></td><td><code class="literal">037</code></td><td><code class="literal">space</code></td><td><code class="literal">' '</code></td></tr><tr><td><code class="literal">exclamation-mark</code></td><td><code class="literal">'!'</code></td><td><code class="literal">quotation-mark</code></td><td><code class="literal">'"'</code></td></tr><tr><td><code class="literal">number-sign</code></td><td><code class="literal">'#'</code></td><td><code class="literal">dollar-sign</code></td><td><code class="literal">'$'</code></td></tr><tr><td><code class="literal">percent-sign</code></td><td><code class="literal">'%'</code></td><td><code class="literal">ampersand</code></td><td><code class="literal">'&amp;'</code></td></tr><tr><td><code class="literal">apostrophe</code></td><td><code class="literal">'\''</code></td><td><code class="literal">left-parenthesis</code></td><td><code class="literal">'('</code></td></tr><tr><td><code class="literal">right-parenthesis</code></td><td><code class="literal">')'</code></td><td><code class="literal">asterisk</code></td><td><code class="literal">'*'</code></td></tr><tr><td><code class="literal">plus-sign</code></td><td><code class="literal">'+'</code></td><td><code class="literal">comma</code></td><td><code class="literal">','</code></td></tr><tr><td><code class="literal">hyphen</code></td><td><code class="literal">'-'</code></td><td><code class="literal">hyphen-minus</code></td><td><code class="literal">'-'</code></td></tr><tr><td><code class="literal">period</code></td><td><code class="literal">'.'</code></td><td><code class="literal">full-stop</code></td><td><code class="literal">'.'</code></td></tr><tr><td><code class="literal">slash</code></td><td><code class="literal">'/'</code></td><td><code class="literal">solidus</code></td><td><code class="literal">'/'</code></td></tr><tr><td><code class="literal">zero</code></td><td><code class="literal">'0'</code></td><td><code class="literal">one</code></td><td><code class="literal">'1'</code></td></tr><tr><td><code class="literal">two</code></td><td><code class="literal">'2'</code></td><td><code class="literal">three</code></td><td><code class="literal">'3'</code></td></tr><tr><td><code class="literal">four</code></td><td><code class="literal">'4'</code></td><td><code class="literal">five</code></td><td><code class="literal">'5'</code></td></tr><tr><td><code class="literal">six</code></td><td><code class="literal">'6'</code></td><td><code class="literal">seven</code></td><td><code class="literal">'7'</code></td></tr><tr><td><code class="literal">eight</code></td><td><code class="literal">'8'</code></td><td><code class="literal">nine</code></td><td><code class="literal">'9'</code></td></tr><tr><td><code class="literal">colon</code></td><td><code class="literal">':'</code></td><td><code class="literal">semicolon</code></td><td><code class="literal">';'</code></td></tr><tr><td><code class="literal">less-than-sign</code></td><td><code class="literal">'&lt;'</code></td><td><code class="literal">equals-sign</code></td><td><code class="literal">'='</code></td></tr><tr><td><code class="literal">greater-than-sign</code></td><td><code class="literal">'&gt;'</code></td><td><code class="literal">question-mark</code></td><td><code class="literal">'?'</code></td></tr><tr><td><code class="literal">commercial-at</code></td><td><code class="literal">'@'</code></td><td><code class="literal">left-square-bracket</code></td><td><code class="literal">'['</code></td></tr><tr><td><code class="literal">backslash</code></td><td><code class="literal">'\\'</code></td><td><code class="literal">reverse-solidus</code></td><td><code class="literal">'\\'</code></td></tr><tr><td><code class="literal">right-square-bracket</code></td><td><code class="literal">']'</code></td><td><code class="literal">circumflex</code></td><td><code class="literal">'^'</code></td></tr><tr><td><code class="literal">circumflex-accent</code></td><td><code class="literal">'^'</code></td><td><code class="literal">underscore</code></td><td><code class="literal">'_'</code></td></tr><tr><td><code class="literal">low-line</code></td><td><code class="literal">'_'</code></td><td><code class="literal">grave-accent</code></td><td><code class="literal">'`'</code></td></tr><tr><td><code class="literal">left-brace</code></td><td><code class="literal">'{'</code></td><td><code class="literal">left-curly-bracket</code></td><td><code class="literal">'{'</code></td></tr><tr><td><code class="literal">vertical-line</code></td><td><code class="literal">'|'</code></td><td><code class="literal">right-brace</code></td><td><code class="literal">'}'</code></td></tr><tr><td><code class="literal">right-curly-bracket</code></td><td><code class="literal">'}'</code></td><td><code class="literal">tilde</code></td><td><code class="literal">'~'</code></td></tr><tr><td><code class="literal">DEL</code></td><td><code class="literal">177</code></td><td> </td><td> </td></tr></tbody></table></div><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT '~' REGEXP '[[.~.]]';</code></strong>                     -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT '~' REGEXP '[[.tilde.]]';</code></strong>                 -&gt; 1
</pre></li><li><p>
            <code class="literal">[=character_class=]</code>
          </p><p>
            Within a bracket expression (written using
            <code class="literal">[</code> and <code class="literal">]</code>),
            <code class="literal">[=character_class=]</code> represents an
            equivalence class. It matches all characters with the same
            collation value, including itself. For example, if
            <code class="literal">o</code> and <code class="literal">(+)</code> are the
            members of an equivalence class, then
            <code class="literal">[[=o=]]</code>, <code class="literal">[[=(+)=]]</code>,
            and <code class="literal">[o(+)]</code> are all synonymous. An
            equivalence class may not be used as an endpoint of a range.
          </p></li><li><p>
            <code class="literal">[:character_class:]</code>
          </p><p>
            Within a bracket expression (written using
            <code class="literal">[</code> and <code class="literal">]</code>),
            <code class="literal">[:character_class:]</code> represents a
            character class that matches all characters belonging to
            that class. The following table lists the standard class
            names. These names stand for the character classes defined
            in the <code class="literal">ctype(3)</code> manual page. A particular
            locale may provide other class names. A character class may
            not be used as an endpoint of a range.
          </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><tbody><tr><td><code class="literal">alnum</code></td><td>Alphanumeric characters</td></tr><tr><td><code class="literal">alpha</code></td><td>Alphabetic characters</td></tr><tr><td><code class="literal">blank</code></td><td>Whitespace characters</td></tr><tr><td><code class="literal">cntrl</code></td><td>Control characters</td></tr><tr><td><code class="literal">digit</code></td><td>Digit characters</td></tr><tr><td><code class="literal">graph</code></td><td>Graphic characters</td></tr><tr><td><code class="literal">lower</code></td><td>Lowercase alphabetic characters</td></tr><tr><td><code class="literal">print</code></td><td>Graphic or space characters</td></tr><tr><td><code class="literal">punct</code></td><td>Punctuation characters</td></tr><tr><td><code class="literal">space</code></td><td>Space, tab, newline, and carriage return</td></tr><tr><td><code class="literal">upper</code></td><td>Uppercase alphabetic characters</td></tr><tr><td><code class="literal">xdigit</code></td><td>Hexadecimal digit characters</td></tr></tbody></table></div><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'justalnums' REGEXP '[[:alnum:]]+';</code></strong>       -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT '!!' REGEXP '[[:alnum:]]+';</code></strong>               -&gt; 0
</pre></li><li><p>
            <code class="literal">[[:&lt;:]]</code>, <code class="literal">[[:&gt;:]]</code>
          </p><p>
            These markers stand for word boundaries. They match the
            beginning and end of words, respectively. A word is a
            sequence of word characters that is not preceded by or
            followed by word characters. A word character is an
            alphanumeric character in the <code class="literal">alnum</code> class
            or an underscore (<code class="literal">_</code>).
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'a word a' REGEXP '[[:&lt;:]]word[[:&gt;:]]';</code></strong>   -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 'a xword a' REGEXP '[[:&lt;:]]word[[:&gt;:]]';</code></strong>  -&gt; 0
</pre></li></ul></div><p>
        To use a literal instance of a special character in a regular
        expression, precede it by two backslash (\) characters. The
        MySQL parser interprets one of the backslashes, and the regular
        expression library interprets the other. For example, to match
        the string <code class="literal">1+2</code> that contains the special
        <code class="literal">+</code> character, only the last of the following
        regular expressions is the correct one:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT '1+2' REGEXP '1+2';</code></strong>                       -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT '1+2' REGEXP '1\+2';</code></strong>                      -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT '1+2' REGEXP '1\\+2';</code></strong>                     -&gt; 1
</pre></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="numeric-functions"></a>Numeric Functions</h2></div></div></div><div class="toc"><dl><dt><span class="section"><a href="mysqlqb_functions.html#arithmetic-functions">Arithmetic Operators</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#mathematical-functions">Mathematical Functions</a></span></dt></dl></div><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_abs"><code class="literal">ABS()</code></a></td><td>Return the absolute value</td></tr><tr><td><a href="mysqlqb_functions.html#function_acos"><code class="literal">ACOS()</code></a></td><td>Return the arc cosine</td></tr><tr><td><a href="mysqlqb_functions.html#function_asin"><code class="literal">ASIN()</code></a></td><td>Return the arc sine</td></tr><tr><td><a href="mysqlqb_functions.html#function_atan2"><code class="literal">ATAN2()</code>, <code class="literal">ATAN()</code></a></td><td>Return the arc tangent of the two arguments</td></tr><tr><td><a href="mysqlqb_functions.html#function_atan"><code class="literal">ATAN()</code></a></td><td>Return the arc tangent</td></tr><tr><td><a href="mysqlqb_functions.html#operator_by"><code class="literal">/</code></a></td><td>Division operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_ceiling"><code class="literal">CEILING()</code>, <code class="literal">CEIL()</code></a></td><td>Return the smallest integer value not less than  the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_cos"><code class="literal">COS()</code></a></td><td>Return the cosine</td></tr><tr><td><a href="mysqlqb_functions.html#function_cot"><code class="literal">COT()</code></a></td><td>Return the cotangent</td></tr><tr><td><a href="mysqlqb_functions.html#function_crc32"><code class="literal">CRC32()</code></a></td><td>Compute a cyclic redundancy check value</td></tr><tr><td><a href="mysqlqb_functions.html#function_degrees"><code class="literal">DEGREES()</code></a></td><td>Convert radians to degrees</td></tr><tr><td><a href="mysqlqb_functions.html#operator_div"><code class="literal">DIV</code></a></td><td>Integer division</td></tr><tr><td><a href="mysqlqb_functions.html#function_exp"><code class="literal">EXP()</code></a></td><td>Raise to the power of</td></tr><tr><td><a href="mysqlqb_functions.html#function_floor"><code class="literal">FLOOR()</code></a></td><td>Return the largest integer value not greater than  the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_ln"><code class="literal">LN()</code></a></td><td>Return the natural logarithm of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_log10"><code class="literal">LOG10()</code></a></td><td>Return the base-10 logarithm of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_log2"><code class="literal">LOG2()</code></a></td><td>Return the base-2 logarithm of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_log"><code class="literal">LOG()</code></a></td><td>  Return the natural logarithm of the first argument  </td></tr><tr><td><a href="mysqlqb_functions.html#operator_minus"><code class="literal">-</code></a></td><td>Minus operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_mod"><code class="literal">MOD()</code></a></td><td>Return the remainder</td></tr><tr><td><a href="mysqlqb_functions.html#operator_mod"><code class="literal">%</code></a></td><td>Modulo operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_pi"><code class="literal">PI()</code></a></td><td>Return the value of pi</td></tr><tr><td><a href="mysqlqb_functions.html#operator_plus"><code class="literal">+</code></a></td><td>Addition operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_pow"><code class="literal">POW()</code>, <code class="literal">POWER()</code></a></td><td>Return the argument raised to the specified power</td></tr><tr><td><a href="mysqlqb_functions.html#function_radians"><code class="literal">RADIANS()</code></a></td><td>Return argument converted to radians</td></tr><tr><td><a href="mysqlqb_functions.html#function_rand"><code class="literal">RAND()</code></a></td><td>Return a random floating-point value</td></tr><tr><td><a href="mysqlqb_functions.html#function_round"><code class="literal">ROUND()</code></a></td><td>Round the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_sign"><code class="literal">SIGN()</code></a></td><td>Return the sign of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_sin"><code class="literal">SIN()</code></a></td><td>Return the sine of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_sqrt"><code class="literal">SQRT()</code></a></td><td>Return the square root of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_tan"><code class="literal">TAN()</code></a></td><td>Return the tangent of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#operator_times"><code class="literal">*</code></a></td><td>Times operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_truncate"><code class="literal">TRUNCATE()</code></a></td><td>Truncate to specified number of decimal places</td></tr><tr><td><a href="mysqlqb_functions.html#operator_unary-minus"><code class="literal">-</code></a></td><td>Change the sign of the argument</td></tr></tbody></table></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="arithmetic-functions"></a>Arithmetic Operators</h3></div></div></div><a id="id2828912" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#operator_by"><code class="literal">/</code></a></td><td>Division operator</td></tr><tr><td><a href="mysqlqb_functions.html#function_ceiling"><code class="literal">CEILING()</code>, <code class="literal">CEIL()</code></a></td><td>Return the smallest integer value not less than  the argument</td></tr><tr><td><a href="mysqlqb_functions.html#operator_div"><code class="literal">DIV</code></a></td><td>Integer division</td></tr><tr><td><a href="mysqlqb_functions.html#operator_minus"><code class="literal">-</code></a></td><td>Minus operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_mod"><code class="literal">%</code></a></td><td>Modulo operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_plus"><code class="literal">+</code></a></td><td>Addition operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_times"><code class="literal">*</code></a></td><td>Times operator</td></tr><tr><td><a href="mysqlqb_functions.html#operator_unary-minus"><code class="literal">-</code></a></td><td>Change the sign of the argument</td></tr></tbody></table></div><p>
        The usual arithmetic operators are available. The precision of
        the result is determined according to the following rules:
      </p><div class="itemizedlist"><ul type="disc"><li><p>
            Note that in the case of <code class="literal">-</code>,
            <code class="literal">+</code>, and <code class="literal">*</code>, the result
            is calculated with <code class="literal">BIGINT</code> (64-bit)
            precision if both arguments are integers.
          </p></li><li><p>
            If one of the arguments is an unsigned integer, and the
            other argument is also an integer, the result is an unsigned
            integer.
          </p></li><li><p>
            If any of the operands of a <code class="literal">+</code>,
            <code class="literal">-</code>, <code class="literal">/</code>,
            <code class="literal">*</code>, <code class="literal">%</code> is a real or
            string value, then the precision of the result is the
            precision of the argument with the maximum precision.
          </p></li><li><p>
            In multiplication and division, the precision of the result
            when using two exact values is the precision of the first
            argument + the value of the
            <code class="literal">div_precision_increment</code> global variable.
            For example, the expression <code class="literal">5.05 / 0.0014</code>
            would have a precision of six decimal places
            (<code class="literal">3607.142857</code>).
          </p></li></ul></div><p>
        These rules are applied for each operation, such that nested
        calculations imply the precision of each component. Hence,
        <code class="literal">(14620 / 9432456) / (24250 / 9432456)</code>, would
        resolve first to <code class="literal">(0.0014) / (0.0026)</code>, with
        the final result having 8 decimal places
        (<code class="literal">0.57692308</code>).
      </p><p>
        Because of these rules and the method they are applied, care
        should be taken to ensure that components and sub-components of
        a calculation use the appropriate level of precision. See
        <a href="mysqlqb_functions.html#cast-functions" title="Cast Functions and Operators">Cast Functions and Operators</a>.
      </p><a id="id2829178" class="indexterm"></a><a id="id2829188" class="indexterm"></a><div class="itemizedlist"><ul type="disc"><li><p><a id="operator_plus"></a>
            <a id="id2829215" class="indexterm"></a>

            <a id="id2829222" class="indexterm"></a>

            <code class="literal">+</code>
          </p><p>
            Addition:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 3+5;</code></strong>
        -&gt; 8
</pre></li><li><p><a id="operator_minus"></a>
            <a id="id2829282" class="indexterm"></a>

            <a id="id2829289" class="indexterm"></a>

            <code class="literal">-</code>
          </p><p>
            Subtraction:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 3-5;</code></strong>
        -&gt; -2
</pre></li><li><p><a id="operator_unary-minus"></a>
            <a id="id2829349" class="indexterm"></a>

            <a id="id2829356" class="indexterm"></a>

            <a id="id2829366" class="indexterm"></a>

            <code class="literal">-</code>
          </p><p>
            Unary minus. This operator changes the sign of the argument.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT - 2;</code></strong>
        -&gt; -2
</pre><p>
            <span class="bold"><strong>Note</strong></span>: If this operator is
            used with a <code class="literal">BIGINT</code>, the return value is
            also a <code class="literal">BIGINT</code>. This means that you should
            avoid using <code class="literal">–</code> on integers that may
            have the value of –2<sup>63</sup>.
          </p></li><li><p><a id="operator_times"></a>
            <a id="id2829454" class="indexterm"></a>

            <a id="id2829461" class="indexterm"></a>

            <code class="literal">*</code>
          </p><p>
            Multiplication:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 3*5;</code></strong>
        -&gt; 15
mysql&gt; <strong class="userinput"><code>SELECT 18014398509481984*18014398509481984.0;</code></strong>
        -&gt; 324518553658426726783156020576256.0
mysql&gt; <strong class="userinput"><code>SELECT 18014398509481984*18014398509481984;</code></strong>
        -&gt; 0
</pre><p>
            The result of the last expression is incorrect because the
            result of the integer multiplication exceeds the 64-bit
            range of <code class="literal">BIGINT</code> calculations. (See
            <a href="http://dev.mysql.com/doc/refman/5.0/en/numeric-types.html" target="_top">Numeric Types</a>.)
          </p></li><li><p><a id="operator_by"></a>
            <a id="id2829549" class="indexterm"></a>

            <a id="id2829556" class="indexterm"></a>

            <code class="literal">/</code>
          </p><p>
            Division:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 3/5;</code></strong>
        -&gt; 0.60
</pre><p>
            Division by zero produces a <code class="literal">NULL</code> result:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 102/(1-1);</code></strong>
        -&gt; NULL
</pre><p>
            A division is calculated with <code class="literal">BIGINT</code>
            arithmetic only if performed in a context where its result
            is converted to an integer.
          </p></li><li><p><a id="operator_div"></a>
            <a id="id2829647" class="indexterm"></a>

            <code class="literal">DIV</code>
          </p><p>
            Integer division. Similar to <code class="literal">FLOOR()</code>, but
            is safe with <code class="literal">BIGINT</code> values.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 5 DIV 2;</code></strong>
        -&gt; 2
</pre></li><li><p><a id="operator_mod"></a>
            <a id="id2829720" class="indexterm"></a>

            <code class="literal"><em class="replaceable"><code>N</code></em> %
            <em class="replaceable"><code>M</code></em></code>
          </p><p>
            Modulo operation. Returns the remainder of
            <em class="replaceable"><code>N</code></em> divided by
            <em class="replaceable"><code>M</code></em>. For more information, see the
            description for the <code class="literal">MOD()</code> function in
            <a href="mysqlqb_functions.html#mathematical-functions" title="Mathematical Functions">Mathematical Functions</a>.
          </p></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="mathematical-functions"></a>Mathematical Functions</h3></div></div></div><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_abs"><code class="literal">ABS()</code></a></td><td>Return the absolute value</td></tr><tr><td><a href="mysqlqb_functions.html#function_acos"><code class="literal">ACOS()</code></a></td><td>Return the arc cosine</td></tr><tr><td><a href="mysqlqb_functions.html#function_asin"><code class="literal">ASIN()</code></a></td><td>Return the arc sine</td></tr><tr><td><a href="mysqlqb_functions.html#function_atan2"><code class="literal">ATAN2()</code>, <code class="literal">ATAN()</code></a></td><td>Return the arc tangent of the two arguments</td></tr><tr><td><a href="mysqlqb_functions.html#function_atan"><code class="literal">ATAN()</code></a></td><td>Return the arc tangent</td></tr><tr><td><a href="mysqlqb_functions.html#function_cos"><code class="literal">COS()</code></a></td><td>Return the cosine</td></tr><tr><td><a href="mysqlqb_functions.html#function_cot"><code class="literal">COT()</code></a></td><td>Return the cotangent</td></tr><tr><td><a href="mysqlqb_functions.html#function_crc32"><code class="literal">CRC32()</code></a></td><td>Compute a cyclic redundancy check value</td></tr><tr><td><a href="mysqlqb_functions.html#function_degrees"><code class="literal">DEGREES()</code></a></td><td>Convert radians to degrees</td></tr><tr><td><a href="mysqlqb_functions.html#function_exp"><code class="literal">EXP()</code></a></td><td>Raise to the power of</td></tr><tr><td><a href="mysqlqb_functions.html#function_floor"><code class="literal">FLOOR()</code></a></td><td>Return the largest integer value not greater than  the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_ln"><code class="literal">LN()</code></a></td><td>Return the natural logarithm of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_log10"><code class="literal">LOG10()</code></a></td><td>Return the base-10 logarithm of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_log2"><code class="literal">LOG2()</code></a></td><td>Return the base-2 logarithm of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_log"><code class="literal">LOG()</code></a></td><td>  Return the natural logarithm of the first argument  </td></tr><tr><td><a href="mysqlqb_functions.html#function_mod"><code class="literal">MOD()</code></a></td><td>Return the remainder</td></tr><tr><td><a href="mysqlqb_functions.html#function_pi"><code class="literal">PI()</code></a></td><td>Return the value of pi</td></tr><tr><td><a href="mysqlqb_functions.html#function_pow"><code class="literal">POW()</code>, <code class="literal">POWER()</code></a></td><td>Return the argument raised to the specified power</td></tr><tr><td><a href="mysqlqb_functions.html#function_radians"><code class="literal">RADIANS()</code></a></td><td>Return argument converted to radians</td></tr><tr><td><a href="mysqlqb_functions.html#function_rand"><code class="literal">RAND()</code></a></td><td>Return a random floating-point value</td></tr><tr><td><a href="mysqlqb_functions.html#function_round"><code class="literal">ROUND()</code></a></td><td>Round the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_sign"><code class="literal">SIGN()</code></a></td><td>Return the sign of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_sin"><code class="literal">SIN()</code></a></td><td>Return the sine of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_sqrt"><code class="literal">SQRT()</code></a></td><td>Return the square root of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_tan"><code class="literal">TAN()</code></a></td><td>Return the tangent of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_truncate"><code class="literal">TRUNCATE()</code></a></td><td>Truncate to specified number of decimal places</td></tr></tbody></table></div><p>
        All mathematical functions return <code class="literal">NULL</code> in the
        event of an error.
      </p><a id="id2830127" class="indexterm"></a><a id="id2830134" class="indexterm"></a><div class="itemizedlist"><ul type="disc"><li><p><a id="function_abs"></a>
            <a id="id2830164" class="indexterm"></a>

            <code class="literal">ABS(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the absolute value of <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ABS(2);</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT ABS(-32);</code></strong>
        -&gt; 32
</pre><p>
            This function is safe to use with <code class="literal">BIGINT</code>
            values.
          </p></li><li><p><a id="function_acos"></a>
            <a id="id2830243" class="indexterm"></a>

            <code class="literal">ACOS(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the arc cosine of <em class="replaceable"><code>X</code></em>, that
            is, the value whose cosine is <em class="replaceable"><code>X</code></em>.
            Returns <code class="literal">NULL</code> if
            <em class="replaceable"><code>X</code></em> is not in the range
            <code class="literal">-1</code> to <code class="literal">1</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ACOS(1);</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT ACOS(1.0001);</code></strong>
        -&gt; NULL
mysql&gt; <strong class="userinput"><code>SELECT ACOS(0);</code></strong>
        -&gt; 1.5707963267949
</pre></li><li><p><a id="function_asin"></a>
            <a id="id2830339" class="indexterm"></a>

            <code class="literal">ASIN(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the arc sine of <em class="replaceable"><code>X</code></em>, that
            is, the value whose sine is <em class="replaceable"><code>X</code></em>.
            Returns <code class="literal">NULL</code> if
            <em class="replaceable"><code>X</code></em> is not in the range
            <code class="literal">-1</code> to <code class="literal">1</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ASIN(0.2);</code></strong>
        -&gt; 0.20135792079033
mysql&gt; <strong class="userinput"><code>SELECT ASIN('foo');</code></strong>

+-------------+
| ASIN('foo') |
+-------------+
|           0 |
+-------------+
1 row in set, 1 warning (0.00 sec)

mysql&gt; <strong class="userinput"><code>SHOW WARNINGS;</code></strong>
+---------+------+-----------------------------------------+
| Level   | Code | Message                                 |
+---------+------+-----------------------------------------+
| Warning | 1292 | Truncated incorrect DOUBLE value: 'foo' |
+---------+------+-----------------------------------------+
</pre></li><li><p><a id="function_atan"></a>
            <a id="id2830443" class="indexterm"></a>

            <code class="literal">ATAN(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the arc tangent of <em class="replaceable"><code>X</code></em>,
            that is, the value whose tangent is
            <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ATAN(2);</code></strong>
        -&gt; 1.1071487177941
mysql&gt; <strong class="userinput"><code>SELECT ATAN(-2);</code></strong>
        -&gt; -1.1071487177941
</pre></li><li><p><a id="function_atan2"></a>
            <a id="id2830521" class="indexterm"></a>

            <code class="literal">ATAN(<em class="replaceable"><code>Y</code></em>,<em class="replaceable"><code>X</code></em>)</code>,
            <code class="literal">ATAN2(<em class="replaceable"><code>Y</code></em>,<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the arc tangent of the two variables
            <em class="replaceable"><code>X</code></em> and
            <em class="replaceable"><code>Y</code></em>. It is similar to calculating
            the arc tangent of <code class="literal"><em class="replaceable"><code>Y</code></em> /
            <em class="replaceable"><code>X</code></em></code>, except that the
            signs of both arguments are used to determine the quadrant
            of the result.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ATAN(-2,2);</code></strong>
        -&gt; -0.78539816339745
mysql&gt; <strong class="userinput"><code>SELECT ATAN2(PI(),0);</code></strong>
        -&gt; 1.5707963267949
</pre></li><li><p><a id="function_ceiling"></a>
            <a id="id2830629" class="indexterm"></a>

            <code class="literal">CEILING(<em class="replaceable"><code>X</code></em>)</code>,
            <code class="literal">CEIL(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the smallest integer value not less than
            <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CEILING(1.23);</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT CEIL(-1.23);</code></strong>
        -&gt; -1
</pre><p>
            These two functions are synonymous. Note that the return
            value is converted to a <code class="literal">BIGINT</code>.
          </p></li><li><p><a id="function_cos"></a>
            <a id="id2830717" class="indexterm"></a>

            <code class="literal">COS(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the cosine of <em class="replaceable"><code>X</code></em>, where
            <em class="replaceable"><code>X</code></em> is given in radians.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT COS(PI());</code></strong>
        -&gt; -1
</pre></li><li><p><a id="function_cot"></a>
            <a id="id2830787" class="indexterm"></a>

            <code class="literal">COT(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the cotangent of <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT COT(12);</code></strong>
        -&gt; -1.5726734063977
mysql&gt; <strong class="userinput"><code>SELECT COT(0);</code></strong>
        -&gt; NULL
</pre></li><li><p><a id="function_crc32"></a>
            <a id="id2830860" class="indexterm"></a>

            <code class="literal">CRC32(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Computes a cyclic redundancy check value and returns a
            32-bit unsigned value. The result is <code class="literal">NULL</code>
            if the argument is <code class="literal">NULL</code>. The argument is
            expected to be a string and (if possible) is treated as one
            if it is not.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CRC32('MySQL');</code></strong>
        -&gt; 3259397556
mysql&gt; <strong class="userinput"><code>SELECT CRC32('mysql');</code></strong>
        -&gt; 2501908538
</pre></li><li><p><a id="function_degrees"></a>
            <a id="id2830940" class="indexterm"></a>

            <code class="literal">DEGREES(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the argument <em class="replaceable"><code>X</code></em>, converted
            from radians to degrees.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DEGREES(PI());</code></strong>
        -&gt; 180
mysql&gt; <strong class="userinput"><code>SELECT DEGREES(PI() / 2);</code></strong>
        -&gt; 90
</pre></li><li><p><a id="function_exp"></a>
            <a id="id2831012" class="indexterm"></a>

            <code class="literal">EXP(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the value of <span class="emphasis"><em>e</em></span> (the base of
            natural logarithms) raised to the power of
            <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT EXP(2);</code></strong>
        -&gt; 7.3890560989307
mysql&gt; <strong class="userinput"><code>SELECT EXP(-2);</code></strong>
        -&gt; 0.13533528323661
mysql&gt; <strong class="userinput"><code>SELECT EXP(0);</code></strong>
        -&gt; 1
</pre></li><li><p><a id="function_floor"></a>
            <a id="id2831095" class="indexterm"></a>

            <code class="literal">FLOOR(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the largest integer value not greater than
            <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT FLOOR(1.23);</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT FLOOR(-1.23);</code></strong>
        -&gt; -2
</pre><p>
            Note that the return value is converted to a
            <code class="literal">BIGINT</code>.
          </p></li><li><p>
            <code class="literal">FORMAT(<em class="replaceable"><code>X</code></em>,<em class="replaceable"><code>D</code></em>)</code>
          </p><p>
            Formats the number <em class="replaceable"><code>X</code></em> to a format
            like <code class="literal">'#,###,###.##'</code>, rounded to
            <em class="replaceable"><code>D</code></em> decimal places, and returns the
            result as a string. For details, see
            <a href="mysqlqb_functions.html#string-functions" title="String Functions">String Functions</a>.
          </p></li><li><p><a id="function_ln"></a>
            <a id="id2831221" class="indexterm"></a>

            <code class="literal">LN(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the natural logarithm of
            <em class="replaceable"><code>X</code></em>; that is, the
            base-<span class="emphasis"><em>e</em></span> logarithm of
            <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LN(2);</code></strong>
        -&gt; 0.69314718055995
mysql&gt; <strong class="userinput"><code>SELECT LN(-2);</code></strong>
        -&gt; NULL
</pre><p>
            This function is synonymous with
            <code class="literal">LOG(<em class="replaceable"><code>X</code></em>)</code>.
          </p></li><li><p><a id="function_log"></a>
            <a id="id2831312" class="indexterm"></a>

            <code class="literal">LOG(<em class="replaceable"><code>X</code></em>)</code>,
            <code class="literal">LOG(<em class="replaceable"><code>B</code></em>,<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            If called with one parameter, this function returns the
            natural logarithm of <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LOG(2);</code></strong>
        -&gt; 0.69314718055995
mysql&gt; <strong class="userinput"><code>SELECT LOG(-2);</code></strong>
        -&gt; NULL
</pre><p>
            If called with two parameters, this function returns the
            logarithm of <em class="replaceable"><code>X</code></em> for an arbitrary
            base <em class="replaceable"><code>B</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LOG(2,65536);</code></strong>
        -&gt; 16
mysql&gt; <strong class="userinput"><code>SELECT LOG(10,100);</code></strong>
        -&gt; 2
</pre><p>
            <code class="literal">LOG(<em class="replaceable"><code>B</code></em>,<em class="replaceable"><code>X</code></em>)</code>
            is equivalent to <code class="literal">LOG(<em class="replaceable"><code>X</code></em>)
            / LOG(<em class="replaceable"><code>B</code></em>)</code>.
          </p></li><li><p><a id="function_log2"></a>
            <a id="id2831445" class="indexterm"></a>

            <code class="literal">LOG2(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the base-2 logarithm of
            <code class="literal"><em class="replaceable"><code>X</code></em></code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LOG2(65536);</code></strong>
        -&gt; 16
mysql&gt; <strong class="userinput"><code>SELECT LOG2(-100);</code></strong>
        -&gt; NULL
</pre><p>
            <code class="literal">LOG2()</code> is useful for finding out how many
            bits a number requires for storage. This function is
            equivalent to the expression
            <code class="literal">LOG(<em class="replaceable"><code>X</code></em>) /
            LOG(2)</code>.
          </p></li><li><p><a id="function_log10"></a>
            <a id="id2831534" class="indexterm"></a>

            <code class="literal">LOG10(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the base-10 logarithm of
            <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LOG10(2);</code></strong>
        -&gt; 0.30102999566398
mysql&gt; <strong class="userinput"><code>SELECT LOG10(100);</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT LOG10(-100);</code></strong>
        -&gt; NULL
</pre><p>
            <code class="literal">LOG10(<em class="replaceable"><code>X</code></em>)</code> is
            equivalent to
            <code class="literal">LOG(10,<em class="replaceable"><code>X</code></em>)</code>.
          </p></li><li><p><a id="function_mod"></a>
            <a id="id2831634" class="indexterm"></a>

            <a id="id2831641" class="indexterm"></a>

            <a id="id2831648" class="indexterm"></a>

            <a id="id2831655" class="indexterm"></a>

            <a id="id2831661" class="indexterm"></a>

            <code class="literal">MOD(<em class="replaceable"><code>N</code></em>,<em class="replaceable"><code>M</code></em>)</code>,
            <code class="literal"><em class="replaceable"><code>N</code></em> %
            <em class="replaceable"><code>M</code></em></code>,
            <code class="literal"><em class="replaceable"><code>N</code></em> MOD
            <em class="replaceable"><code>M</code></em></code>
          </p><p>
            Modulo operation. Returns the remainder of
            <em class="replaceable"><code>N</code></em> divided by
            <em class="replaceable"><code>M</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MOD(234, 10);</code></strong>
        -&gt; 4
mysql&gt; <strong class="userinput"><code>SELECT 253 % 7;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT MOD(29,9);</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT 29 MOD 9;</code></strong>
        -&gt; 2
</pre><p>
            This function is safe to use with <code class="literal">BIGINT</code>
            values.
          </p><p>
            <code class="literal">MOD()</code> also works on values that have a
            fractional part and returns the exact remainder after
            division:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MOD(34.5,3);</code></strong>
        -&gt; 1.5
</pre><p>
            <code class="literal">MOD(<em class="replaceable"><code>N</code></em>,0)</code>
            returns <code class="literal">NULL</code>.
          </p></li><li><p><a id="function_pi"></a>
            <a id="id2831804" class="indexterm"></a>

            <code class="literal">PI()</code>
          </p><p>
            Returns the value of π (pi). The default number of
            decimal places displayed is seven, but MySQL uses the full
            double-precision value internally.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT PI();</code></strong>
        -&gt; 3.141593
mysql&gt; <strong class="userinput"><code>SELECT PI()+0.000000000000000000;</code></strong>
        -&gt; 3.141592653589793116
</pre></li><li><p><a id="function_pow"></a>
            <a id="id2831882" class="indexterm"></a>

            <a id="id2831889" class="indexterm"></a>

            <code class="literal">POW(<em class="replaceable"><code>X</code></em>,<em class="replaceable"><code>Y</code></em>)</code>,
            <code class="literal">POWER(<em class="replaceable"><code>X</code></em>,<em class="replaceable"><code>Y</code></em>)</code>
          </p><p>
            Returns the value of <em class="replaceable"><code>X</code></em> raised to
            the power of <em class="replaceable"><code>Y</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT POW(2,2);</code></strong>
        -&gt; 4
mysql&gt; <strong class="userinput"><code>SELECT POW(2,-2);</code></strong>
        -&gt; 0.25
</pre></li><li><p><a id="function_radians"></a>
            <a id="id2831978" class="indexterm"></a>

            <code class="literal">RADIANS(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the argument <em class="replaceable"><code>X</code></em>, converted
            from degrees to radians. (Note that π radians equals 180
            degrees.)
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT RADIANS(90);</code></strong>
        -&gt; 1.5707963267949
</pre></li><li><p><a id="function_rand"></a>
            <a id="id2832049" class="indexterm"></a>

            <code class="literal">RAND()</code>,
            <code class="literal">RAND(<em class="replaceable"><code>N</code></em>)</code>
          </p><p>
            Returns a random floating-point value
            <em class="replaceable"><code>v</code></em> in the range
            <code class="literal">0</code> &lt;= <em class="replaceable"><code>v</code></em> &lt;
            <code class="literal">1.0</code>. If a constant integer argument
            <em class="replaceable"><code>N</code></em> is specified, it is used as the
            seed value, which produces a repeatable sequence of column
            values.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT RAND();</code></strong>
        -&gt; 0.9233482386203
mysql&gt; <strong class="userinput"><code>SELECT RAND(20);</code></strong>
        -&gt; 0.15888261251047
mysql&gt; <strong class="userinput"><code>SELECT RAND(20);</code></strong>
        -&gt; 0.15888261251047
mysql&gt; <strong class="userinput"><code>SELECT RAND();</code></strong>
        -&gt; 0.63553050033332
mysql&gt; <strong class="userinput"><code>SELECT RAND();</code></strong>
        -&gt; 0.70100469486881
mysql&gt; <strong class="userinput"><code>SELECT RAND(20);</code></strong>
        -&gt; 0.15888261251047
</pre><p>
            The effect of using a non-constant argument is undefined. As
            of MySQL 5.0.13, non-constant arguments are disallowed.
          </p><p>
            To obtain a random integer <em class="replaceable"><code>R</code></em> in
            the range <em class="replaceable"><code>i</code></em> &lt;=
            <em class="replaceable"><code>R</code></em> &lt;
            <em class="replaceable"><code>j</code></em>, use the expression
            <code class="literal">FLOOR(<em class="replaceable"><code>i</code></em> + RAND() *
            (<em class="replaceable"><code>j</code></em> –
            <em class="replaceable"><code>i</code></em>)</code>. For example, to
            obtain a random integer in the range the range
            <code class="literal">7</code> &lt;= <em class="replaceable"><code>R</code></em> &lt;
            <code class="literal">12</code>, you could use the following
            statement:
          </p><pre class="programlisting">SELECT FLOOR(7 + (RAND() * 5));
</pre><p>
            You cannot use a column with <code class="literal">RAND()</code>
            values in an <code class="literal">ORDER BY</code> clause, because
            <code class="literal">ORDER BY</code> would evaluate the column
            multiple times. However, you can retrieve rows in random
            order like this:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT * FROM <em class="replaceable"><code>tbl_name</code></em> ORDER BY RAND();</code></strong>
</pre><p>
            <code class="literal">ORDER BY RAND()</code> combined with
            <code class="literal">LIMIT</code> is useful for selecting a random
            sample from a set of rows:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT * FROM table1, table2 WHERE a=b AND c&lt;d</code></strong> -&gt; <strong class="userinput"><code>ORDER BY RAND() LIMIT 1000;</code></strong>
</pre><p>
            Note that <code class="literal">RAND()</code> in a
            <code class="literal">WHERE</code> clause is re-evaluated every time
            the <code class="literal">WHERE</code> is executed.
          </p><p>
            <code class="literal">RAND()</code> is not meant to be a perfect
            random generator, but instead is a fast way to generate
            <span class="foreignphrase"><em class="foreignphrase">ad hoc</em></span> random numbers which
            is portable between platforms for the same MySQL version.
          </p></li><li><p><a id="function_round"></a>
            <a id="id2832307" class="indexterm"></a>

            <code class="literal">ROUND(<em class="replaceable"><code>X</code></em>)</code>,
            <code class="literal">ROUND(<em class="replaceable"><code>X</code></em>,<em class="replaceable"><code>D</code></em>)</code>
          </p><p>
            Rounds the argument <em class="replaceable"><code>X</code></em> to
            <em class="replaceable"><code>D</code></em> decimal places. The rounding
            algorithm depends on the data type of
            <em class="replaceable"><code>X</code></em>. <em class="replaceable"><code>D</code></em>
            defaults to 0 if not specified. <em class="replaceable"><code>D</code></em>
            can be negative to cause <em class="replaceable"><code>D</code></em> digits
            left of the decimal point of the value
            <em class="replaceable"><code>X</code></em> to become zero.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ROUND(-1.23);</code></strong>
        -&gt; -1
mysql&gt; <strong class="userinput"><code>SELECT ROUND(-1.58);</code></strong>
        -&gt; -2
mysql&gt; <strong class="userinput"><code>SELECT ROUND(1.58);</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT ROUND(1.298, 1);</code></strong>
        -&gt; 1.3
mysql&gt; <strong class="userinput"><code>SELECT ROUND(1.298, 0);</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT ROUND(23.298, -1);</code></strong>
        -&gt; 20
</pre><p>
            The return type is the same type as that of the first
            argument (assuming that it is integer, double, or decimal).
            This means that for an integer argument, the result is an
            integer (no decimal places).
          </p><p>
            Before MySQL 5.0.3, the behavior of
            <code class="literal">ROUND()</code> when the argument is halfway
            between two integers depends on the C library
            implementation. Different implementations round to the
            nearest even number, always up, always down, or always
            toward zero. If you need one kind of rounding, you should
            use a well-defined function such as
            <code class="literal">TRUNCATE()</code> or <code class="literal">FLOOR()</code>
            instead.
          </p><p>
            As of MySQL 5.0.3, <code class="literal">ROUND()</code> uses the
            precision math library for exact-value arguments when the
            first argument is a decimal value:
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                For exact-value numbers, <code class="literal">ROUND()</code> uses
                the “<span class="quote">round half up</span>” or “<span class="quote">round toward
                nearest</span>” rule: A value with a fractional part of
                .5 or greater is rounded up to the next integer if
                positive or down to the next integer if negative. (In
                other words, it is rounded away from zero.) A value with
                a fractional part less than .5 is rounded down to the
                next integer if positive or up to the next integer if
                negative.
              </p></li><li><p>
                For approximate-value numbers, the result depends on the
                C library. On many systems, this means that
                <code class="literal">ROUND()</code> uses the "round to nearest
                even" rule: A value with any fractional part is rounded
                to the nearest even integer.
              </p></li></ul></div><p>
            The following example shows how rounding differs for exact
            and approximate values:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ROUND(2.5), ROUND(25E-1);</code></strong>
+------------+--------------+
| ROUND(2.5) | ROUND(25E-1) |
+------------+--------------+
| 3          |            2 |
+------------+--------------+
</pre><p>
            For more information, see <a href="http://dev.mysql.com/doc/refman/5.0/en/precision-math.html" target="_top">Precision Math</a>.
          </p></li><li><p><a id="function_sign"></a>
            <a id="id2832541" class="indexterm"></a>

            <code class="literal">SIGN(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the sign of the argument as <code class="literal">-1</code>,
            <code class="literal">0</code>, or <code class="literal">1</code>, depending on
            whether <em class="replaceable"><code>X</code></em> is negative, zero, or
            positive.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SIGN(-32);</code></strong>
        -&gt; -1
mysql&gt; <strong class="userinput"><code>SELECT SIGN(0);</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT SIGN(234);</code></strong>
        -&gt; 1
</pre></li><li><p><a id="function_sin"></a>
            <a id="id2832628" class="indexterm"></a>

            <code class="literal">SIN(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the sine of <em class="replaceable"><code>X</code></em>, where
            <em class="replaceable"><code>X</code></em> is given in radians.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SIN(PI());</code></strong>
        -&gt; 1.2246063538224e-16
mysql&gt; <strong class="userinput"><code>SELECT ROUND(SIN(PI()));</code></strong>
        -&gt; 0
</pre></li><li><p><a id="function_sqrt"></a>
            <a id="id2832705" class="indexterm"></a>

            <code class="literal">SQRT(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the square root of a non-negative number
            <em class="replaceable"><code>X</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SQRT(4);</code></strong>
        -&gt; 2
mysql&gt; <strong class="userinput"><code>SELECT SQRT(20);</code></strong>
        -&gt; 4.4721359549996
mysql&gt; <strong class="userinput"><code>SELECT SQRT(-16);</code></strong>
        -&gt; NULL        
</pre></li><li><p><a id="function_tan"></a>
            <a id="id2832782" class="indexterm"></a>

            <code class="literal">TAN(<em class="replaceable"><code>X</code></em>)</code>
          </p><p>
            Returns the tangent of <em class="replaceable"><code>X</code></em>, where
            <em class="replaceable"><code>X</code></em> is given in radians.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TAN(PI());</code></strong>
        -&gt; -1.2246063538224e-16
mysql&gt; <strong class="userinput"><code>SELECT TAN(PI()+1);</code></strong>
        -&gt; 1.5574077246549
</pre></li><li><p><a id="function_truncate"></a>
            <a id="id2832860" class="indexterm"></a>

            <code class="literal">TRUNCATE(<em class="replaceable"><code>X</code></em>,<em class="replaceable"><code>D</code></em>)</code>
          </p><p>
            Returns the number <em class="replaceable"><code>X</code></em>, truncated
            to <em class="replaceable"><code>D</code></em> decimal places. If
            <em class="replaceable"><code>D</code></em> is <code class="literal">0</code>, the
            result has no decimal point or fractional part.
            <em class="replaceable"><code>D</code></em> can be negative to cause
            <em class="replaceable"><code>D</code></em> digits left of the decimal
            point of the value <em class="replaceable"><code>X</code></em> to become
            zero.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TRUNCATE(1.223,1);</code></strong>
        -&gt; 1.2
mysql&gt; <strong class="userinput"><code>SELECT TRUNCATE(1.999,1);</code></strong>
        -&gt; 1.9
mysql&gt; <strong class="userinput"><code>SELECT TRUNCATE(1.999,0);</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT TRUNCATE(-1.999,1);</code></strong>
        -&gt; -1.9
mysql&gt; <strong class="userinput"><code>SELECT TRUNCATE(122,-2);</code></strong>
       -&gt; 100
mysql&gt; <strong class="userinput"><code>SELECT TRUNCATE(10.28*100,0);</code></strong>
       -&gt; 1028
</pre><p>
            All numbers are rounded toward zero.
          </p></li></ul></div></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="date-and-time-functions"></a>Date and Time Functions</h2></div></div></div><a id="id2832982" class="indexterm"></a><a id="id2832989" class="indexterm"></a><p>
      This section describes the functions that can be used to
      manipulate temporal values. See
      <a href="http://dev.mysql.com/doc/refman/5.0/en/date-and-time-types.html" target="_top">Date and Time Types</a>, for a description of the
      range of values each date and time type has and the valid formats
      in which values may be specified.
    </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_adddate"><code class="literal">ADDDATE()</code></a></td><td>Add dates</td></tr><tr><td><a href="mysqlqb_functions.html#function_addtime"><code class="literal">ADDTIME()</code></a></td><td>Add time</td></tr><tr><td><a href="mysqlqb_functions.html#function_convert-tz"><code class="literal">CONVERT_TZ()</code></a></td><td>Convert from one timezone to another</td></tr><tr><td><a href="mysqlqb_functions.html#function_curdate"><code class="literal">CURDATE()</code></a></td><td>Return the current date</td></tr><tr><td><a href="mysqlqb_functions.html#function_current-date"><code class="literal">CURRENT_DATE()</code>, <code class="literal">CURRENT_DATE</code></a></td><td>Synonyms for CURDATE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_current-time"><code class="literal">CURRENT_TIME()</code>, <code class="literal">CURRENT_TIME</code></a></td><td>Synonyms for CURTIME()</td></tr><tr><td><a href="mysqlqb_functions.html#function_current-timestamp"><code class="literal">CURRENT_TIMESTAMP()</code>, <code class="literal">CURRENT_TIMESTAMP</code></a></td><td>Synonyms for NOW()</td></tr><tr><td><a href="mysqlqb_functions.html#function_curtime"><code class="literal">CURTIME()</code></a></td><td>Return the current time</td></tr><tr><td><a href="mysqlqb_functions.html#function_date-add"><code class="literal">DATE_ADD()</code></a></td><td>Add two dates</td></tr><tr><td><a href="mysqlqb_functions.html#function_date-format"><code class="literal">DATE_FORMAT()</code></a></td><td>Format date as specified</td></tr><tr><td><a href="mysqlqb_functions.html#function_date-sub"><code class="literal">DATE_SUB()</code></a></td><td>Subtract two dates</td></tr><tr><td><a href="mysqlqb_functions.html#function_date"><code class="literal">DATE()</code></a></td><td>Extract the date part of a date or datetime expression</td></tr><tr><td><a href="mysqlqb_functions.html#function_datediff"><code class="literal">DATEDIFF()</code></a></td><td>Subtract two dates</td></tr><tr><td><a href="mysqlqb_functions.html#function_day"><code class="literal">DAY()</code></a></td><td>Synonym for DAYOFMONTH()</td></tr><tr><td><a href="mysqlqb_functions.html#function_dayname"><code class="literal">DAYNAME()</code></a></td><td>Return the name of the weekday</td></tr><tr><td><a href="mysqlqb_functions.html#function_dayofmonth"><code class="literal">DAYOFMONTH()</code></a></td><td>Return the day of the month (1-31)</td></tr><tr><td><a href="mysqlqb_functions.html#function_dayofweek"><code class="literal">DAYOFWEEK()</code></a></td><td>Return the weekday index of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_dayofyear"><code class="literal">DAYOFYEAR()</code></a></td><td>Return the day of the year (1-366)</td></tr><tr><td><a href="mysqlqb_functions.html#function_extract"><code class="literal">EXTRACT</code></a></td><td>Extract part of a date</td></tr><tr><td><a href="mysqlqb_functions.html#function_from-days"><code class="literal">FROM_DAYS()</code></a></td><td>Convert a day number to a date</td></tr><tr><td><a href="mysqlqb_functions.html#function_from-unixtime"><code class="literal">FROM_UNIXTIME()</code></a></td><td>Format date as a UNIX timestamp</td></tr><tr><td><a href="mysqlqb_functions.html#function_get-format"><code class="literal">GET_FORMAT()</code></a></td><td>Return a date format string</td></tr><tr><td><a href="mysqlqb_functions.html#function_hour"><code class="literal">HOUR()</code></a></td><td>Extract the hour</td></tr><tr><td><a href="mysqlqb_functions.html#function_last-day"><code class="literal">LAST_DAY</code></a></td><td>Return the last day of the month for the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_localtime"><code class="literal">LOCALTIME()</code>, <code class="literal">LOCALTIME</code></a></td><td>Synonym for NOW()</td></tr><tr><td><a href="mysqlqb_functions.html#function_localtimestamp"><code class="literal">LOCALTIMESTAMP</code>, <code class="literal">LOCALTIMESTAMP()</code></a></td><td>Synonym for NOW()</td></tr><tr><td><a href="mysqlqb_functions.html#function_makedate"><code class="literal">MAKEDATE()</code></a></td><td>Create a date from the year and day of year</td></tr><tr><td><a href="mysqlqb_functions.html#function_maketime"><code class="literal">MAKETIME</code></a></td><td>MAKETIME()</td></tr><tr><td><a href="mysqlqb_functions.html#function_microsecond"><code class="literal">MICROSECOND()</code></a></td><td>Return the microseconds from argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_minute"><code class="literal">MINUTE()</code></a></td><td>Return the minute from the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_month"><code class="literal">MONTH()</code></a></td><td>Return the month from the date passed</td></tr><tr><td><a href="mysqlqb_functions.html#function_monthname"><code class="literal">MONTHNAME()</code></a></td><td>Return the name of the month</td></tr><tr><td><a href="mysqlqb_functions.html#function_now"><code class="literal">NOW()</code></a></td><td>Return the current date and time</td></tr><tr><td><a href="mysqlqb_functions.html#function_period-add"><code class="literal">PERIOD_ADD()</code></a></td><td>Add a period to a year-month</td></tr><tr><td><a href="mysqlqb_functions.html#function_period-diff"><code class="literal">PERIOD_DIFF()</code></a></td><td>Return the number of months between periods</td></tr><tr><td><a href="mysqlqb_functions.html#function_quarter"><code class="literal">QUARTER()</code></a></td><td>Return the quarter from a date argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_sec-to-time"><code class="literal">SEC_TO_TIME()</code></a></td><td>Converts seconds to 'HH:MM:SS' format</td></tr><tr><td><a href="mysqlqb_functions.html#function_second"><code class="literal">SECOND()</code></a></td><td>Return the second (0-59)</td></tr><tr><td><a href="mysqlqb_functions.html#function_str-to-date"><code class="literal">STR_TO_DATE()</code></a></td><td>Convert a string to a date</td></tr><tr><td><a href="mysqlqb_functions.html#function_subdate"><code class="literal">SUBDATE()</code></a></td><td>When invoked with three arguments a synonym for DATE_SUB()</td></tr><tr><td><a href="mysqlqb_functions.html#function_subtime"><code class="literal">SUBTIME()</code></a></td><td>Subtract times</td></tr><tr><td><a href="mysqlqb_functions.html#function_sysdate"><code class="literal">SYSDATE()</code></a></td><td>Return the time at which the function executes</td></tr><tr><td><a href="mysqlqb_functions.html#function_time-format"><code class="literal">TIME_FORMAT()</code></a></td><td>Format as time</td></tr><tr><td><a href="mysqlqb_functions.html#function_time-to-sec"><code class="literal">TIME_TO_SEC()</code></a></td><td>Return the argument converted to seconds</td></tr><tr><td><a href="mysqlqb_functions.html#function_time"><code class="literal">TIME()</code></a></td><td>Extract the time portion of the expression passed</td></tr><tr><td><a href="mysqlqb_functions.html#function_timediff"><code class="literal">TIMEDIFF()</code></a></td><td>Subtract time</td></tr><tr><td><a href="mysqlqb_functions.html#function_timestamp"><code class="literal">TIMESTAMP()</code></a></td><td>With a single argument, this function returns the date or  datetime expression. With two arguments, the sum of the arguments</td></tr><tr><td><a href="mysqlqb_functions.html#function_timestampadd"><code class="literal">TIMESTAMPADD()</code></a></td><td>Add an interval to a datetime expression</td></tr><tr><td><a href="mysqlqb_functions.html#function_timestampdiff"><code class="literal">TIMESTAMPDIFF()</code></a></td><td>Subtract an interval from a datetime expression</td></tr><tr><td><a href="mysqlqb_functions.html#function_to-days"><code class="literal">TO_DAYS()</code></a></td><td>Return the date argument converted to days</td></tr><tr><td><a href="mysqlqb_functions.html#function_unix-timestamp"><code class="literal">UNIX_TIMESTAMP()</code></a></td><td>Return a UNIX timestamp</td></tr><tr><td><a href="mysqlqb_functions.html#function_utc-date"><code class="literal">UTC_DATE()</code></a></td><td>Return the current UTC date</td></tr><tr><td><a href="mysqlqb_functions.html#function_utc-time"><code class="literal">UTC_TIME()</code></a></td><td>Return the current UTC time</td></tr><tr><td><a href="mysqlqb_functions.html#function_utc-timestamp"><code class="literal">UTC_TIMESTAMP()</code></a></td><td>Return the current UTC date and time</td></tr><tr><td><a href="mysqlqb_functions.html#function_week"><code class="literal">WEEK()</code></a></td><td>Return the week number</td></tr><tr><td><a href="mysqlqb_functions.html#function_weekday"><code class="literal">WEEKDAY()</code></a></td><td>Return the weekday index</td></tr><tr><td><a href="mysqlqb_functions.html#function_weekofyear"><code class="literal">WEEKOFYEAR()</code></a></td><td>Return the calendar week of the date (1-53)</td></tr><tr><td><a href="mysqlqb_functions.html#function_year"><code class="literal">YEAR()</code></a></td><td>Return the year</td></tr><tr><td><a href="mysqlqb_functions.html#function_yearweek"><code class="literal">YEARWEEK()</code></a></td><td>Return the year and week</td></tr></tbody></table></div><p>
      Here is an example that uses date functions. The following query
      selects all rows with a <em class="replaceable"><code>date_col</code></em> value
      from within the last 30 days:
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT <em class="replaceable"><code>something</code></em> FROM <em class="replaceable"><code>tbl_name</code></em></code></strong>
    -&gt; <strong class="userinput"><code>WHERE DATE_SUB(CURDATE(),INTERVAL 30 DAY) &lt;= <em class="replaceable"><code>date_col</code></em>;</code></strong>
</pre><p>
      Note that the query also selects rows with dates that lie in the
      future.
    </p><p>
      Functions that expect date values usually accept datetime values
      and ignore the time part. Functions that expect time values
      usually accept datetime values and ignore the date part.
    </p><p>
      Functions that return the current date or time each are evaluated
      only once per query at the start of query execution. This means
      that multiple references to a function such as
      <code class="literal">NOW()</code> within a single query always produce the
      same result (for our purposes a single query also includes a call
      to a stored routine or trigger and all sub-routines called by that
      routine/trigger). This principle also applies to
      <code class="literal">CURDATE()</code>, <code class="literal">CURTIME()</code>,
      <code class="literal">UTC_DATE()</code>, <code class="literal">UTC_TIME()</code>,
      <code class="literal">UTC_TIMESTAMP()</code>, and to any of their synonyms.
    </p><p>
      The <code class="literal">CURRENT_TIMESTAMP()</code>,
      <code class="literal">CURRENT_TIME()</code>,
      <code class="literal">CURRENT_DATE()</code>, and
      <code class="literal">FROM_UNIXTIME()</code> functions return values in the
      connection's current time zone, which is available as the value of
      the <code class="literal">time_zone</code> system variable. In addition,
      <code class="literal">UNIX_TIMESTAMP()</code> assumes that its argument is a
      datetime value in the current time zone. See
      <a href="http://dev.mysql.com/doc/refman/5.0/en/time-zone-support.html" target="_top">MySQL Server Time Zone Support</a>.
    </p><p>
      Some date functions can be used with “<span class="quote">zero</span>” dates or
      incomplete dates such as <code class="literal">'2001-11-00'</code>, whereas
      others cannot. Functions that extract parts of dates typically
      work with incomplete dates. For example:
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DAYOFMONTH('2001-11-00'), MONTH('2005-00-00');</code></strong>
        -&gt; 0, 0
</pre><p>
      Other functions expect complete dates and return
      <code class="literal">NULL</code> for incomplete dates. These include
      functions that perform date arithmetic or that map parts of dates
      to names. For example:
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('2006-05-00',INTERVAL 1 DAY);</code></strong>
        -&gt; NULL
mysql&gt; <strong class="userinput"><code>SELECT DAYNAME('2006-05-00');</code></strong>
        -&gt; NULL
</pre><div class="itemizedlist"><ul type="disc"><li><p><a id="function_adddate"></a>
          <a id="id2833893" class="indexterm"></a>

          <code class="literal">ADDDATE(<em class="replaceable"><code>date</code></em>,INTERVAL
          <em class="replaceable"><code>expr</code></em>
          <em class="replaceable"><code>unit</code></em>)</code>,
          <code class="literal">ADDDATE(<em class="replaceable"><code>expr</code></em>,<em class="replaceable"><code>days</code></em>)</code>
        </p><p>
          When invoked with the <code class="literal">INTERVAL</code> form of the
          second argument, <code class="literal">ADDDATE()</code> is a synonym for
          <code class="literal">DATE_ADD()</code>. The related function
          <code class="literal">SUBDATE()</code> is a synonym for
          <code class="literal">DATE_SUB()</code>. For information on the
          <code class="literal">INTERVAL</code> <em class="replaceable"><code>unit</code></em>
          argument, see the discussion for
          <code class="literal">DATE_ADD()</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('1998-01-02', INTERVAL 31 DAY);</code></strong>
        -&gt; '1998-02-02'
mysql&gt; <strong class="userinput"><code>SELECT ADDDATE('1998-01-02', INTERVAL 31 DAY);</code></strong>
        -&gt; '1998-02-02'
</pre><p>
          When invoked with the <em class="replaceable"><code>days</code></em> form of
          the second argument, MySQL treats it as an integer number of
          days to be added to <em class="replaceable"><code>expr</code></em>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ADDDATE('1998-01-02', 31);</code></strong>
        -&gt; '1998-02-02'
</pre></li><li><p><a id="function_addtime"></a>
          <a id="id2834036" class="indexterm"></a>

          <code class="literal">ADDTIME(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>)</code>
        </p><p>
          <code class="literal">ADDTIME()</code> adds
          <em class="replaceable"><code>expr2</code></em> to
          <em class="replaceable"><code>expr1</code></em> and returns the result.
          <em class="replaceable"><code>expr1</code></em> is a time or datetime
          expression, and <em class="replaceable"><code>expr2</code></em> is a time
          expression.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ADDTIME('1997-12-31 23:59:59.999999',</code></strong>
    -&gt;                <strong class="userinput"><code>'1 1:1:1.000002');</code></strong>
        -&gt; '1998-01-02 01:01:01.000001'
mysql&gt; <strong class="userinput"><code>SELECT ADDTIME('01:00:00.999999', '02:00:00.999998');</code></strong>
        -&gt; '03:00:01.999997'
</pre></li><li><p><a id="function_convert-tz"></a>
          <a id="id2834142" class="indexterm"></a>

          <code class="literal">CONVERT_TZ(<em class="replaceable"><code>dt</code></em>,<em class="replaceable"><code>from_tz</code></em>,<em class="replaceable"><code>to_tz</code></em>)</code>
        </p><p>
          <code class="literal">CONVERT_TZ()</code> converts a datetime value
          <em class="replaceable"><code>dt</code></em> from the time zone given by
          <em class="replaceable"><code>from_tz</code></em> to the time zone given by
          <em class="replaceable"><code>to_tz</code></em> and returns the resulting
          value. Time zones are specified as described in
          <a href="http://dev.mysql.com/doc/refman/5.0/en/time-zone-support.html" target="_top">MySQL Server Time Zone Support</a>. This function returns
          <code class="literal">NULL</code> if the arguments are invalid.
        </p><p>
          If the value falls out of the supported range of the
          <code class="literal">TIMESTAMP</code> type when converted fom
          <em class="replaceable"><code>from_tz</code></em> to UTC, no conversion
          occurs. The <code class="literal">TIMESTAMP</code> range is described in
          <a href="http://dev.mysql.com/doc/refman/5.0/en/date-and-time-type-overview.html" target="_top">Overview of Date and Time Types</a>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CONVERT_TZ('2004-01-01 12:00:00','GMT','MET');</code></strong>
        -&gt; '2004-01-01 13:00:00'
mysql&gt; <strong class="userinput"><code>SELECT CONVERT_TZ('2004-01-01 12:00:00','+00:00','+10:00');</code></strong>
        -&gt; '2004-01-01 22:00:00'
</pre><p>
          <span class="bold"><strong>Note</strong></span>: To use named time zones
          such as <code class="literal">'MET'</code> or
          <code class="literal">'Europe/Moscow'</code>, the time zone tables must
          be properly set up. See <a href="http://dev.mysql.com/doc/refman/5.0/en/time-zone-support.html" target="_top">MySQL Server Time Zone Support</a>,
          for instructions.
        </p><p>
          If you intend to use <code class="literal">CONVERT_TZ()</code> while
          other tables are locked with <code class="literal">LOCK TABLES</code>,
          you must also lock the <code class="literal">mysql.time_zone_name</code>
          table.
        </p></li><li><p><a id="function_curdate"></a>
          <a id="id2834318" class="indexterm"></a>

          <code class="literal">CURDATE()</code>
        </p><p>
          Returns the current date as a value in
          <code class="literal">'YYYY-MM-DD'</code> or <code class="literal">YYYYMMDD</code>
          format, depending on whether the function is used in a string
          or numeric context.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CURDATE();</code></strong>
        -&gt; '1997-12-15'
mysql&gt; <strong class="userinput"><code>SELECT CURDATE() + 0;</code></strong>
        -&gt; 19971215
</pre></li><li><p><a id="function_current-date"></a>
          <a id="id2834394" class="indexterm"></a>

          <code class="literal">CURRENT_DATE</code>,
          <code class="literal">CURRENT_DATE()</code>
        </p><p>
          <code class="literal">CURRENT_DATE</code> and
          <code class="literal">CURRENT_DATE()</code> are synonyms for
          <code class="literal">CURDATE()</code>.
        </p></li><li><p><a id="function_curtime"></a>
          <a id="id2834457" class="indexterm"></a>

          <code class="literal">CURTIME()</code>
        </p><p>
          Returns the current time as a value in
          <code class="literal">'HH:MM:SS'</code> or <code class="literal">HHMMSS</code>
          format, depending on whether the function is used in a string
          or numeric context. The value is expressed in the current time
          zone.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CURTIME();</code></strong>
        -&gt; '23:50:26'
mysql&gt; <strong class="userinput"><code>SELECT CURTIME() + 0;</code></strong>
        -&gt; 235026
</pre></li><li><p><a id="function_current-time"></a>
          <a id="id2834533" class="indexterm"></a>

          <code class="literal">CURRENT_TIME</code>,
          <code class="literal">CURRENT_TIME()</code>
        </p><p>
          <code class="literal">CURRENT_TIME</code> and
          <code class="literal">CURRENT_TIME()</code> are synonyms for
          <code class="literal">CURTIME()</code>.
        </p></li><li><p><a id="function_current-timestamp"></a>
          <a id="id2834597" class="indexterm"></a>

          <code class="literal">CURRENT_TIMESTAMP</code>,
          <code class="literal">CURRENT_TIMESTAMP()</code>
        </p><p>
          <code class="literal">CURRENT_TIMESTAMP</code> and
          <code class="literal">CURRENT_TIMESTAMP()</code> are synonyms for
          <code class="literal">NOW()</code>.
        </p></li><li><p><a id="function_date"></a>
          <a id="id2834667" class="indexterm"></a>

          <code class="literal">DATE(<em class="replaceable"><code>expr</code></em>)</code>
        </p><p>
          Extracts the date part of the date or datetime expression
          <em class="replaceable"><code>expr</code></em>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATE('2003-12-31 01:02:03');</code></strong>
        -&gt; '2003-12-31'
</pre></li><li><p><a id="function_datediff"></a>
          <a id="id2834736" class="indexterm"></a>

          <code class="literal">DATEDIFF(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>)</code>
        </p><p>
          <code class="literal">DATEDIFF()</code> returns
          <em class="replaceable"><code>expr1</code></em> –
          <em class="replaceable"><code>expr2</code></em> expressed as a value in days
          from one date to the other. <em class="replaceable"><code>expr1</code></em>
          and <em class="replaceable"><code>expr2</code></em> are date or date-and-time
          expressions. Only the date parts of the values are used in the
          calculation.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATEDIFF('1997-12-31 23:59:59','1997-12-30');</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT DATEDIFF('1997-11-30 23:59:59','1997-12-31');</code></strong>
        -&gt; -31
</pre></li><li><p><a id="function_date-add"></a>
          <a id="id2834846" class="indexterm"></a>

          <a id="id2834852" class="indexterm"></a>

          <code class="literal">DATE_ADD(<em class="replaceable"><code>date</code></em>,INTERVAL
          <em class="replaceable"><code>expr</code></em>
          <em class="replaceable"><code>unit</code></em>)</code>,
          <code class="literal">DATE_SUB(<em class="replaceable"><code>date</code></em>,INTERVAL
          <em class="replaceable"><code>expr</code></em>
          <em class="replaceable"><code>unit</code></em>)</code>
        </p><p>
          These functions perform date arithmetic.
          <em class="replaceable"><code>date</code></em> is a
          <code class="literal">DATETIME</code> or <code class="literal">DATE</code> value
          specifying the starting date. <em class="replaceable"><code>expr</code></em>
          is an expression specifying the interval value to be added or
          subtracted from the starting date.
          <em class="replaceable"><code>expr</code></em> is a string; it may start with
          a ‘<code class="literal">-</code>’ for negative intervals.
          <em class="replaceable"><code>unit</code></em> is a keyword indicating the
          units in which the expression should be interpreted.
        </p><p>
          The <code class="literal">INTERVAL</code> keyword and the
          <em class="replaceable"><code>unit</code></em> specifier are not case
          sensitive.
        </p><p>
          The following table shows the expected form of the
          <em class="replaceable"><code>expr</code></em> argument for each
          <em class="replaceable"><code>unit</code></em> value.
        </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><tbody><tr><td><em class="replaceable"><code>unit</code></em> <span class="bold"><strong>Value</strong></span></td><td><span class="bold"><strong>Expected</strong></span>
                  <em class="replaceable"><code>expr</code></em>
                  <span class="bold"><strong>Format</strong></span></td></tr><tr><td><code class="literal">MICROSECOND</code></td><td><code class="literal">MICROSECONDS</code></td></tr><tr><td><code class="literal">SECOND</code></td><td><code class="literal">SECONDS</code></td></tr><tr><td><code class="literal">MINUTE</code></td><td><code class="literal">MINUTES</code></td></tr><tr><td><code class="literal">HOUR</code></td><td><code class="literal">HOURS</code></td></tr><tr><td><code class="literal">DAY</code></td><td><code class="literal">DAYS</code></td></tr><tr><td><code class="literal">WEEK</code></td><td><code class="literal">WEEKS</code></td></tr><tr><td><code class="literal">MONTH</code></td><td><code class="literal">MONTHS</code></td></tr><tr><td><code class="literal">QUARTER</code></td><td><code class="literal">QUARTERS</code></td></tr><tr><td><code class="literal">YEAR</code></td><td><code class="literal">YEARS</code></td></tr><tr><td><code class="literal">SECOND_MICROSECOND</code></td><td><code class="literal">'SECONDS.MICROSECONDS'</code></td></tr><tr><td><code class="literal">MINUTE_MICROSECOND</code></td><td><code class="literal">'MINUTES.MICROSECONDS'</code></td></tr><tr><td><code class="literal">MINUTE_SECOND</code></td><td><code class="literal">'MINUTES:SECONDS'</code></td></tr><tr><td><code class="literal">HOUR_MICROSECOND</code></td><td><code class="literal">'HOURS.MICROSECONDS'</code></td></tr><tr><td><code class="literal">HOUR_SECOND</code></td><td><code class="literal">'HOURS:MINUTES:SECONDS'</code></td></tr><tr><td><code class="literal">HOUR_MINUTE</code></td><td><code class="literal">'HOURS:MINUTES'</code></td></tr><tr><td><code class="literal">DAY_MICROSECOND</code></td><td><code class="literal">'DAYS.MICROSECONDS'</code></td></tr><tr><td><code class="literal">DAY_SECOND</code></td><td><code class="literal">'DAYS HOURS:MINUTES:SECONDS'</code></td></tr><tr><td><code class="literal">DAY_MINUTE</code></td><td><code class="literal">'DAYS HOURS:MINUTES'</code></td></tr><tr><td><code class="literal">DAY_HOUR</code></td><td><code class="literal">'DAYS HOURS'</code></td></tr><tr><td><code class="literal">YEAR_MONTH</code></td><td><code class="literal">'YEARS-MONTHS'</code></td></tr></tbody></table></div><p>
          The values <code class="literal">QUARTER</code> and
          <code class="literal">WEEK</code> are available beginning with MySQL
          5.0.0.
        </p><p>
          MySQL allows any punctuation delimiter in the
          <em class="replaceable"><code>expr</code></em> format. Those shown in the
          table are the suggested delimiters. If the
          <em class="replaceable"><code>date</code></em> argument is a
          <code class="literal">DATE</code> value and your calculations involve
          only <code class="literal">YEAR</code>, <code class="literal">MONTH</code>, and
          <code class="literal">DAY</code> parts (that is, no time parts), the
          result is a <code class="literal">DATE</code> value. Otherwise, the
          result is a <code class="literal">DATETIME</code> value.
        </p><p>
          Date arithmetic also can be performed using
          <code class="literal">INTERVAL</code> together with the
          <code class="literal">+</code> or <code class="literal">-</code> operator:
        </p><pre class="programlisting"><code class="literal">date</code> + INTERVAL <em class="replaceable"><code>expr</code></em> <em class="replaceable"><code>unit</code></em>
<code class="literal">date</code> - INTERVAL <em class="replaceable"><code>expr</code></em> <em class="replaceable"><code>unit</code></em>
</pre><p>
          <code class="literal">INTERVAL <em class="replaceable"><code>expr</code></em>
          <em class="replaceable"><code>unit</code></em></code> is allowed on either
          side of the <code class="literal">+</code> operator if the expression on
          the other side is a date or datetime value. For the
          <code class="literal">-</code> operator, <code class="literal">INTERVAL
          <em class="replaceable"><code>expr</code></em>
          <em class="replaceable"><code>unit</code></em></code> is allowed only on
          the right side, because it makes no sense to subtract a date
          or datetime value from an interval.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT '1997-12-31 23:59:59' + INTERVAL 1 SECOND;</code></strong>
        -&gt; '1998-01-01 00:00:00'
mysql&gt; <strong class="userinput"><code>SELECT INTERVAL 1 DAY + '1997-12-31';</code></strong>
        -&gt; '1998-01-01'
mysql&gt; <strong class="userinput"><code>SELECT '1998-01-01' - INTERVAL 1 SECOND;</code></strong>
        -&gt; '1997-12-31 23:59:59'
mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('1997-12-31 23:59:59',</code></strong>
    -&gt;                 <strong class="userinput"><code>INTERVAL 1 SECOND);</code></strong>
        -&gt; '1998-01-01 00:00:00'
mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('1997-12-31 23:59:59',</code></strong>
    -&gt;                 <strong class="userinput"><code>INTERVAL 1 DAY);</code></strong>
        -&gt; '1998-01-01 23:59:59'
mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('1997-12-31 23:59:59',</code></strong>
    -&gt;                 <strong class="userinput"><code>INTERVAL '1:1' MINUTE_SECOND);</code></strong>
        -&gt; '1998-01-01 00:01:00'
mysql&gt; <strong class="userinput"><code>SELECT DATE_SUB('1998-01-01 00:00:00',</code></strong>
    -&gt;                 <strong class="userinput"><code>INTERVAL '1 1:1:1' DAY_SECOND);</code></strong>
        -&gt; '1997-12-30 22:58:59'
mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('1998-01-01 00:00:00',</code></strong>
    -&gt;                 <strong class="userinput"><code>INTERVAL '-1 10' DAY_HOUR);</code></strong>
        -&gt; '1997-12-30 14:00:00'
mysql&gt; <strong class="userinput"><code>SELECT DATE_SUB('1998-01-02', INTERVAL 31 DAY);</code></strong>
        -&gt; '1997-12-02'
mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('1992-12-31 23:59:59.000002',</code></strong>
    -&gt;            <strong class="userinput"><code>INTERVAL '1.999999' SECOND_MICROSECOND);</code></strong>
        -&gt; '1993-01-01 00:00:01.000001'
</pre><p>
          If you specify an interval value that is too short (does not
          include all the interval parts that would be expected from the
          <em class="replaceable"><code>unit</code></em> keyword), MySQL assumes that
          you have left out the leftmost parts of the interval value.
          For example, if you specify a <em class="replaceable"><code>unit</code></em>
          of <code class="literal">DAY_SECOND</code>, the value of
          <em class="replaceable"><code>expr</code></em> is expected to have days,
          hours, minutes, and seconds parts. If you specify a value like
          <code class="literal">'1:10'</code>, MySQL assumes that the days and
          hours parts are missing and the value represents minutes and
          seconds. In other words, <code class="literal">'1:10' DAY_SECOND</code>
          is interpreted in such a way that it is equivalent to
          <code class="literal">'1:10' MINUTE_SECOND</code>. This is analogous to
          the way that MySQL interprets <code class="literal">TIME</code> values
          as representing elapsed time rather than as a time of day.
        </p><p>
          If you add to or subtract from a date value something that
          contains a time part, the result is automatically converted to
          a datetime value:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('1999-01-01', INTERVAL 1 DAY);</code></strong>
        -&gt; '1999-01-02'
mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('1999-01-01', INTERVAL 1 HOUR);</code></strong>
        -&gt; '1999-01-01 01:00:00'
</pre><p>
          If you add <code class="literal">MONTH</code>,
          <code class="literal">YEAR_MONTH</code>, or <code class="literal">YEAR</code> and
          the resulting date has a day that is larger than the maximum
          day for the new month, the day is adjusted to the maximum days
          in the new month:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('1998-01-30', INTERVAL 1 MONTH);</code></strong>
        -&gt; '1998-02-28'
</pre><p>
          Date arithmetic operations require complete dates and do not
          work with incomplete dates such as
          <code class="literal">'2006-07-00'</code> or badly malformed dates:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATE_ADD('2006-07-00', INTERVAL 1 DAY);</code></strong>
        -&gt; NULL
mysql&gt; <strong class="userinput"><code>SELECT '2005-03-32' + INTERVAL 1 MONTH;</code></strong>
        -&gt; NULL
</pre></li><li><p><a id="function_date-format"></a>
          <a id="id2835619" class="indexterm"></a>

          <code class="literal">DATE_FORMAT(<em class="replaceable"><code>date</code></em>,<em class="replaceable"><code>format</code></em>)</code>
        </p><p>
          Formats the <em class="replaceable"><code>date</code></em> value according to
          the <em class="replaceable"><code>format</code></em> string.
        </p><p>
          The following specifiers may be used in the
          <em class="replaceable"><code>format</code></em> string. The
          ‘<code class="literal">%</code>’ character is required
          before format specifier characters.
        </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>Specifier</strong></span></td><td><span class="bold"><strong>Description</strong></span></td></tr><tr><td><code class="literal">%a</code></td><td>Abbreviated weekday name
                  (<code class="literal">Sun</code>..<code class="literal">Sat</code>)</td></tr><tr><td><code class="literal">%b</code></td><td>Abbreviated month name (<code class="literal">Jan</code>..<code class="literal">Dec</code>)</td></tr><tr><td><code class="literal">%c</code></td><td>Month, numeric (<code class="literal">0</code>..<code class="literal">12</code>)</td></tr><tr><td><code class="literal">%D</code></td><td>Day of the month with English suffix (<code class="literal">0th</code>,
                  <code class="literal">1st</code>, <code class="literal">2nd</code>,
                  <code class="literal">3rd</code>, …)</td></tr><tr><td><code class="literal">%d</code></td><td>Day of the month, numeric (<code class="literal">00</code>..<code class="literal">31</code>)</td></tr><tr><td><code class="literal">%e</code></td><td>Day of the month, numeric (<code class="literal">0</code>..<code class="literal">31</code>)</td></tr><tr><td><code class="literal">%f</code></td><td>Microseconds (<code class="literal">000000</code>..<code class="literal">999999</code>)</td></tr><tr><td><code class="literal">%H</code></td><td>Hour (<code class="literal">00</code>..<code class="literal">23</code>)</td></tr><tr><td><code class="literal">%h</code></td><td>Hour (<code class="literal">01</code>..<code class="literal">12</code>)</td></tr><tr><td><code class="literal">%I</code></td><td>Hour (<code class="literal">01</code>..<code class="literal">12</code>)</td></tr><tr><td><code class="literal">%i</code></td><td>Minutes, numeric (<code class="literal">00</code>..<code class="literal">59</code>)</td></tr><tr><td><code class="literal">%j</code></td><td>Day of year (<code class="literal">001</code>..<code class="literal">366</code>)</td></tr><tr><td><code class="literal">%k</code></td><td>Hour (<code class="literal">0</code>..<code class="literal">23</code>)</td></tr><tr><td><code class="literal">%l</code></td><td>Hour (<code class="literal">1</code>..<code class="literal">12</code>)</td></tr><tr><td><code class="literal">%M</code></td><td>Month name (<code class="literal">January</code>..<code class="literal">December</code>)</td></tr><tr><td><code class="literal">%m</code></td><td>Month, numeric (<code class="literal">00</code>..<code class="literal">12</code>)</td></tr><tr><td><code class="literal">%p</code></td><td><code class="literal">AM</code> or <code class="literal">PM</code></td></tr><tr><td><code class="literal">%r</code></td><td>Time, 12-hour (<code class="literal">hh:mm:ss</code> followed by
                  <code class="literal">AM</code> or <code class="literal">PM</code>)</td></tr><tr><td><code class="literal">%S</code></td><td>Seconds (<code class="literal">00</code>..<code class="literal">59</code>)</td></tr><tr><td><code class="literal">%s</code></td><td>Seconds (<code class="literal">00</code>..<code class="literal">59</code>)</td></tr><tr><td><code class="literal">%T</code></td><td>Time, 24-hour (<code class="literal">hh:mm:ss</code>)</td></tr><tr><td><code class="literal">%U</code></td><td>Week (<code class="literal">00</code>..<code class="literal">53</code>), where Sunday is the
                  first day of the week</td></tr><tr><td><code class="literal">%u</code></td><td>Week (<code class="literal">00</code>..<code class="literal">53</code>), where Monday is the
                  first day of the week</td></tr><tr><td><code class="literal">%V</code></td><td>Week (<code class="literal">01</code>..<code class="literal">53</code>), where Sunday is the
                  first day of the week; used with <code class="literal">%X</code></td></tr><tr><td><code class="literal">%v</code></td><td>Week (<code class="literal">01</code>..<code class="literal">53</code>), where Monday is the
                  first day of the week; used with <code class="literal">%x</code></td></tr><tr><td><code class="literal">%W</code></td><td>Weekday name (<code class="literal">Sunday</code>..<code class="literal">Saturday</code>)</td></tr><tr><td><code class="literal">%w</code></td><td>Day of the week
                  (<code class="literal">0</code>=Sunday..<code class="literal">6</code>=Saturday)</td></tr><tr><td><code class="literal">%X</code></td><td>Year for the week where Sunday is the first day of the week, numeric,
                  four digits; used with <code class="literal">%V</code></td></tr><tr><td><code class="literal">%x</code></td><td>Year for the week, where Monday is the first day of the week, numeric,
                  four digits; used with <code class="literal">%v</code></td></tr><tr><td><code class="literal">%Y</code></td><td>Year, numeric, four digits</td></tr><tr><td><code class="literal">%y</code></td><td>Year, numeric (two digits)</td></tr><tr><td><code class="literal">%%</code></td><td>A literal ‘<code class="literal">%</code>’ character</td></tr><tr><td><code class="literal">%<em class="replaceable"><code>x</code></em></code></td><td><em class="replaceable"><code>x</code></em>, for any
                  ‘<em class="replaceable"><code>x</code></em>’ not listed
                  above</td></tr></tbody></table></div><p>
          Ranges for the month and day specifiers begin with zero due to
          the fact that MySQL allows the storing of incomplete dates
          such as <code class="literal">'2004-00-00'</code>.
        </p><p>
          As of MySQL 5.0.25, the language used for day and month names
          and abbreviations is controlled by the value of the
          <code class="literal">lc_time_names</code> system variable
          (<a href="http://dev.mysql.com/doc/refman/5.0/en/locale-support.html" target="_top">MySQL Server Locale Support</a>).
        </p><p>
          As of MySQL 5.0.36, <code class="literal">DATE_FORMAT()</code> returns a
          string with a character set and collation given by
          <code class="literal">character_set_connection</code> and
          <code class="literal">collation_connection</code> so that it can return
          month and weekday names containing non-ASCII characters.
          Before 5.0.36, the return value is a binary string.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATE_FORMAT('1997-10-04 22:23:00', '%W %M %Y');</code></strong>
        -&gt; 'Saturday October 1997'
mysql&gt; <strong class="userinput"><code>SELECT DATE_FORMAT('1997-10-04 22:23:00', '%H:%i:%s');</code></strong>
        -&gt; '22:23:00'
mysql&gt; <strong class="userinput"><code>SELECT DATE_FORMAT('1997-10-04 22:23:00',</code></strong>
                          '%D %y %a %d %m %b %j');
        -&gt; '4th 97 Sat 04 10 Oct 277'
mysql&gt; <strong class="userinput"><code>SELECT DATE_FORMAT('1997-10-04 22:23:00',</code></strong>
                          '%H %k %I %r %T %S %w');
        -&gt; '22 22 10 10:23:00 PM 22:23:00 00 6'
mysql&gt; <strong class="userinput"><code>SELECT DATE_FORMAT('1999-01-01', '%X %V');</code></strong>
        -&gt; '1998 52'
mysql&gt; <strong class="userinput"><code>SELECT DATE_FORMAT('2006-06-00', '%d');</code></strong>
        -&gt; '00'
</pre></li><li><p><a id="function_date-sub"></a>
          <a id="id2836404" class="indexterm"></a>

          <code class="literal">DATE_SUB(<em class="replaceable"><code>date</code></em>,INTERVAL
          <em class="replaceable"><code>expr</code></em>
          <em class="replaceable"><code>unit</code></em>)</code>
        </p><p>
          See <code class="literal">DATE_ADD()</code>.
        </p></li><li><p><a id="function_day"></a>
          <a id="id2836466" class="indexterm"></a>

          <code class="literal">DAY(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          <code class="literal">DAY()</code> is a synonym for
          <code class="literal">DAYOFMONTH()</code>.
        </p></li><li><p><a id="function_dayname"></a>
          <a id="id2836525" class="indexterm"></a>

          <code class="literal">DAYNAME(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the name of the weekday for
          <em class="replaceable"><code>date</code></em>. As of MySQL 5.0.25, the
          language used for the name is controlled by the value of the
          <code class="literal">lc_time_names</code> system variable
          (<a href="http://dev.mysql.com/doc/refman/5.0/en/locale-support.html" target="_top">MySQL Server Locale Support</a>).
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DAYNAME('1998-02-05');</code></strong>
        -&gt; 'Thursday'
</pre></li><li><p><a id="function_dayofmonth"></a>
          <a id="id2836606" class="indexterm"></a>

          <code class="literal">DAYOFMONTH(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the day of the month for
          <em class="replaceable"><code>date</code></em>, in the range
          <code class="literal">0</code> to <code class="literal">31</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DAYOFMONTH('1998-02-03');</code></strong>
        -&gt; 3
</pre></li><li><p><a id="function_dayofweek"></a>
          <a id="id2836681" class="indexterm"></a>

          <code class="literal">DAYOFWEEK(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the weekday index for <em class="replaceable"><code>date</code></em>
          (<code class="literal">1</code> = Sunday, <code class="literal">2</code> = Monday,
          …, <code class="literal">7</code> = Saturday). These index values
          correspond to the ODBC standard.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DAYOFWEEK('1998-02-03');</code></strong>
        -&gt; 3
</pre></li><li><p><a id="function_dayofyear"></a>
          <a id="id2836762" class="indexterm"></a>

          <code class="literal">DAYOFYEAR(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the day of the year for
          <em class="replaceable"><code>date</code></em>, in the range
          <code class="literal">1</code> to <code class="literal">366</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DAYOFYEAR('1998-02-03');</code></strong>
        -&gt; 34
</pre></li><li><p><a id="function_extract"></a>
          <a id="id2836837" class="indexterm"></a>

          <code class="literal">EXTRACT(<em class="replaceable"><code>unit</code></em> FROM
          <em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          The <code class="literal">EXTRACT()</code> function uses the same kinds
          of unit specifiers as <code class="literal">DATE_ADD()</code> or
          <code class="literal">DATE_SUB()</code>, but extracts parts from the
          date rather than performing date arithmetic.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT EXTRACT(YEAR FROM '1999-07-02');</code></strong>
       -&gt; 1999
mysql&gt; <strong class="userinput"><code>SELECT EXTRACT(YEAR_MONTH FROM '1999-07-02 01:02:03');</code></strong>
       -&gt; 199907
mysql&gt; <strong class="userinput"><code>SELECT EXTRACT(DAY_MINUTE FROM '1999-07-02 01:02:03');</code></strong>
       -&gt; 20102
mysql&gt; <strong class="userinput"><code>SELECT EXTRACT(MICROSECOND</code></strong>
    -&gt;                <strong class="userinput"><code>FROM '2003-01-02 10:30:00.000123');</code></strong>
        -&gt; 123
</pre></li><li><p><a id="function_from-days"></a>
          <a id="id2836939" class="indexterm"></a>

          <code class="literal">FROM_DAYS(<em class="replaceable"><code>N</code></em>)</code>
        </p><p>
          Given a day number <em class="replaceable"><code>N</code></em>, returns a
          <code class="literal">DATE</code> value.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT FROM_DAYS(729669);</code></strong>
        -&gt; '1997-10-07'
</pre><p>
          Use <code class="literal">FROM_DAYS()</code> with caution on old dates.
          It is not intended for use with values that precede the advent
          of the Gregorian calendar (1582). See
          <a href="mysqlqb_functions.html#mysql-calendar" title="What Calendar Is Used By MySQL?">What Calendar Is Used By MySQL?</a>.
        </p></li><li><p><a id="function_from-unixtime"></a>
          <a id="id2837026" class="indexterm"></a>

          <code class="literal">FROM_UNIXTIME(<em class="replaceable"><code>unix_timestamp</code></em>)</code>,
          <code class="literal">FROM_UNIXTIME(<em class="replaceable"><code>unix_timestamp</code></em>,<em class="replaceable"><code>format</code></em>)</code>
        </p><p>
          Returns a representation of the
          <em class="replaceable"><code>unix_timestamp</code></em> argument as a value
          in <code class="literal">'YYYY-MM-DD HH:MM:SS'</code> or
          <code class="literal">YYYYMMDDHHMMSS</code> format, depending on whether
          the function is used in a string or numeric context. The value
          is expressed in the current time zone.
          <em class="replaceable"><code>unix_timestamp</code></em> is an internal
          timestamp value such as is produced by the
          <code class="literal">UNIX_TIMESTAMP()</code> function.
        </p><p>
          If <em class="replaceable"><code>format</code></em> is given, the result is
          formatted according to the <em class="replaceable"><code>format</code></em>
          string, which is used the same way as listed in the entry for
          the <code class="literal">DATE_FORMAT()</code> function.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT FROM_UNIXTIME(875996580);</code></strong>
        -&gt; '1997-10-04 22:23:00'
mysql&gt; <strong class="userinput"><code>SELECT FROM_UNIXTIME(875996580) + 0;</code></strong>
        -&gt; 19971004222300
mysql&gt; <strong class="userinput"><code>SELECT FROM_UNIXTIME(UNIX_TIMESTAMP(),</code></strong>
    -&gt;                      <strong class="userinput"><code>'%Y %D %M %h:%i:%s %x');</code></strong>
        -&gt; '2003 6th August 06:22:58 2003'
</pre><p>
          Note: If you use <code class="literal">UNIX_TIMESTAMP()</code> and
          <code class="literal">FROM_UNIXTIME()</code> to convert between
          <code class="literal">TIMESTAMP</code> values and Unix timestamp values,
          the conversion is lossy because the mapping is not one-to-one
          in both directions. For details, see the description of the
          <code class="literal">UNIX_TIMESTAMP()</code> function.
        </p></li><li><p><a id="function_get-format"></a>
          <a id="id2837184" class="indexterm"></a>

          <code class="literal">GET_FORMAT(DATE|TIME|DATETIME,
          'EUR'|'USA'|'JIS'|'ISO'|'INTERNAL')</code>
        </p><p>
          Returns a format string. This function is useful in
          combination with the <code class="literal">DATE_FORMAT()</code> and the
          <code class="literal">STR_TO_DATE()</code> functions.
        </p><p>
          The possible values for the first and second arguments result
          in several possible format strings (for the specifiers used,
          see the table in the <code class="literal">DATE_FORMAT()</code> function
          description). ISO format refers to ISO 9075, not ISO 8601.
        </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>Function Call</strong></span></td><td><span class="bold"><strong>Result</strong></span></td></tr><tr><td><code class="literal">GET_FORMAT(DATE,'USA')</code></td><td><code class="literal">'%m.%d.%Y'</code></td></tr><tr><td><code class="literal">GET_FORMAT(DATE,'JIS')</code></td><td><code class="literal">'%Y-%m-%d'</code></td></tr><tr><td><code class="literal">GET_FORMAT(DATE,'ISO')</code></td><td><code class="literal">'%Y-%m-%d'</code></td></tr><tr><td><code class="literal">GET_FORMAT(DATE,'EUR')</code></td><td><code class="literal">'%d.%m.%Y'</code></td></tr><tr><td><code class="literal">GET_FORMAT(DATE,'INTERNAL')</code></td><td><code class="literal">'%Y%m%d'</code></td></tr><tr><td><code class="literal">GET_FORMAT(DATETIME,'USA')</code></td><td><code class="literal">'%Y-%m-%d %H.%i.%s'</code></td></tr><tr><td><code class="literal">GET_FORMAT(DATETIME,'JIS')</code></td><td><code class="literal">'%Y-%m-%d %H:%i:%s'</code></td></tr><tr><td><code class="literal">GET_FORMAT(DATETIME,'ISO')</code></td><td><code class="literal">'%Y-%m-%d %H:%i:%s'</code></td></tr><tr><td><code class="literal">GET_FORMAT(DATETIME,'EUR')</code></td><td><code class="literal">'%Y-%m-%d %H.%i.%s'</code></td></tr><tr><td><code class="literal">GET_FORMAT(DATETIME,'INTERNAL')</code></td><td><code class="literal">'%Y%m%d%H%i%s'</code></td></tr><tr><td><code class="literal">GET_FORMAT(TIME,'USA')</code></td><td><code class="literal">'%h:%i:%s %p'</code></td></tr><tr><td><code class="literal">GET_FORMAT(TIME,'JIS')</code></td><td><code class="literal">'%H:%i:%s'</code></td></tr><tr><td><code class="literal">GET_FORMAT(TIME,'ISO')</code></td><td><code class="literal">'%H:%i:%s'</code></td></tr><tr><td><code class="literal">GET_FORMAT(TIME,'EUR')</code></td><td><code class="literal">'%H.%i.%s'</code></td></tr><tr><td><code class="literal">GET_FORMAT(TIME,'INTERNAL')</code></td><td><code class="literal">'%H%i%s'</code></td></tr></tbody></table></div><p>
          <code class="literal">TIMESTAMP</code> can also be used as the first
          argument to <code class="literal">GET_FORMAT()</code>, in which case the
          function returns the same values as for
          <code class="literal">DATETIME</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATE_FORMAT('2003-10-03',GET_FORMAT(DATE,'EUR'));</code></strong>
        -&gt; '03.10.2003'
mysql&gt; <strong class="userinput"><code>SELECT STR_TO_DATE('10.31.2003',GET_FORMAT(DATE,'USA'));</code></strong>
        -&gt; '2003-10-31'
</pre></li><li><p><a id="function_hour"></a>
          <a id="id2837532" class="indexterm"></a>

          <code class="literal">HOUR(<em class="replaceable"><code>time</code></em>)</code>
        </p><p>
          Returns the hour for <em class="replaceable"><code>time</code></em>. The
          range of the return value is <code class="literal">0</code> to
          <code class="literal">23</code> for time-of-day values. However, the
          range of <code class="literal">TIME</code> values actually is much
          larger, so <code class="literal">HOUR</code> can return values greater
          than <code class="literal">23</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT HOUR('10:05:03');</code></strong>
        -&gt; 10
mysql&gt; <strong class="userinput"><code>SELECT HOUR('272:59:59');</code></strong>
        -&gt; 272
</pre></li><li><p><a id="function_last-day"></a>
          <a id="id2837624" class="indexterm"></a>

          <code class="literal">LAST_DAY(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Takes a date or datetime value and returns the corresponding
          value for the last day of the month. Returns
          <code class="literal">NULL</code> if the argument is invalid.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LAST_DAY('2003-02-05');</code></strong>
        -&gt; '2003-02-28'
mysql&gt; <strong class="userinput"><code>SELECT LAST_DAY('2004-02-05');</code></strong>
        -&gt; '2004-02-29'
mysql&gt; <strong class="userinput"><code>SELECT LAST_DAY('2004-01-01 01:01:01');</code></strong>
        -&gt; '2004-01-31'
mysql&gt; <strong class="userinput"><code>SELECT LAST_DAY('2003-03-32');</code></strong>
        -&gt; NULL
</pre></li><li><p><a id="function_localtime"></a>
          <a id="id2837711" class="indexterm"></a>

          <code class="literal">LOCALTIME</code>, <code class="literal">LOCALTIME()</code>
        </p><p>
          <code class="literal">LOCALTIME</code> and
          <code class="literal">LOCALTIME()</code> are synonyms for
          <code class="literal">NOW()</code>.
        </p></li><li><p><a id="function_localtimestamp"></a>
          <a id="id2837774" class="indexterm"></a>

          <code class="literal">LOCALTIMESTAMP</code>,
          <code class="literal">LOCALTIMESTAMP()</code>
        </p><p>
          <code class="literal">LOCALTIMESTAMP</code> and
          <code class="literal">LOCALTIMESTAMP()</code> are synonyms for
          <code class="literal">NOW()</code>.
        </p></li><li><p><a id="function_makedate"></a>
          <a id="id2837837" class="indexterm"></a>

          <code class="literal">MAKEDATE(<em class="replaceable"><code>year</code></em>,<em class="replaceable"><code>dayofyear</code></em>)</code>
        </p><p>
          Returns a date, given year and day-of-year values.
          <em class="replaceable"><code>dayofyear</code></em> must be greater than 0 or
          the result is <code class="literal">NULL</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MAKEDATE(2001,31), MAKEDATE(2001,32);</code></strong>
        -&gt; '2001-01-31', '2001-02-01'
mysql&gt; <strong class="userinput"><code>SELECT MAKEDATE(2001,365), MAKEDATE(2004,365);</code></strong>
        -&gt; '2001-12-31', '2004-12-30'
mysql&gt; <strong class="userinput"><code>SELECT MAKEDATE(2001,0);</code></strong>
        -&gt; NULL
</pre></li><li><p><a id="function_maketime"></a>
          <a id="id2837925" class="indexterm"></a>

          <code class="literal">MAKETIME(<em class="replaceable"><code>hour</code></em>,<em class="replaceable"><code>minute</code></em>,<em class="replaceable"><code>second</code></em>)</code>
        </p><p>
          Returns a time value calculated from the
          <em class="replaceable"><code>hour</code></em>,
          <em class="replaceable"><code>minute</code></em>, and
          <em class="replaceable"><code>second</code></em> arguments.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MAKETIME(12,15,30);</code></strong>
        -&gt; '12:15:30'
</pre></li><li><p><a id="function_microsecond"></a>
          <a id="id2838008" class="indexterm"></a>

          <code class="literal">MICROSECOND(<em class="replaceable"><code>expr</code></em>)</code>
        </p><p>
          Returns the microseconds from the time or datetime expression
          <em class="replaceable"><code>expr</code></em> as a number in the range from
          <code class="literal">0</code> to <code class="literal">999999</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MICROSECOND('12:00:00.123456');</code></strong>
        -&gt; 123456
mysql&gt; <strong class="userinput"><code>SELECT MICROSECOND('1997-12-31 23:59:59.000010');</code></strong>
        -&gt; 10
</pre></li><li><p><a id="function_minute"></a>
          <a id="id2838090" class="indexterm"></a>

          <code class="literal">MINUTE(<em class="replaceable"><code>time</code></em>)</code>
        </p><p>
          Returns the minute for <em class="replaceable"><code>time</code></em>, in the
          range <code class="literal">0</code> to <code class="literal">59</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MINUTE('98-02-03 10:05:03');</code></strong>
        -&gt; 5
</pre></li><li><p><a id="function_month"></a>
          <a id="id2838165" class="indexterm"></a>

          <code class="literal">MONTH(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the month for <em class="replaceable"><code>date</code></em>, in the
          range <code class="literal">0</code> to <code class="literal">12</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MONTH('1998-02-03');</code></strong>
        -&gt; 2
</pre></li><li><p><a id="function_monthname"></a>
          <a id="id2838240" class="indexterm"></a>

          <code class="literal">MONTHNAME(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the full name of the month for
          <em class="replaceable"><code>date</code></em>. As of MySQL 5.0.25, the
          language used for the name is controlled by the value of the
          <code class="literal">lc_time_names</code> system variable
          (<a href="http://dev.mysql.com/doc/refman/5.0/en/locale-support.html" target="_top">MySQL Server Locale Support</a>).
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MONTHNAME('1998-02-05');</code></strong>
        -&gt; 'February'
</pre></li><li><p><a id="function_now"></a>
          <a id="id2838321" class="indexterm"></a>

          <code class="literal">NOW()</code>
        </p><p>
          Returns the current date and time as a value in
          <code class="literal">'YYYY-MM-DD HH:MM:SS'</code> or
          <code class="literal">YYYYMMDDHHMMSS</code> format, depending on whether
          the function is used in a string or numeric context. The value
          is expressed in the current time zone.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT NOW();</code></strong>
        -&gt; '1997-12-15 23:50:26'
mysql&gt; <strong class="userinput"><code>SELECT NOW() + 0;</code></strong>
        -&gt; 19971215235026
</pre><p>
          <code class="literal">NOW()</code> returns a constant time that
          indicates the time at which the statement began to execute.
          (Within a stored routine or trigger, <code class="literal">NOW()</code>
          returns the time at which the routine or triggering statement
          began to execute.) This differs from the behavior for
          <code class="literal">SYSDATE()</code>, which returns the exact time at
          which it executes as of MySQL 5.0.13.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT NOW(), SLEEP(2), NOW();</code></strong>
+---------------------+----------+---------------------+
| NOW()               | SLEEP(2) | NOW()               |
+---------------------+----------+---------------------+
| 2006-04-12 13:47:36 |        0 | 2006-04-12 13:47:36 |
+---------------------+----------+---------------------+

mysql&gt; <strong class="userinput"><code>SELECT SYSDATE(), SLEEP(2), SYSDATE();</code></strong>
+---------------------+----------+---------------------+
| SYSDATE()           | SLEEP(2) | SYSDATE()           |
+---------------------+----------+---------------------+
| 2006-04-12 13:47:44 |        0 | 2006-04-12 13:47:46 |
+---------------------+----------+---------------------+
</pre><p>
          See the description for <code class="literal">SYSDATE()</code> for
          additional information about the differences between the two
          functions.
        </p></li><li><p><a id="function_period-add"></a>
          <a id="id2838454" class="indexterm"></a>

          <code class="literal">PERIOD_ADD(<em class="replaceable"><code>P</code></em>,<em class="replaceable"><code>N</code></em>)</code>
        </p><p>
          Adds <em class="replaceable"><code>N</code></em> months to period
          <em class="replaceable"><code>P</code></em> (in the format
          <code class="literal">YYMM</code> or <code class="literal">YYYYMM</code>). Returns
          a value in the format <code class="literal">YYYYMM</code>. Note that the
          period argument <em class="replaceable"><code>P</code></em> is
          <span class="emphasis"><em>not</em></span> a date value.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT PERIOD_ADD(9801,2);</code></strong>
        -&gt; 199803
</pre></li><li><p><a id="function_period-diff"></a>
          <a id="id2838549" class="indexterm"></a>

          <code class="literal">PERIOD_DIFF(<em class="replaceable"><code>P1</code></em>,<em class="replaceable"><code>P2</code></em>)</code>
        </p><p>
          Returns the number of months between periods
          <em class="replaceable"><code>P1</code></em> and
          <em class="replaceable"><code>P2</code></em>. <em class="replaceable"><code>P1</code></em>
          and <em class="replaceable"><code>P2</code></em> should be in the format
          <code class="literal">YYMM</code> or <code class="literal">YYYYMM</code>. Note
          that the period arguments <em class="replaceable"><code>P1</code></em> and
          <em class="replaceable"><code>P2</code></em> are <span class="emphasis"><em>not</em></span>
          date values.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT PERIOD_DIFF(9802,199703);</code></strong>
        -&gt; 11
</pre></li><li><p><a id="function_quarter"></a>
          <a id="id2838650" class="indexterm"></a>

          <code class="literal">QUARTER(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the quarter of the year for
          <em class="replaceable"><code>date</code></em>, in the range
          <code class="literal">1</code> to <code class="literal">4</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT QUARTER('98-04-01');</code></strong>
        -&gt; 2
</pre></li><li><p><a id="function_second"></a>
          <a id="id2838725" class="indexterm"></a>

          <code class="literal">SECOND(<em class="replaceable"><code>time</code></em>)</code>
        </p><p>
          Returns the second for <em class="replaceable"><code>time</code></em>, in the
          range <code class="literal">0</code> to <code class="literal">59</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SECOND('10:05:03');</code></strong>
        -&gt; 3
</pre></li><li><p><a id="function_sec-to-time"></a>
          <a id="id2838800" class="indexterm"></a>

          <code class="literal">SEC_TO_TIME(<em class="replaceable"><code>seconds</code></em>)</code>
        </p><p>
          Returns the <em class="replaceable"><code>seconds</code></em> argument,
          converted to hours, minutes, and seconds, as a value in
          <code class="literal">'HH:MM:SS'</code> or <code class="literal">HHMMSS</code>
          format, depending on whether the function is used in a string
          or numeric context.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SEC_TO_TIME(2378);</code></strong>
        -&gt; '00:39:38'
mysql&gt; <strong class="userinput"><code>SELECT SEC_TO_TIME(2378) + 0;</code></strong>
        -&gt; 3938
</pre></li><li><p><a id="function_str-to-date"></a>
          <a id="id2838884" class="indexterm"></a>

          <code class="literal">STR_TO_DATE(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>format</code></em>)</code>
        </p><p>
          This is the inverse of the <code class="literal">DATE_FORMAT()</code>
          function. It takes a string <em class="replaceable"><code>str</code></em> and
          a format string <em class="replaceable"><code>format</code></em>.
          <code class="literal">STR_TO_DATE()</code> returns a
          <code class="literal">DATETIME</code> value if the format string
          contains both date and time parts, or a
          <code class="literal">DATE</code> or <code class="literal">TIME</code> value if
          the string contains only date or time parts.
        </p><p>
          The date, time, or datetime values contained in
          <em class="replaceable"><code>str</code></em> should be given in the format
          indicated by <em class="replaceable"><code>format</code></em>. For the
          specifiers that can be used in
          <em class="replaceable"><code>format</code></em>, see the
          <code class="literal">DATE_FORMAT()</code> function description. If
          <em class="replaceable"><code>str</code></em> contains an illegal date, time,
          or datetime value, <code class="literal">STR_TO_DATE()</code> returns
          <code class="literal">NULL</code>. Starting from MySQL 5.0.3, an illegal
          value also produces a warning.
        </p><p>
          Range checking on the parts of date values is as described in
          <a href="http://dev.mysql.com/doc/refman/5.0/en/datetime.html" target="_top">The <code class="literal">DATETIME</code>, <code class="literal">DATE</code>, and  <code class="literal">TIMESTAMP</code> Types</a>. This means, for example, that
          “<span class="quote">zero</span>” dates or dates with part values of 0 are
          allowed unless the SQL mode is set to disallow such values.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT STR_TO_DATE('00/00/0000', '%m/%d/%Y');</code></strong>
        -&gt; '0000-00-00'
mysql&gt; <strong class="userinput"><code>SELECT STR_TO_DATE('04/31/2004', '%m/%d/%Y');</code></strong>
        -&gt; '2004-04-31'
</pre><p>
          <span class="bold"><strong>Note</strong></span>: You cannot use format
          <code class="literal">"%X%V"</code> to convert a year-week string to a
          date because the combination of a year and week does not
          uniquely identify a year and month if the week crosses a month
          boundary. To convert a year-week to a date, then you should
          also specify the weekday:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT STR_TO_DATE('200442 Monday', '%X%V %W');</code></strong>
        -&gt; '2004-10-18'
</pre></li><li><p><a id="function_subdate"></a>
          <a id="id2839101" class="indexterm"></a>

          <code class="literal">SUBDATE(<em class="replaceable"><code>date</code></em>,INTERVAL
          <em class="replaceable"><code>expr</code></em>
          <em class="replaceable"><code>unit</code></em>)</code>,
          <code class="literal">SUBDATE(<em class="replaceable"><code>expr</code></em>,<em class="replaceable"><code>days</code></em>)</code>
        </p><p>
          When invoked with the <code class="literal">INTERVAL</code> form of the
          second argument, <code class="literal">SUBDATE()</code> is a synonym for
          <code class="literal">DATE_SUB()</code>. For information on the
          <code class="literal">INTERVAL</code> <em class="replaceable"><code>unit</code></em>
          argument, see the discussion for
          <code class="literal">DATE_ADD()</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATE_SUB('1998-01-02', INTERVAL 31 DAY);</code></strong>
        -&gt; '1997-12-02'
mysql&gt; <strong class="userinput"><code>SELECT SUBDATE('1998-01-02', INTERVAL 31 DAY);</code></strong>
        -&gt; '1997-12-02'
</pre><p>
          The second form allows the use of an integer value for
          <em class="replaceable"><code>days</code></em>. In such cases, it is
          interpreted as the number of days to be subtracted from the
          date or datetime expression <em class="replaceable"><code>expr</code></em>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SUBDATE('1998-01-02 12:00:00', 31);</code></strong>
        -&gt; '1997-12-02 12:00:00'
</pre></li><li><p><a id="function_subtime"></a>
          <a id="id2839233" class="indexterm"></a>

          <code class="literal">SUBTIME(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>)</code>
        </p><p>
          <code class="literal">SUBTIME()</code> returns
          <em class="replaceable"><code>expr1</code></em> –
          <em class="replaceable"><code>expr2</code></em> expressed as a value in the
          same format as <em class="replaceable"><code>expr1</code></em>.
          <em class="replaceable"><code>expr1</code></em> is a time or datetime
          expression, and <em class="replaceable"><code>expr2</code></em> is a time
          expression.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SUBTIME('1997-12-31 23:59:59.999999','1 1:1:1.000002');</code></strong>
        -&gt; '1997-12-30 22:58:58.999997'
mysql&gt; <strong class="userinput"><code>SELECT SUBTIME('01:00:00.999999', '02:00:00.999998');</code></strong>
        -&gt; '-00:59:59.999999'
</pre></li><li><p><a id="function_sysdate"></a>
          <a id="id2839340" class="indexterm"></a>

          <code class="literal">SYSDATE()</code>
        </p><p>
          Returns the current date and time as a value in
          <code class="literal">'YYYY-MM-DD HH:MM:SS'</code> or
          <code class="literal">YYYYMMDDHHMMSS</code> format, depending on whether
          the function is used in a string or numeric context.
        </p><p>
          As of MySQL 5.0.13, <code class="literal">SYSDATE()</code> returns the
          time at which it executes. This differs from the behavior for
          <code class="literal">NOW()</code>, which returns a constant time that
          indicates the time at which the statement began to execute.
          (Within a stored routine or trigger, <code class="literal">NOW()</code>
          returns the time at which the routine or triggering statement
          began to execute.)
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT NOW(), SLEEP(2), NOW();</code></strong>
+---------------------+----------+---------------------+
| NOW()               | SLEEP(2) | NOW()               |
+---------------------+----------+---------------------+
| 2006-04-12 13:47:36 |        0 | 2006-04-12 13:47:36 |
+---------------------+----------+---------------------+

mysql&gt; <strong class="userinput"><code>SELECT SYSDATE(), SLEEP(2), SYSDATE();</code></strong>
+---------------------+----------+---------------------+
| SYSDATE()           | SLEEP(2) | SYSDATE()           |
+---------------------+----------+---------------------+
| 2006-04-12 13:47:44 |        0 | 2006-04-12 13:47:46 |
+---------------------+----------+---------------------+
</pre><p>
          In addition, the <code class="literal">SET TIMESTAMP</code> statement
          affects the value returned by <code class="literal">NOW()</code> but not
          by <code class="literal">SYSDATE()</code>. This means that timestamp
          settings in the binary log have no effect on invocations of
          <code class="literal">SYSDATE()</code>.
        </p><p>
          Because <code class="literal">SYSDATE()</code> can return different
          values even within the same statement, and is not affected by
          <code class="literal">SET TIMESTAMP</code>, it is non-deterministic and
          therefore unsafe for replication. If that is a problem, you
          can start the server with the
          <code class="option">--sysdate-is-now</code> option to cause
          <code class="literal">SYSDATE()</code> to be an alias for
          <code class="literal">NOW()</code>.
        </p></li><li><p><a id="function_time"></a>
          <a id="id2839500" class="indexterm"></a>

          <code class="literal">TIME(<em class="replaceable"><code>expr</code></em>)</code>
        </p><p>
          Extracts the time part of the time or datetime expression
          <em class="replaceable"><code>expr</code></em> and returns it as a string.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TIME('2003-12-31 01:02:03');</code></strong>
        -&gt; '01:02:03'
mysql&gt; <strong class="userinput"><code>SELECT TIME('2003-12-31 01:02:03.000123');</code></strong>
        -&gt; '01:02:03.000123'
</pre></li><li><p><a id="function_timediff"></a>
          <a id="id2839575" class="indexterm"></a>

          <code class="literal">TIMEDIFF(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>)</code>
        </p><p>
          <code class="literal">TIMEDIFF()</code> returns
          <em class="replaceable"><code>expr1</code></em> –
          <em class="replaceable"><code>expr2</code></em> expressed as a time value.
          <em class="replaceable"><code>expr1</code></em> and
          <em class="replaceable"><code>expr2</code></em> are time or date-and-time
          expressions, but both must be of the same type.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TIMEDIFF('2000:01:01 00:00:00',</code></strong>
    -&gt;                 <strong class="userinput"><code>'2000:01:01 00:00:00.000001');</code></strong>
        -&gt; '-00:00:00.000001'
mysql&gt; <strong class="userinput"><code>SELECT TIMEDIFF('1997-12-31 23:59:59.000001',</code></strong>
    -&gt;                 <strong class="userinput"><code>'1997-12-30 01:01:01.000002');</code></strong>
        -&gt; '46:58:57.999999'
</pre></li><li><p><a id="function_timestamp"></a>
          <a id="id2839693" class="indexterm"></a>

          <code class="literal">TIMESTAMP(<em class="replaceable"><code>expr</code></em>)</code>,
          <code class="literal">TIMESTAMP(<em class="replaceable"><code>expr1</code></em>,<em class="replaceable"><code>expr2</code></em>)</code>
        </p><p>
          With a single argument, this function returns the date or
          datetime expression <em class="replaceable"><code>expr</code></em> as a
          datetime value. With two arguments, it adds the time
          expression <em class="replaceable"><code>expr2</code></em> to the date or
          datetime expression <em class="replaceable"><code>expr1</code></em> and
          returns the result as a datetime value.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TIMESTAMP('2003-12-31');</code></strong>
        -&gt; '2003-12-31 00:00:00'
mysql&gt; <strong class="userinput"><code>SELECT TIMESTAMP('2003-12-31 12:00:00','12:00:00');</code></strong>
        -&gt; '2004-01-01 00:00:00'
</pre></li><li><p><a id="function_timestampadd"></a>
          <a id="id2839791" class="indexterm"></a>

          <code class="literal">TIMESTAMPADD(<em class="replaceable"><code>unit</code></em>,<em class="replaceable"><code>interval</code></em>,<em class="replaceable"><code>datetime_expr</code></em>)</code>
        </p><p>
          Adds the integer expression
          <em class="replaceable"><code>interval</code></em> to the date or datetime
          expression <em class="replaceable"><code>datetime_expr</code></em>. The unit
          for <em class="replaceable"><code>interval</code></em> is given by the
          <em class="replaceable"><code>unit</code></em> argument, which should be one
          of the following values: <code class="literal">FRAC_SECOND</code>,
          <code class="literal">SECOND</code>, <code class="literal">MINUTE</code>,
          <code class="literal">HOUR</code>, <code class="literal">DAY</code>,
          <code class="literal">WEEK</code>, <code class="literal">MONTH</code>,
          <code class="literal">QUARTER</code>, or <code class="literal">YEAR</code>.
        </p><p>
          The <em class="replaceable"><code>unit</code></em> value may be specified
          using one of keywords as shown, or with a prefix of
          <code class="literal">SQL_TSI_</code>. For example,
          <code class="literal">DAY</code> and <code class="literal">SQL_TSI_DAY</code> both
          are legal.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TIMESTAMPADD(MINUTE,1,'2003-01-02');</code></strong>
        -&gt; '2003-01-02 00:01:00'
mysql&gt; <strong class="userinput"><code>SELECT TIMESTAMPADD(WEEK,1,'2003-01-02');</code></strong>
        -&gt; '2003-01-09'
</pre><p>
          <code class="literal">TIMESTAMPADD()</code> is available as of MySQL
          5.0.0.
        </p></li><li><p><a id="function_timestampdiff"></a>
          <a id="id2839948" class="indexterm"></a>

          <code class="literal">TIMESTAMPDIFF(<em class="replaceable"><code>unit</code></em>,<em class="replaceable"><code>datetime_expr1</code></em>,<em class="replaceable"><code>datetime_expr2</code></em>)</code>
        </p><p>
          Returns the integer difference between the date or datetime
          expressions <em class="replaceable"><code>datetime_expr1</code></em> and
          <em class="replaceable"><code>datetime_expr2</code></em>. The unit for the
          result is given by the <em class="replaceable"><code>unit</code></em>
          argument. The legal values for <em class="replaceable"><code>unit</code></em>
          are the same as those listed in the description of the
          <code class="literal">TIMESTAMPADD()</code> function.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TIMESTAMPDIFF(MONTH,'2003-02-01','2003-05-01');</code></strong>
        -&gt; 3
mysql&gt; <strong class="userinput"><code>SELECT TIMESTAMPDIFF(YEAR,'2002-05-01','2001-01-01');</code></strong>
        -&gt; -1
</pre><p>
          <code class="literal">TIMESTAMPDIFF()</code> is available as of MySQL
          5.0.0.
        </p></li><li><p><a id="function_time-format"></a>
          <a id="id2840055" class="indexterm"></a>

          <code class="literal">TIME_FORMAT(<em class="replaceable"><code>time</code></em>,<em class="replaceable"><code>format</code></em>)</code>
        </p><p>
          This is used like the <code class="literal">DATE_FORMAT()</code>
          function, but the <em class="replaceable"><code>format</code></em> string may
          contain format specifiers only for hours, minutes, and
          seconds. Other specifiers produce a <code class="literal">NULL</code>
          value or <code class="literal">0</code>.
        </p><p>
          If the <em class="replaceable"><code>time</code></em> value contains an hour
          part that is greater than <code class="literal">23</code>, the
          <code class="literal">%H</code> and <code class="literal">%k</code> hour format
          specifiers produce a value larger than the usual range of
          <code class="literal">0..23</code>. The other hour format specifiers
          produce the hour value modulo 12.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TIME_FORMAT('100:00:00', '%H %k %h %I %l');</code></strong>
        -&gt; '100 100 04 04 4'
</pre></li><li><p><a id="function_time-to-sec"></a>
          <a id="id2840167" class="indexterm"></a>

          <code class="literal">TIME_TO_SEC(<em class="replaceable"><code>time</code></em>)</code>
        </p><p>
          Returns the <em class="replaceable"><code>time</code></em> argument,
          converted to seconds.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TIME_TO_SEC('22:23:00');</code></strong>
        -&gt; 80580
mysql&gt; <strong class="userinput"><code>SELECT TIME_TO_SEC('00:39:38');</code></strong>
        -&gt; 2378
</pre></li><li><p><a id="function_to-days"></a>
          <a id="id2840240" class="indexterm"></a>

          <code class="literal">TO_DAYS(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Given a date <em class="replaceable"><code>date</code></em>, returns a day
          number (the number of days since year 0).
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TO_DAYS(950501);</code></strong>
        -&gt; 728779
mysql&gt; <strong class="userinput"><code>SELECT TO_DAYS('1997-10-07');</code></strong>
        -&gt; 729669
</pre><p>
          <code class="literal">TO_DAYS()</code> is not intended for use with
          values that precede the advent of the Gregorian calendar
          (1582), because it does not take into account the days that
          were lost when the calendar was changed. For dates before 1582
          (and possibly a later year in other locales), results from
          this function are not reliable. See
          <a href="mysqlqb_functions.html#mysql-calendar" title="What Calendar Is Used By MySQL?">What Calendar Is Used By MySQL?</a>, for details.
        </p><p>
          Remember that MySQL converts two-digit year values in dates to
          four-digit form using the rules in
          <a href="http://dev.mysql.com/doc/refman/5.0/en/date-and-time-types.html" target="_top">Date and Time Types</a>. For example,
          <code class="literal">'1997-10-07'</code> and
          <code class="literal">'97-10-07'</code> are seen as identical dates:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT TO_DAYS('1997-10-07'), TO_DAYS('97-10-07');</code></strong>
        -&gt; 729669, 729669
</pre></li><li><p><a id="function_unix-timestamp"></a>
          <a id="id2840362" class="indexterm"></a>

          <code class="literal">UNIX_TIMESTAMP()</code>,
          <code class="literal">UNIX_TIMESTAMP(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          If called with no argument, returns a Unix timestamp (seconds
          since <code class="literal">'1970-01-01 00:00:00'</code> UTC) as an
          unsigned integer. If <code class="literal">UNIX_TIMESTAMP()</code> is
          called with a <em class="replaceable"><code>date</code></em> argument, it
          returns the value of the argument as seconds since
          <code class="literal">'1970-01-01 00:00:00'</code> UTC.
          <em class="replaceable"><code>date</code></em> may be a
          <code class="literal">DATE</code> string, a <code class="literal">DATETIME</code>
          string, a <code class="literal">TIMESTAMP</code>, or a number in the
          format <code class="literal">YYMMDD</code> or
          <code class="literal">YYYYMMDD</code>. The server interprets
          <em class="replaceable"><code>date</code></em> as a value in the current time
          zone and converts it to an internal value in UTC. Clients can
          set their time zone as described in
          <a href="http://dev.mysql.com/doc/refman/5.0/en/time-zone-support.html" target="_top">MySQL Server Time Zone Support</a>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UNIX_TIMESTAMP();</code></strong>
        -&gt; 882226357
mysql&gt; <strong class="userinput"><code>SELECT UNIX_TIMESTAMP('1997-10-04 22:23:00');</code></strong>
        -&gt; 875996580
</pre><p>
          When <code class="literal">UNIX_TIMESTAMP</code> is used on a
          <code class="literal">TIMESTAMP</code> column, the function returns the
          internal timestamp value directly, with no implicit
          “<span class="quote">string-to-Unix-timestamp</span>” conversion. If you
          pass an out-of-range date to
          <code class="literal">UNIX_TIMESTAMP()</code>, it returns
          <code class="literal">0</code>.
        </p><p>
          Note: If you use <code class="literal">UNIX_TIMESTAMP()</code> and
          <code class="literal">FROM_UNIXTIME()</code> to convert between
          <code class="literal">TIMESTAMP</code> values and Unix timestamp values,
          the conversion is lossy because the mapping is not one-to-one
          in both directions. For example, due to conventions for local
          time zone changes, it is possible for two
          <code class="literal">UNIX_TIMESTAMP()</code> to map two
          <code class="literal">TIMESTAMP</code> values to the same Unix timestamp
          value. <code class="literal">FROM_UNIXTIME()</code> will map that value
          back to only one of the original <code class="literal">TIMESTAMP</code>
          values. Here is an example, using <code class="literal">TIMESTAMP</code>
          values in the <code class="literal">CET</code> time zone:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UNIX_TIMESTAMP('2005-03-27 03:00:00');</code></strong>
+---------------------------------------+
| UNIX_TIMESTAMP('2005-03-27 03:00:00') |
+---------------------------------------+
|                            1111885200 |
+---------------------------------------+
mysql&gt; <strong class="userinput"><code>SELECT UNIX_TIMESTAMP('2005-03-27 02:00:00');</code></strong>
+---------------------------------------+
| UNIX_TIMESTAMP('2005-03-27 02:00:00') |
+---------------------------------------+
|                            1111885200 |
+---------------------------------------+
mysql&gt; <strong class="userinput"><code>SELECT FROM_UNIXTIME(1111885200);</code></strong>
+---------------------------+
| FROM_UNIXTIME(1111885200) |
+---------------------------+
| 2005-03-27 03:00:00       |
+---------------------------+
</pre><p>
          If you want to subtract <code class="literal">UNIX_TIMESTAMP()</code>
          columns, you might want to cast the result to signed integers.
          See <a href="mysqlqb_functions.html#cast-functions" title="Cast Functions and Operators">Cast Functions and Operators</a>.
        </p></li><li><p><a id="function_utc-date"></a>
          <a id="id2840611" class="indexterm"></a>

          <code class="literal">UTC_DATE</code>, <code class="literal">UTC_DATE()</code>
        </p><p>
          Returns the current UTC date as a value in
          <code class="literal">'YYYY-MM-DD'</code> or <code class="literal">YYYYMMDD</code>
          format, depending on whether the function is used in a string
          or numeric context.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UTC_DATE(), UTC_DATE() + 0;</code></strong>
        -&gt; '2003-08-14', 20030814
</pre></li><li><p><a id="function_utc-time"></a>
          <a id="id2840686" class="indexterm"></a>

          <code class="literal">UTC_TIME</code>, <code class="literal">UTC_TIME()</code>
        </p><p>
          Returns the current UTC time as a value in
          <code class="literal">'HH:MM:SS'</code> or <code class="literal">HHMMSS</code>
          format, depending on whether the function is used in a string
          or numeric context.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UTC_TIME(), UTC_TIME() + 0;</code></strong>
        -&gt; '18:07:53', 180753
</pre></li><li><p><a id="function_utc-timestamp"></a>
          <a id="id2840760" class="indexterm"></a>

          <code class="literal">UTC_TIMESTAMP</code>,
          <code class="literal">UTC_TIMESTAMP()</code>
        </p><p>
          Returns the current UTC date and time as a value in
          <code class="literal">'YYYY-MM-DD HH:MM:SS'</code> or
          <code class="literal">YYYYMMDDHHMMSS</code> format, depending on whether
          the function is used in a string or numeric context.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UTC_TIMESTAMP(), UTC_TIMESTAMP() + 0;</code></strong>
        -&gt; '2003-08-14 18:08:04', 20030814180804
</pre></li><li><p><a id="function_week"></a>
          <a id="id2840837" class="indexterm"></a>

          <code class="literal">WEEK(<em class="replaceable"><code>date</code></em>[,<em class="replaceable"><code>mode</code></em>])</code>
        </p><p>
          This function returns the week number for
          <em class="replaceable"><code>date</code></em>. The two-argument form of
          <code class="literal">WEEK()</code> allows you to specify whether the
          week starts on Sunday or Monday and whether the return value
          should be in the range from <code class="literal">0</code> to
          <code class="literal">53</code> or from <code class="literal">1</code> to
          <code class="literal">53</code>. If the <em class="replaceable"><code>mode</code></em>
          argument is omitted, the value of the
          <code class="literal">default_week_format</code> system variable is
          used. See <a href="http://dev.mysql.com/doc/refman/5.0/en/server-system-variables.html" target="_top">System Variables</a>.
        </p><p>
          The following table describes how the
          <em class="replaceable"><code>mode</code></em> argument works.
        </p><div class="informaltable"><table border="1"><colgroup><col /><col /><col /><col /></colgroup><tbody><tr><td> </td><td><span class="bold"><strong>First day</strong></span></td><td> </td><td> </td></tr><tr><td><span class="bold"><strong>Mode</strong></span></td><td><span class="bold"><strong>of week</strong></span></td><td><span class="bold"><strong>Range</strong></span></td><td><span class="bold"><strong>Week 1 is the first week …</strong></span></td></tr><tr><td>0</td><td>Sunday</td><td>0-53</td><td>with a Sunday in this year</td></tr><tr><td>1</td><td>Monday</td><td>0-53</td><td>with more than 3 days this year</td></tr><tr><td>2</td><td>Sunday</td><td>1-53</td><td>with a Sunday in this year</td></tr><tr><td>3</td><td>Monday</td><td>1-53</td><td>with more than 3 days this year</td></tr><tr><td>4</td><td>Sunday</td><td>0-53</td><td>with more than 3 days this year</td></tr><tr><td>5</td><td>Monday</td><td>0-53</td><td>with a Monday in this year</td></tr><tr><td>6</td><td>Sunday</td><td>1-53</td><td>with more than 3 days this year</td></tr><tr><td>7</td><td>Monday</td><td>1-53</td><td>with a Monday in this year</td></tr></tbody></table></div><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT WEEK('1998-02-20');</code></strong>
        -&gt; 7
mysql&gt; <strong class="userinput"><code>SELECT WEEK('1998-02-20',0);</code></strong>
        -&gt; 7
mysql&gt; <strong class="userinput"><code>SELECT WEEK('1998-02-20',1);</code></strong>
        -&gt; 8
mysql&gt; <strong class="userinput"><code>SELECT WEEK('1998-12-31',1);</code></strong>
        -&gt; 53
</pre><p>
          Note that if a date falls in the last week of the previous
          year, MySQL returns <code class="literal">0</code> if you do not use
          <code class="literal">2</code>, <code class="literal">3</code>,
          <code class="literal">6</code>, or <code class="literal">7</code> as the optional
          <em class="replaceable"><code>mode</code></em> argument:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT YEAR('2000-01-01'), WEEK('2000-01-01',0);</code></strong>
        -&gt; 2000, 0
</pre><p>
          One might argue that MySQL should return <code class="literal">52</code>
          for the <code class="literal">WEEK()</code> function, because the given
          date actually occurs in the 52nd week of 1999. We decided to
          return <code class="literal">0</code> instead because we want the
          function to return “<span class="quote">the week number in the given
          year.</span>” This makes use of the <code class="literal">WEEK()</code>
          function reliable when combined with other functions that
          extract a date part from a date.
        </p><p>
          If you would prefer the result to be evaluated with respect to
          the year that contains the first day of the week for the given
          date, use <code class="literal">0</code>, <code class="literal">2</code>,
          <code class="literal">5</code>, or <code class="literal">7</code> as the optional
          <em class="replaceable"><code>mode</code></em> argument.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT WEEK('2000-01-01',2);</code></strong>
        -&gt; 52
</pre><p>
          Alternatively, use the <code class="literal">YEARWEEK()</code> function:
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT YEARWEEK('2000-01-01');</code></strong>
        -&gt; 199952
mysql&gt; <strong class="userinput"><code>SELECT MID(YEARWEEK('2000-01-01'),5,2);</code></strong>
        -&gt; '52'
</pre></li><li><p><a id="function_weekday"></a>
          <a id="id2841296" class="indexterm"></a>

          <code class="literal">WEEKDAY(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the weekday index for <em class="replaceable"><code>date</code></em>
          (<code class="literal">0</code> = Monday, <code class="literal">1</code> =
          Tuesday, … <code class="literal">6</code> = Sunday).
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT WEEKDAY('1998-02-03 22:23:00');</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT WEEKDAY('1997-11-05');</code></strong>
        -&gt; 2
</pre></li><li><p><a id="function_weekofyear"></a>
          <a id="id2841380" class="indexterm"></a>

          <code class="literal">WEEKOFYEAR(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the calendar week of the date as a number in the range
          from <code class="literal">1</code> to <code class="literal">53</code>.
          <code class="literal">WEEKOFYEAR()</code> is a compatibility function
          that is equivalent to
          <code class="literal">WEEK(<em class="replaceable"><code>date</code></em>,3)</code>.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT WEEKOFYEAR('1998-02-20');</code></strong>
        -&gt; 8
</pre></li><li><p><a id="function_year"></a>
          <a id="id2841464" class="indexterm"></a>

          <code class="literal">YEAR(<em class="replaceable"><code>date</code></em>)</code>
        </p><p>
          Returns the year for <em class="replaceable"><code>date</code></em>, in the
          range <code class="literal">1000</code> to <code class="literal">9999</code>, or
          <code class="literal">0</code> for the “<span class="quote">zero</span>” date.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT YEAR('98-02-03');</code></strong>
        -&gt; 1998
</pre></li><li><p><a id="function_yearweek"></a>
          <a id="id2841547" class="indexterm"></a>

          <code class="literal">YEARWEEK(<em class="replaceable"><code>date</code></em>)</code>,
          <code class="literal">YEARWEEK(<em class="replaceable"><code>date</code></em>,<em class="replaceable"><code>mode</code></em>)</code>
        </p><p>
          Returns year and week for a date. The
          <em class="replaceable"><code>mode</code></em> argument works exactly like
          the <em class="replaceable"><code>mode</code></em> argument to
          <code class="literal">WEEK()</code>. The year in the result may be
          different from the year in the date argument for the first and
          the last week of the year.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT YEARWEEK('1987-01-01');</code></strong>
        -&gt; 198653
</pre><p>
          Note that the week number is different from what the
          <code class="literal">WEEK()</code> function would return
          (<code class="literal">0</code>) for optional arguments
          <code class="literal">0</code> or <code class="literal">1</code>, as
          <code class="literal">WEEK()</code> then returns the week in the context
          of the given year.
        </p></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="mysql-calendar"></a>What Calendar Is Used By MySQL?</h2></div></div></div><a id="id2841654" class="indexterm"></a><p>
      MySQL uses what is known as a <em class="firstterm">proleptic Gregorian
      calendar</em>.
    </p><p>
      Every country that has switched from the Julian to the Gregorian
      calendar has had to discard at least ten days during the switch.
      To see how this works, consider the month of October 1582, when
      the first Julian-to-Gregorian switch occurred:
    </p><div class="informaltable"><table border="1"><colgroup><col /><col /><col /><col /><col /><col /><col /></colgroup><tbody><tr><td>Monday</td><td>Tuesday</td><td>Wednesday</td><td>Thursday</td><td>Friday</td><td>Saturday</td><td>Sunday</td></tr><tr><td>1</td><td>2</td><td>3</td><td>4</td><td>15</td><td>16</td><td>17</td></tr><tr><td>18</td><td>19</td><td>20</td><td>21</td><td>22</td><td>23</td><td>24</td></tr><tr><td>25</td><td>26</td><td>27</td><td>28</td><td>29</td><td>30</td><td>31</td></tr></tbody></table></div><p>
      There are no dates between October 4 and October 15. This
      discontinuity is called the <em class="firstterm">cutover</em>. Any
      dates before the cutover are Julian, and any dates following the
      cutover are Gregorian. Dates during a cutover are non-existent.
    </p><p>
      A calendar applied to dates when it wasn't actually in use is
      called <em class="firstterm">proleptic</em>. Thus, if we assume there
      was never a cutover and Gregorian rules always rule, we have a
      proleptic Gregorian calendar. This is what is used by MySQL, as is
      required by standard SQL. For this reason, dates prior to the
      cutover stored as MySQL <code class="literal">DATE</code> or
      <code class="literal">DATETIME</code> values must be adjusted to compensate
      for the difference. It is important to realize that the cutover
      did not occur at the same time in all countries, and that the
      later it happened, the more days were lost. For example, in Great
      Britain, it took place in 1752, when Wednesday September 2 was
      followed by Thursday September 14. Russia remained on the Julian
      calendar until 1918, losing 13 days in the process, and what is
      popularly referred to as its “<span class="quote">October Revolution</span>”
      occurred in November according to the Gregorian calendar.
    </p></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="fulltext-search"></a>Full-Text Search Functions</h2></div></div></div><div class="toc"><dl><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-boolean">Boolean Full-Text Searches</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-query-expansion">Full-Text Searches with Query Expansion</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-stopwords">Full-Text Stopwords</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-restrictions">Full-Text Restrictions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#fulltext-fine-tuning">Fine-Tuning MySQL Full-Text Search</a></span></dt></dl></div><a id="id2841890" class="indexterm"></a><a id="id2841900" class="indexterm"></a><a id="id2841907" class="indexterm"></a><a id="id2841914" class="indexterm"></a><pre class="programlisting">MATCH (<em class="replaceable"><code>col1</code></em>,<em class="replaceable"><code>col2</code></em>,...) AGAINST (<em class="replaceable"><code>expr</code></em> [<em class="replaceable"><code>search_modifier</code></em>])

<em class="replaceable"><code>search_modifier:</code></em> { IN BOOLEAN MODE | WITH QUERY EXPANSION }
</pre><p>
      MySQL has support for full-text indexing and searching:
    </p><div class="itemizedlist"><ul type="disc"><li><p>
          A full-text index in MySQL is an index of type
          <code class="literal">FULLTEXT</code>.
        </p></li><li><p>
          Full-text indexes can be used only with
          <code class="literal">MyISAM</code> tables, and can be created only for
          <code class="literal">CHAR</code>, <code class="literal">VARCHAR</code>, or
          <code class="literal">TEXT</code> columns.
        </p></li><li><p>
          A <code class="literal">FULLTEXT</code> index definition can be given in
          the <code class="literal">CREATE TABLE</code> statement when a table is
          created, or added later using <code class="literal">ALTER TABLE</code>
          or <code class="literal">CREATE INDEX</code>.
        </p></li><li><p>
          For large datasets, it is much faster to load your data into a
          table that has no <code class="literal">FULLTEXT</code> index and then
          create the index after that, than to load data into a table
          that has an existing <code class="literal">FULLTEXT</code> index.
        </p></li></ul></div><p>
      Full-text searching is performed using <code class="literal">MATCH() ...
      AGAINST</code> syntax. <code class="literal">MATCH()</code> takes a
      comma-separated list that names the columns to be searched.
      <code class="literal">AGAINST</code> takes a string to search for, and an
      optional modifier that indicates what type of search to perform.
      The search string must be a literal string, not a variable or a
      column name. There are three types of full-text searches:
    </p><div class="itemizedlist"><ul type="disc"><li><p>
          A boolean search interprets the search string using the rules
          of a special query language. The string contains the words to
          search for. It can also contain operators that specify
          requirements such that a word must be present or absent in
          matching rows, or that it should be weighted higher or lower
          than usual. Common words such as “<span class="quote">some</span>” or
          “<span class="quote">then</span>” are stopwords and do not match if present
          in the search string. The <code class="literal">IN BOOLEAN MODE</code>
          modifier specifies a boolean search. For more information, see
          <a href="mysqlqb_functions.html#fulltext-boolean" title="Boolean Full-Text Searches">Boolean Full-Text Searches</a>.
        </p></li><li><p>
          A natural language search interprets the search string as a
          phrase in natural human language (a phrase in free text).
          There are no special operators. The stopword list applies. In
          addition, words that are present in more than 50% of the rows
          are considered common and do not match. Full-text searches are
          natural language searches if no modifier is given.
        </p></li><li><p>
          A query expansion search is a modification of a natural
          language search. The search string is used to perform a
          natural language search. Then words from the most relevant
          rows returned by the search are added to the search string and
          the search is done again. The query returns the rows from the
          second search. The <code class="literal">WITH QUERY EXPANSION</code>
          modifier specifies a query expansion search. For more
          information, see <a href="mysqlqb_functions.html#fulltext-query-expansion" title="Full-Text Searches with Query Expansion">Full-Text Searches with Query Expansion</a>.
        </p></li></ul></div><p>
      Constraints on full-text searching are listed in
      <a href="mysqlqb_functions.html#fulltext-restrictions" title="Full-Text Restrictions">Full-Text Restrictions</a>.
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>CREATE TABLE articles (</code></strong>
    -&gt;   <strong class="userinput"><code>id INT UNSIGNED AUTO_INCREMENT NOT NULL PRIMARY KEY,</code></strong>
    -&gt;   <strong class="userinput"><code>title VARCHAR(200),</code></strong>
    -&gt;   <strong class="userinput"><code>body TEXT,</code></strong>
    -&gt;   <strong class="userinput"><code>FULLTEXT (title,body)</code></strong>
    -&gt; <strong class="userinput"><code>);</code></strong>
Query OK, 0 rows affected (0.00 sec)

mysql&gt; <strong class="userinput"><code>INSERT INTO articles (title,body) VALUES</code></strong>
    -&gt; <strong class="userinput"><code>('MySQL Tutorial','DBMS stands for DataBase ...'),</code></strong>
    -&gt; <strong class="userinput"><code>('How To Use MySQL Well','After you went through a ...'),</code></strong>
    -&gt; <strong class="userinput"><code>('Optimizing MySQL','In this tutorial we will show ...'),</code></strong>
    -&gt; <strong class="userinput"><code>('1001 MySQL Tricks','1. Never run mysqld as root. 2. ...'),</code></strong>
    -&gt; <strong class="userinput"><code>('MySQL vs. YourSQL','In the following database comparison ...'),</code></strong>
    -&gt; <strong class="userinput"><code>('MySQL Security','When configured properly, MySQL ...');</code></strong>
Query OK, 6 rows affected (0.00 sec)
Records: 6  Duplicates: 0  Warnings: 0

mysql&gt; <strong class="userinput"><code>SELECT * FROM articles</code></strong>
    -&gt; <strong class="userinput"><code>WHERE MATCH (title,body) AGAINST ('database');</code></strong>
+----+-------------------+------------------------------------------+
| id | title             | body                                     |
+----+-------------------+------------------------------------------+
|  5 | MySQL vs. YourSQL | In the following database comparison ... |
|  1 | MySQL Tutorial    | DBMS stands for DataBase ...             |
+----+-------------------+------------------------------------------+
2 rows in set (0.00 sec)
</pre><p>
      The <code class="literal">MATCH()</code> function performs a natural
      language search for a string against a <em class="firstterm">text
      collection</em>. A collection is a set of one or more
      columns included in a <code class="literal">FULLTEXT</code> index. The
      search string is given as the argument to
      <code class="literal">AGAINST()</code>. For each row in the table,
      <code class="literal">MATCH()</code> returns a relevance value; that is, a
      similarity measure between the search string and the text in that
      row in the columns named in the <code class="literal">MATCH()</code> list.
    </p><p>
      By default, the search is performed in case-insensitive fashion.
      However, you can perform a case-sensitive full-text search by
      using a binary collation for the indexed columns. For example, a
      column that uses the <code class="literal">latin1</code> character set of
      can be assigned a collation of <code class="literal">latin1_bin</code> to
      make it case sensitive for full-text searches.
    </p><p>
      When <code class="literal">MATCH()</code> is used in a
      <code class="literal">WHERE</code> clause, as in the example shown earlier,
      the rows returned are automatically sorted with the highest
      relevance first. Relevance values are non-negative floating-point
      numbers. Zero relevance means no similarity. Relevance is computed
      based on the number of words in the row, the number of unique
      words in that row, the total number of words in the collection,
      and the number of documents (rows) that contain a particular word.
    </p><p>
      To simply count matches, you could use a query like this:
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT COUNT(*) FROM articles</code></strong>
    -&gt; <strong class="userinput"><code>WHERE MATCH (title,body)</code></strong>
    -&gt; <strong class="userinput"><code>AGAINST ('database');</code></strong>
+----------+
| COUNT(*) |
+----------+
|        2 | 
+----------+
1 row in set (0.00 sec)
</pre><p>
      However, you might find it quicker to rewrite the query as
      follows:
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT</code></strong>
    -&gt; <strong class="userinput"><code>COUNT(IF(MATCH (title,body) AGAINST ('database'), 1, NULL))</code></strong>
    -&gt; <strong class="userinput"><code>AS count</code></strong>
    -&gt; <strong class="userinput"><code>FROM articles;</code></strong>
+-------+
| count |
+-------+
|     2 | 
+-------+
1 row in set (0.00 sec)
</pre><p>
      The first query sorts the results by relevance whereas the second
      does not. However, the second query performs a full table scan and
      the first does not. The first may be faster if the search matches
      few rows; otherwise, the second may be faster because it would
      read many rows anyway.
    </p><p>
      For natural-language full-text searches, it is a requirement that
      the columns named in the <code class="literal">MATCH()</code> function be
      the same columns included in some <code class="literal">FULLTEXT</code>
      index in your table. For the preceding query, note that the
      columns named in the <code class="literal">MATCH()</code> function
      (<code class="literal">title</code> and <code class="literal">body</code>) are the
      same as those named in the definition of the
      <code class="literal">article</code> table's <code class="literal">FULLTEXT</code>
      index. If you wanted to search the <code class="literal">title</code> or
      <code class="literal">body</code> separately, you would need to create
      separate <code class="literal">FULLTEXT</code> indexes for each column.
    </p><p>
      It is also possible to perform a boolean search or a search with
      query expansion. These search types are described in
      <a href="mysqlqb_functions.html#fulltext-boolean" title="Boolean Full-Text Searches">Boolean Full-Text Searches</a>, and
      <a href="mysqlqb_functions.html#fulltext-query-expansion" title="Full-Text Searches with Query Expansion">Full-Text Searches with Query Expansion</a>.
    </p><p>
      A full-text search that uses an index can name columns only from a
      single table in the <code class="literal">MATCH()</code> clause because an
      index cannot span multiple tables. A boolean search can be done in
      the absence of an index (albeit more slowly), in which case it is
      possible to name columns from multiple tables.
    </p><p>
      The preceding example is a basic illustration that shows how to
      use the <code class="literal">MATCH()</code> function where rows are
      returned in order of decreasing relevance. The next example shows
      how to retrieve the relevance values explicitly. Returned rows are
      not ordered because the <code class="literal">SELECT</code> statement
      includes neither <code class="literal">WHERE</code> nor <code class="literal">ORDER
      BY</code> clauses:
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT id, MATCH (title,body) AGAINST ('Tutorial')</code></strong>
    -&gt; <strong class="userinput"><code>FROM articles;</code></strong>
+----+-----------------------------------------+
| id | MATCH (title,body) AGAINST ('Tutorial') |
+----+-----------------------------------------+
|  1 |                        0.65545833110809 |
|  2 |                                       0 |
|  3 |                        0.66266459226608 |
|  4 |                                       0 |
|  5 |                                       0 |
|  6 |                                       0 |
+----+-----------------------------------------+
6 rows in set (0.00 sec)
</pre><p>
      The following example is more complex. The query returns the
      relevance values and it also sorts the rows in order of decreasing
      relevance. To achieve this result, you should specify
      <code class="literal">MATCH()</code> twice: once in the
      <code class="literal">SELECT</code> list and once in the
      <code class="literal">WHERE</code> clause. This causes no additional
      overhead, because the MySQL optimizer notices that the two
      <code class="literal">MATCH()</code> calls are identical and invokes the
      full-text search code only once.
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT id, body, MATCH (title,body) AGAINST</code></strong>
    -&gt; <strong class="userinput"><code>('Security implications of running MySQL as root') AS score</code></strong>
    -&gt; <strong class="userinput"><code>FROM articles WHERE MATCH (title,body) AGAINST</code></strong>
    -&gt; <strong class="userinput"><code>('Security implications of running MySQL as root');</code></strong>
+----+-------------------------------------+-----------------+
| id | body                                | score           |
+----+-------------------------------------+-----------------+
|  4 | 1. Never run mysqld as root. 2. ... | 1.5219271183014 |
|  6 | When configured properly, MySQL ... | 1.3114095926285 |
+----+-------------------------------------+-----------------+
2 rows in set (0.00 sec)
</pre><p>
      The MySQL <code class="literal">FULLTEXT</code> implementation regards any
      sequence of true word characters (letters, digits, and
      underscores) as a word. That sequence may also contain apostrophes
      (‘<code class="literal">'</code>’), but not more than one in a
      row. This means that <code class="literal">aaa'bbb</code> is regarded as one
      word, but <code class="literal">aaa''bbb</code> is regarded as two words.
      Apostrophes at the beginning or the end of a word are stripped by
      the <code class="literal">FULLTEXT</code> parser;
      <code class="literal">'aaa'bbb'</code> would be parsed as
      <code class="literal">aaa'bbb</code>.
    </p><p>
      The <code class="literal">FULLTEXT</code> parser determines where words
      start and end by looking for certain delimiter characters; for
      example, ‘<code class="literal"> </code>’ (space),
      ‘<code class="literal">,</code>’ (comma), and
      ‘<code class="literal">.</code>’ (period). If words are not
      separated by delimiters (as in, for example, Chinese), the
      <code class="literal">FULLTEXT</code> parser cannot determine where a word
      begins or ends. To be able to add words or other indexed terms in
      such languages to a <code class="literal">FULLTEXT</code> index, you must
      preprocess them so that they are separated by some arbitrary
      delimiter such as ‘<code class="literal">"</code>’.
    </p><p>
      Some words are ignored in full-text searches:
    </p><div class="itemizedlist"><ul type="disc"><li><p>
          Any word that is too short is ignored. The default minimum
          length of words that are found by full-text searches is four
          characters.
        </p></li><li><p>
          Words in the stopword list are ignored. A stopword is a word
          such as “<span class="quote">the</span>” or “<span class="quote">some</span>” that is so
          common that it is considered to have zero semantic value.
          There is a built-in stopword list, but it can be overwritten
          by a user-defined list.
        </p></li></ul></div><p>
      The default stopword list is given in
      <a href="mysqlqb_functions.html#fulltext-stopwords" title="Full-Text Stopwords">Full-Text Stopwords</a>. The default minimum word
      length and stopword list can be changed as described in
      <a href="mysqlqb_functions.html#fulltext-fine-tuning" title="Fine-Tuning MySQL Full-Text Search">Fine-Tuning MySQL Full-Text Search</a>.
    </p><p>
      Every correct word in the collection and in the query is weighted
      according to its significance in the collection or query.
      Consequently, a word that is present in many documents has a lower
      weight (and may even have a zero weight), because it has lower
      semantic value in this particular collection. Conversely, if the
      word is rare, it receives a higher weight. The weights of the
      words are combined to compute the relevance of the row.
    </p><p>
      Such a technique works best with large collections (in fact, it
      was carefully tuned this way). For very small tables, word
      distribution does not adequately reflect their semantic value, and
      this model may sometimes produce bizarre results. For example,
      although the word “<span class="quote">MySQL</span>” is present in every row of
      the <code class="literal">articles</code> table shown earlier, a search for
      the word produces no results:
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT * FROM articles</code></strong>
    -&gt; <strong class="userinput"><code>WHERE MATCH (title,body) AGAINST ('MySQL');</code></strong>
Empty set (0.00 sec)
</pre><p>
      The search result is empty because the word “<span class="quote">MySQL</span>”
      is present in at least 50% of the rows. As such, it is effectively
      treated as a stopword. For large datasets, this is the most
      desirable behavior: A natural language query should not return
      every second row from a 1GB table. For small datasets, it may be
      less desirable.
    </p><p>
      A word that matches half of the rows in a table is less likely to
      locate relevant documents. In fact, it most likely finds plenty of
      irrelevant documents. We all know this happens far too often when
      we are trying to find something on the Internet with a search
      engine. It is with this reasoning that rows containing the word
      are assigned a low semantic value for <span class="emphasis"><em>the particular
      dataset in which they occur</em></span>. A given word may exceed
      the 50% threshold in one dataset but not another.
    </p><p>
      The 50% threshold has a significant implication when you first try
      full-text searching to see how it works: If you create a table and
      insert only one or two rows of text into it, every word in the
      text occurs in at least 50% of the rows. As a result, no search
      returns any results. Be sure to insert at least three rows, and
      preferably many more. Users who need to bypass the 50% limitation
      can use the boolean search mode; see
      <a href="mysqlqb_functions.html#fulltext-boolean" title="Boolean Full-Text Searches">Boolean Full-Text Searches</a>.
    </p><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="fulltext-boolean"></a>Boolean Full-Text Searches</h3></div></div></div><p>
        MySQL can perform boolean full-text searches using the
        <code class="literal">IN BOOLEAN MODE</code> modifier:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT * FROM articles WHERE MATCH (title,body)</code></strong>
    -&gt; <strong class="userinput"><code>AGAINST ('+MySQL -YourSQL' IN BOOLEAN MODE);</code></strong>
+----+-----------------------+-------------------------------------+
| id | title                 | body                                |
+----+-----------------------+-------------------------------------+
|  1 | MySQL Tutorial        | DBMS stands for DataBase ...        |
|  2 | How To Use MySQL Well | After you went through a ...        |
|  3 | Optimizing MySQL      | In this tutorial we will show ...   |
|  4 | 1001 MySQL Tricks     | 1. Never run mysqld as root. 2. ... |
|  6 | MySQL Security        | When configured properly, MySQL ... |
+----+-----------------------+-------------------------------------+
</pre><p>
        The <code class="literal">+</code> and <code class="literal">-</code> operators
        indicate that a word is required to be present or absent,
        respectively, for a match to occur. Thus, this query retrieves
        all the rows that contain the word “<span class="quote">MySQL</span>” but that
        do <span class="emphasis"><em>not</em></span> contain the word
        “<span class="quote">YourSQL</span>”.
      </p><p>
        Boolean full-text searches have these characteristics:
      </p><div class="itemizedlist"><ul type="disc"><li><p>
            They do not use the 50% threshold.
          </p></li><li><p>
            They do not automatically sort rows in order of decreasing
            relevance. You can see this from the preceding query result:
            The row with the highest relevance is the one that contains
            “<span class="quote">MySQL</span>” twice, but it is listed last, not
            first.
          </p></li><li><p>
            They can work even without a <code class="literal">FULLTEXT</code>
            index, although a search executed in this fashion would be
            quite slow.
          </p></li><li><p>
            The minimum and maximum word length full-text parameters
            apply.
          </p></li><li><p>
            The stopword list applies.
          </p></li></ul></div><p>
        The boolean full-text search capability supports the following
        operators:
      </p><div class="itemizedlist"><ul type="disc"><li><p>
            <code class="literal">+</code>
          </p><p>
            A leading plus sign indicates that this word
            <span class="emphasis"><em>must</em></span> be present in each row that is
            returned.
          </p></li><li><p>
            <code class="literal">-</code>
          </p><p>
            A leading minus sign indicates that this word must
            <span class="emphasis"><em>not</em></span> be present in any of the rows that
            are returned.
          </p><p>
            Note: The <code class="literal">-</code> operator acts only to exclude
            rows that are otherwise matched by other search terms. Thus,
            a boolean-mode search that contains only terms preceded by
            <code class="literal">-</code> returns an empty result. It does not
            return “<span class="quote">all rows except those containing any of the
            excluded terms.</span>”
          </p></li><li><p>
            (no operator)
          </p><p>
            By default (when neither <code class="literal">+</code> nor
            <code class="literal">-</code> is specified) the word is optional, but
            the rows that contain it are rated higher. This mimics the
            behavior of <code class="literal">MATCH() ... AGAINST()</code> without
            the <code class="literal">IN BOOLEAN MODE</code> modifier.
          </p></li><li><p>
            <code class="literal">&gt; &lt;</code>
          </p><p>
            These two operators are used to change a word's contribution
            to the relevance value that is assigned to a row. The
            <code class="literal">&gt;</code> operator increases the contribution
            and the <code class="literal">&lt;</code> operator decreases it. See
            the example following this list.
          </p></li><li><p>
            <code class="literal">( )</code>
          </p><p>
            Parentheses group words into subexpressions. Parenthesized
            groups can be nested.
          </p></li><li><p>
            <code class="literal">~</code>
          </p><p>
            A leading tilde acts as a negation operator, causing the
            word's contribution to the row's relevance to be negative.
            This is useful for marking “<span class="quote">noise</span>” words. A row
            containing such a word is rated lower than others, but is
            not excluded altogether, as it would be with the
            <code class="literal">-</code> operator.
          </p></li><li><p>
            <code class="literal">*</code>
          </p><p>
            The asterisk serves as the truncation (or wildcard)
            operator. Unlike the other operators, it should be
            <span class="emphasis"><em>appended</em></span> to the word to be affected.
            Words match if they begin with the word preceding the
            <code class="literal">*</code> operator.
          </p><p>
            If a stopword or too-short word is specified with the
            truncation operator, it will not be stripped from a boolean
            query. For example, a search for <code class="literal">'+word
            +stopword*'</code> will likely return fewer rows than a
            search for <code class="literal">'+word +stopword'</code> because the
            former query remains as is and requires
            <code class="literal">stopword*</code> to be present in a document.
            The latter query is transformed to <code class="literal">+word</code>.
          </p></li><li><p>
            <code class="literal">"</code>
          </p><p>
            A phrase that is enclosed within double quote
            (‘<code class="literal">"</code>’) characters matches only
            rows that contain the phrase <span class="emphasis"><em>literally, as it was
            typed</em></span>. The full-text engine splits the phrase
            into words, performs a search in the
            <code class="literal">FULLTEXT</code> index for the words. Prior to
            MySQL 5.0.3, the engine then performed a substring search
            for the phrase in the records that were found, so the match
            must include non-word characters in the phrase. As of MySQL
            5.0.3, non-word characters need not be matched exactly:
            Phrase searching requires only that matches contain exactly
            the same words as the phrase and in the same order. For
            example, <code class="literal">"test phrase"</code> matches
            <code class="literal">"test, phrase"</code> in MySQL 5.0.3, but not
            before.
          </p><p>
            If the phrase contains no words that are in the index, the
            result is empty. For example, if all words are either
            stopwords or shorter than the minimum length of indexed
            words, the result is empty.
          </p></li></ul></div><p>
        The following examples demonstrate some search strings that use
        boolean full-text operators:
      </p><div class="itemizedlist"><ul type="disc"><li><p>
            <code class="literal">'apple banana'</code>
          </p><p>
            Find rows that contain at least one of the two words.
          </p></li><li><p>
            <code class="literal">'+apple +juice'</code>
          </p><p>
            Find rows that contain both words.
          </p></li><li><p>
            <code class="literal">'+apple macintosh'</code>
          </p><p>
            Find rows that contain the word “<span class="quote">apple</span>”, but
            rank rows higher if they also contain
            “<span class="quote">macintosh</span>”.
          </p></li><li><p>
            <code class="literal">'+apple -macintosh'</code>
          </p><p>
            Find rows that contain the word “<span class="quote">apple</span>” but not
            “<span class="quote">macintosh</span>”.
          </p></li><li><p>
            <code class="literal">'+apple ~macintosh'</code>
          </p><p>
            Find rows that contain the word “<span class="quote">apple</span>”, but if
            the row also contains the word “<span class="quote">macintosh</span>”,
            rate it lower than if row does not. This is
            “<span class="quote">softer</span>” than a search for <code class="literal">'+apple
            -macintosh'</code>, for which the presence of
            “<span class="quote">macintosh</span>” causes the row not to be returned
            at all.
          </p></li><li><p>
            <code class="literal">'+apple +(&gt;turnover &lt;strudel)'</code>
          </p><p>
            Find rows that contain the words “<span class="quote">apple</span>” and
            “<span class="quote">turnover</span>”, or “<span class="quote">apple</span>” and
            “<span class="quote">strudel</span>” (in any order), but rank “<span class="quote">apple
            turnover</span>” higher than “<span class="quote">apple strudel</span>”.
          </p></li><li><p>
            <code class="literal">'apple*'</code>
          </p><p>
            Find rows that contain words such as “<span class="quote">apple</span>”,
            “<span class="quote">apples</span>”, “<span class="quote">applesauce</span>”, or
            “<span class="quote">applet</span>”.
          </p></li><li><p>
            <code class="literal">'"some words"'</code>
          </p><p>
            Find rows that contain the exact phrase “<span class="quote">some
            words</span>” (for example, rows that contain “<span class="quote">some
            words of wisdom</span>” but not “<span class="quote">some noise
            words</span>”). Note that the
            ‘<code class="literal">"</code>’ characters that enclose
            the phrase are operator characters that delimit the phrase.
            They are not the quotes that enclose the search string
            itself.
          </p></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="fulltext-query-expansion"></a>Full-Text Searches with Query Expansion</h3></div></div></div><p>
        Full-text search supports query expansion (and in particular,
        its variant “<span class="quote">blind query expansion</span>”). This is
        generally useful when a search phrase is too short, which often
        means that the user is relying on implied knowledge that the
        full-text search engine lacks. For example, a user searching for
        “<span class="quote">database</span>” may really mean that
        “<span class="quote">MySQL</span>”, “<span class="quote">Oracle</span>”, “<span class="quote">DB2</span>”,
        and “<span class="quote">RDBMS</span>” all are phrases that should match
        “<span class="quote">databases</span>” and should be returned, too. This is
        implied knowledge.
      </p><p>
        Blind query expansion (also known as automatic relevance
        feedback) is enabled by adding <code class="literal">WITH QUERY
        EXPANSION</code> following the search phrase. It works by
        performing the search twice, where the search phrase for the
        second search is the original search phrase concatenated with
        the few most highly relevant documents from the first search.
        Thus, if one of these documents contains the word
        “<span class="quote">databases</span>” and the word “<span class="quote">MySQL</span>”, the
        second search finds the documents that contain the word
        “<span class="quote">MySQL</span>” even if they do not contain the word
        “<span class="quote">database</span>”. The following example shows this
        difference:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT * FROM articles</code></strong>
    -&gt; <strong class="userinput"><code>WHERE MATCH (title,body) AGAINST ('database');</code></strong>
+----+-------------------+------------------------------------------+
| id | title             | body                                     |
+----+-------------------+------------------------------------------+
|  5 | MySQL vs. YourSQL | In the following database comparison ... |
|  1 | MySQL Tutorial    | DBMS stands for DataBase ...             |
+----+-------------------+------------------------------------------+
2 rows in set (0.00 sec)

mysql&gt; <strong class="userinput"><code>SELECT * FROM articles</code></strong>
    -&gt; <strong class="userinput"><code>WHERE MATCH (title,body)</code></strong>
    -&gt; <strong class="userinput"><code>AGAINST ('database' WITH QUERY EXPANSION);</code></strong>
+----+-------------------+------------------------------------------+
| id | title             | body                                     |
+----+-------------------+------------------------------------------+
|  1 | MySQL Tutorial    | DBMS stands for DataBase ...             |
|  5 | MySQL vs. YourSQL | In the following database comparison ... |
|  3 | Optimizing MySQL  | In this tutorial we will show ...        |
+----+-------------------+------------------------------------------+
3 rows in set (0.00 sec)
</pre><p>
        Another example could be searching for books by Georges Simenon
        about Maigret, when a user is not sure how to spell
        “<span class="quote">Maigret</span>”. A search for “<span class="quote">Megre and the
        reluctant witnesses</span>” finds only “<span class="quote">Maigret and the
        Reluctant Witnesses</span>” without query expansion. A search
        with query expansion finds all books with the word
        “<span class="quote">Maigret</span>” on the second pass.
      </p><p>
        <span class="bold"><strong>Note</strong></span>: Because blind query
        expansion tends to increase noise significantly by returning
        non-relevant documents, it is meaningful to use only when a
        search phrase is rather short.
      </p></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="fulltext-stopwords"></a>Full-Text Stopwords</h3></div></div></div><p>
        The following table shows the default list of full-text
        stopwords.
      </p><div class="informaltable"><table border="1"><colgroup><col /><col /><col /><col /><col /></colgroup><tbody><tr><td>a's</td><td>able</td><td>about</td><td>above</td><td>according</td></tr><tr><td>accordingly</td><td>across</td><td>actually</td><td>after</td><td>afterwards</td></tr><tr><td>again</td><td>against</td><td>ain't</td><td>all</td><td>allow</td></tr><tr><td>allows</td><td>almost</td><td>alone</td><td>along</td><td>already</td></tr><tr><td>also</td><td>although</td><td>always</td><td>am</td><td>among</td></tr><tr><td>amongst</td><td>an</td><td>and</td><td>another</td><td>any</td></tr><tr><td>anybody</td><td>anyhow</td><td>anyone</td><td>anything</td><td>anyway</td></tr><tr><td>anyways</td><td>anywhere</td><td>apart</td><td>appear</td><td>appreciate</td></tr><tr><td>appropriate</td><td>are</td><td>aren't</td><td>around</td><td>as</td></tr><tr><td>aside</td><td>ask</td><td>asking</td><td>associated</td><td>at</td></tr><tr><td>available</td><td>away</td><td>awfully</td><td>be</td><td>became</td></tr><tr><td>because</td><td>become</td><td>becomes</td><td>becoming</td><td>been</td></tr><tr><td>before</td><td>beforehand</td><td>behind</td><td>being</td><td>believe</td></tr><tr><td>below</td><td>beside</td><td>besides</td><td>best</td><td>better</td></tr><tr><td>between</td><td>beyond</td><td>both</td><td>brief</td><td>but</td></tr><tr><td>by</td><td>c'mon</td><td>c's</td><td>came</td><td>can</td></tr><tr><td>can't</td><td>cannot</td><td>cant</td><td>cause</td><td>causes</td></tr><tr><td>certain</td><td>certainly</td><td>changes</td><td>clearly</td><td>co</td></tr><tr><td>com</td><td>come</td><td>comes</td><td>concerning</td><td>consequently</td></tr><tr><td>consider</td><td>considering</td><td>contain</td><td>containing</td><td>contains</td></tr><tr><td>corresponding</td><td>could</td><td>couldn't</td><td>course</td><td>currently</td></tr><tr><td>definitely</td><td>described</td><td>despite</td><td>did</td><td>didn't</td></tr><tr><td>different</td><td>do</td><td>does</td><td>doesn't</td><td>doing</td></tr><tr><td>don't</td><td>done</td><td>down</td><td>downwards</td><td>during</td></tr><tr><td>each</td><td>edu</td><td>eg</td><td>eight</td><td>either</td></tr><tr><td>else</td><td>elsewhere</td><td>enough</td><td>entirely</td><td>especially</td></tr><tr><td>et</td><td>etc</td><td>even</td><td>ever</td><td>every</td></tr><tr><td>everybody</td><td>everyone</td><td>everything</td><td>everywhere</td><td>ex</td></tr><tr><td>exactly</td><td>example</td><td>except</td><td>far</td><td>few</td></tr><tr><td>fifth</td><td>first</td><td>five</td><td>followed</td><td>following</td></tr><tr><td>follows</td><td>for</td><td>former</td><td>formerly</td><td>forth</td></tr><tr><td>four</td><td>from</td><td>further</td><td>furthermore</td><td>get</td></tr><tr><td>gets</td><td>getting</td><td>given</td><td>gives</td><td>go</td></tr><tr><td>goes</td><td>going</td><td>gone</td><td>got</td><td>gotten</td></tr><tr><td>greetings</td><td>had</td><td>hadn't</td><td>happens</td><td>hardly</td></tr><tr><td>has</td><td>hasn't</td><td>have</td><td>haven't</td><td>having</td></tr><tr><td>he</td><td>he's</td><td>hello</td><td>help</td><td>hence</td></tr><tr><td>her</td><td>here</td><td>here's</td><td>hereafter</td><td>hereby</td></tr><tr><td>herein</td><td>hereupon</td><td>hers</td><td>herself</td><td>hi</td></tr><tr><td>him</td><td>himself</td><td>his</td><td>hither</td><td>hopefully</td></tr><tr><td>how</td><td>howbeit</td><td>however</td><td>i'd</td><td>i'll</td></tr><tr><td>i'm</td><td>i've</td><td>ie</td><td>if</td><td>ignored</td></tr><tr><td>immediate</td><td>in</td><td>inasmuch</td><td>inc</td><td>indeed</td></tr><tr><td>indicate</td><td>indicated</td><td>indicates</td><td>inner</td><td>insofar</td></tr><tr><td>instead</td><td>into</td><td>inward</td><td>is</td><td>isn't</td></tr><tr><td>it</td><td>it'd</td><td>it'll</td><td>it's</td><td>its</td></tr><tr><td>itself</td><td>just</td><td>keep</td><td>keeps</td><td>kept</td></tr><tr><td>know</td><td>knows</td><td>known</td><td>last</td><td>lately</td></tr><tr><td>later</td><td>latter</td><td>latterly</td><td>least</td><td>less</td></tr><tr><td>lest</td><td>let</td><td>let's</td><td>like</td><td>liked</td></tr><tr><td>likely</td><td>little</td><td>look</td><td>looking</td><td>looks</td></tr><tr><td>ltd</td><td>mainly</td><td>many</td><td>may</td><td>maybe</td></tr><tr><td>me</td><td>mean</td><td>meanwhile</td><td>merely</td><td>might</td></tr><tr><td>more</td><td>moreover</td><td>most</td><td>mostly</td><td>much</td></tr><tr><td>must</td><td>my</td><td>myself</td><td>name</td><td>namely</td></tr><tr><td>nd</td><td>near</td><td>nearly</td><td>necessary</td><td>need</td></tr><tr><td>needs</td><td>neither</td><td>never</td><td>nevertheless</td><td>new</td></tr><tr><td>next</td><td>nine</td><td>no</td><td>nobody</td><td>non</td></tr><tr><td>none</td><td>noone</td><td>nor</td><td>normally</td><td>not</td></tr><tr><td>nothing</td><td>novel</td><td>now</td><td>nowhere</td><td>obviously</td></tr><tr><td>of</td><td>off</td><td>often</td><td>oh</td><td>ok</td></tr><tr><td>okay</td><td>old</td><td>on</td><td>once</td><td>one</td></tr><tr><td>ones</td><td>only</td><td>onto</td><td>or</td><td>other</td></tr><tr><td>others</td><td>otherwise</td><td>ought</td><td>our</td><td>ours</td></tr><tr><td>ourselves</td><td>out</td><td>outside</td><td>over</td><td>overall</td></tr><tr><td>own</td><td>particular</td><td>particularly</td><td>per</td><td>perhaps</td></tr><tr><td>placed</td><td>please</td><td>plus</td><td>possible</td><td>presumably</td></tr><tr><td>probably</td><td>provides</td><td>que</td><td>quite</td><td>qv</td></tr><tr><td>rather</td><td>rd</td><td>re</td><td>really</td><td>reasonably</td></tr><tr><td>regarding</td><td>regardless</td><td>regards</td><td>relatively</td><td>respectively</td></tr><tr><td>right</td><td>said</td><td>same</td><td>saw</td><td>say</td></tr><tr><td>saying</td><td>says</td><td>second</td><td>secondly</td><td>see</td></tr><tr><td>seeing</td><td>seem</td><td>seemed</td><td>seeming</td><td>seems</td></tr><tr><td>seen</td><td>self</td><td>selves</td><td>sensible</td><td>sent</td></tr><tr><td>serious</td><td>seriously</td><td>seven</td><td>several</td><td>shall</td></tr><tr><td>she</td><td>should</td><td>shouldn't</td><td>since</td><td>six</td></tr><tr><td>so</td><td>some</td><td>somebody</td><td>somehow</td><td>someone</td></tr><tr><td>something</td><td>sometime</td><td>sometimes</td><td>somewhat</td><td>somewhere</td></tr><tr><td>soon</td><td>sorry</td><td>specified</td><td>specify</td><td>specifying</td></tr><tr><td>still</td><td>sub</td><td>such</td><td>sup</td><td>sure</td></tr><tr><td>t's</td><td>take</td><td>taken</td><td>tell</td><td>tends</td></tr><tr><td>th</td><td>than</td><td>thank</td><td>thanks</td><td>thanx</td></tr><tr><td>that</td><td>that's</td><td>thats</td><td>the</td><td>their</td></tr><tr><td>theirs</td><td>them</td><td>themselves</td><td>then</td><td>thence</td></tr><tr><td>there</td><td>there's</td><td>thereafter</td><td>thereby</td><td>therefore</td></tr><tr><td>therein</td><td>theres</td><td>thereupon</td><td>these</td><td>they</td></tr><tr><td>they'd</td><td>they'll</td><td>they're</td><td>they've</td><td>think</td></tr><tr><td>third</td><td>this</td><td>thorough</td><td>thoroughly</td><td>those</td></tr><tr><td>though</td><td>three</td><td>through</td><td>throughout</td><td>thru</td></tr><tr><td>thus</td><td>to</td><td>together</td><td>too</td><td>took</td></tr><tr><td>toward</td><td>towards</td><td>tried</td><td>tries</td><td>truly</td></tr><tr><td>try</td><td>trying</td><td>twice</td><td>two</td><td>un</td></tr><tr><td>under</td><td>unfortunately</td><td>unless</td><td>unlikely</td><td>until</td></tr><tr><td>unto</td><td>up</td><td>upon</td><td>us</td><td>use</td></tr><tr><td>used</td><td>useful</td><td>uses</td><td>using</td><td>usually</td></tr><tr><td>value</td><td>various</td><td>very</td><td>via</td><td>viz</td></tr><tr><td>vs</td><td>want</td><td>wants</td><td>was</td><td>wasn't</td></tr><tr><td>way</td><td>we</td><td>we'd</td><td>we'll</td><td>we're</td></tr><tr><td>we've</td><td>welcome</td><td>well</td><td>went</td><td>were</td></tr><tr><td>weren't</td><td>what</td><td>what's</td><td>whatever</td><td>when</td></tr><tr><td>whence</td><td>whenever</td><td>where</td><td>where's</td><td>whereafter</td></tr><tr><td>whereas</td><td>whereby</td><td>wherein</td><td>whereupon</td><td>wherever</td></tr><tr><td>whether</td><td>which</td><td>while</td><td>whither</td><td>who</td></tr><tr><td>who's</td><td>whoever</td><td>whole</td><td>whom</td><td>whose</td></tr><tr><td>why</td><td>will</td><td>willing</td><td>wish</td><td>with</td></tr><tr><td>within</td><td>without</td><td>won't</td><td>wonder</td><td>would</td></tr><tr><td>would</td><td>wouldn't</td><td>yes</td><td>yet</td><td>you</td></tr><tr><td>you'd</td><td>you'll</td><td>you're</td><td>you've</td><td>your</td></tr><tr><td>yours</td><td>yourself</td><td>yourselves</td><td>zero</td><td> </td></tr></tbody></table></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="fulltext-restrictions"></a>Full-Text Restrictions</h3></div></div></div><div class="itemizedlist"><ul type="disc"><li><p>
            Full-text searches are supported for
            <code class="literal">MyISAM</code> tables only.
          </p></li><li><p>
            Full-text searches can be used with most multi-byte
            character sets. The exception is that for Unicode, the
            <code class="literal">utf8</code> character set can be used, but not
            the <code class="literal">ucs2</code> character set.
          </p></li><li><p>
            Ideographic languages such as Chinese and Japanese do not
            have word delimiters. Therefore, the
            <code class="literal">FULLTEXT</code> parser <span class="emphasis"><em>cannot
            determine where words begin and end in these and other such
            languages</em></span>. The implications of this and some
            workarounds for the problem are described in
            <a href="mysqlqb_functions.html#fulltext-search" title="Full-Text Search Functions">Full-Text Search Functions</a>.
          </p></li><li><p>
            Although the use of multiple character sets within a single
            table is supported, all columns in a
            <code class="literal">FULLTEXT</code> index must use the same
            character set and collation.
          </p></li><li><p>
            The <code class="literal">MATCH()</code> column list must match
            exactly the column list in some <code class="literal">FULLTEXT</code>
            index definition for the table, unless this
            <code class="literal">MATCH()</code> is <code class="literal">IN BOOLEAN
            MODE</code>. Boolean-mode searches can be done on
            non-indexed columns, although they are likely to be slow.
          </p></li><li><p>
            The argument to <code class="literal">AGAINST()</code> must be a
            constant string.
          </p></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="fulltext-fine-tuning"></a>Fine-Tuning MySQL Full-Text Search</h3></div></div></div><p>
        MySQL's full-text search capability has few user-tunable
        parameters. You can exert more control over full-text searching
        behavior if you have a MySQL source distribution because some
        changes require source code modifications. See
        <a href="http://dev.mysql.com/doc/refman/5.0/en/installing-source.html" target="_top">MySQL Installation Using a Source Distribution</a>.
      </p><p>
        Note that full-text search is carefully tuned for the most
        effectiveness. Modifying the default behavior in most cases can
        actually decrease effectiveness. <span class="emphasis"><em>Do not alter the
        MySQL sources unless you know what you are doing</em></span>.
      </p><p>
        Most full-text variables described in this section must be set
        at server startup time. A server restart is required to change
        them; they cannot be modified while the server is running.
      </p><p>
        Some variable changes require that you rebuild the
        <code class="literal">FULLTEXT</code> indexes in your tables. Instructions
        for doing this are given at the end of this section.
      </p><div class="itemizedlist"><ul type="disc"><li><p>
            The minimum and maximum lengths of words to be indexed are
            defined by the <code class="literal">ft_min_word_len</code> and
            <code class="literal">ft_max_word_len</code> system variables. (See
            <a href="http://dev.mysql.com/doc/refman/5.0/en/server-system-variables.html" target="_top">System Variables</a>.) The default
            minimum value is four characters; the default maximum is
            version dependent. If you change either value, you must
            rebuild your <code class="literal">FULLTEXT</code> indexes. For
            example, if you want three-character words to be searchable,
            you can set the <code class="literal">ft_min_word_len</code> variable
            by putting the following lines in an option file:
          </p><pre class="programlisting">[mysqld]
ft_min_word_len=3
</pre><p>
            Then you must restart the server and rebuild your
            <code class="literal">FULLTEXT</code> indexes. Note particularly the
            remarks regarding <span><strong class="command">myisamchk</strong></span> in the
            instructions following this list.
          </p></li><li><p>
            <a id="id2846051" class="indexterm"></a>

            <a id="id2846062" class="indexterm"></a>

            To override the default stopword list, set the
            <code class="literal">ft_stopword_file</code> system variable. (See
            <a href="http://dev.mysql.com/doc/refman/5.0/en/server-system-variables.html" target="_top">System Variables</a>.) The variable
            value should be the pathname of the file containing the
            stopword list, or the empty string to disable stopword
            filtering. After changing the value of this variable or the
            contents of the stopword file, restart the server and
            rebuild your <code class="literal">FULLTEXT</code> indexes.
          </p><p>
            The stopword list is free-form. That is, you may use any
            non-alphanumeric character such as newline, space, or comma
            to separate stopwords. Exceptions are the underscore
            character (‘<code class="literal">_</code>’) and a single
            apostrophe (‘<code class="literal">'</code>’) which are
            treated as part of a word. The character set of the stopword
            list is the server's default character set; see
            <a href="http://dev.mysql.com/doc/refman/5.0/en/charset-server.html" target="_top">Server Character Set and Collation</a>.
          </p></li><li><p>
            The 50% threshold for natural language searches is
            determined by the particular weighting scheme chosen. To
            disable it, look for the following line in
            <code class="filename">myisam/ftdefs.h</code>:
          </p><pre class="programlisting">#define GWS_IN_USE GWS_PROB
</pre><p>
            Change that line to this:
          </p><pre class="programlisting">#define GWS_IN_USE GWS_FREQ
</pre><p>
            Then recompile MySQL. There is no need to rebuild the
            indexes in this case. <span class="bold"><strong>Note</strong></span>:
            By making this change, you <span class="emphasis"><em>severely</em></span>
            decrease MySQL's ability to provide adequate relevance
            values for the <code class="literal">MATCH()</code> function. If you
            really need to search for such common words, it would be
            better to search using <code class="literal">IN BOOLEAN MODE</code>
            instead, which does not observe the 50% threshold.
          </p></li><li><p>
            To change the operators used for boolean full-text searches,
            set the <code class="literal">ft_boolean_syntax</code> system
            variable. This variable can be changed while the server is
            running, but you must have the <code class="literal">SUPER</code>
            privilege to do so. No rebuilding of indexes is necessary in
            this case. See <a href="http://dev.mysql.com/doc/refman/5.0/en/server-system-variables.html" target="_top">System Variables</a>,
            which describes the rules governing how to set this
            variable.
          </p></li><li><p>
            If you want to change the set of characters that are
            considered word characters, you can do so in two ways.
            Suppose that you want to treat the hyphen character ('-') as
            a word character. Use either of these methods:
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                Modify the MySQL source: In
                <code class="filename">myisam/ftdefs.h</code>, see the
                <code class="literal">true_word_char()</code> and
                <code class="literal">misc_word_char()</code> macros. Add
                <code class="literal">'-'</code> to one of those macros and
                recompile MySQL.
              </p></li><li><p>
                Modify a character set file: This requires no
                recompilation. The <code class="literal">true_word_char()</code>
                macro uses a “<span class="quote">character type</span>” table to
                distinguish letters and numbers from other characters. .
                You can edit the
                <code class="literal">&lt;ctype&gt;&lt;map&gt;</code> contents in
                one of the character set XML files to specify that
                <code class="literal">'-'</code> is a “<span class="quote">letter.</span>” Then
                use the given character set for your
                <code class="literal">FULLTEXT</code> indexes.
              </p></li></ul></div><p>
            After making the modification, you must rebuild the indexes
            for each table that contains any <code class="literal">FULLTEXT</code>
            indexes.
          </p></li></ul></div><p>
        If you modify full-text variables that affect indexing
        (<code class="literal">ft_min_word_len</code>,
        <code class="literal">ft_max_word_len</code>, or
        <code class="literal">ft_stopword_file</code>), or if you change the
        stopword file itself, you must rebuild your
        <code class="literal">FULLTEXT</code> indexes after making the changes and
        restarting the server. To rebuild the indexes in this case, it
        is sufficient to do a <code class="literal">QUICK</code> repair operation:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>REPAIR TABLE <em class="replaceable"><code>tbl_name</code></em> QUICK;</code></strong>
</pre><p>
        Each table that contains any <code class="literal">FULLTEXT</code> index
        must be repaired as just shown. Otherwise, queries for the table
        may yield incorrect results, and modifications to the table will
        cause the server to see the table as corrupt and in need of
        repair.
      </p><p>
        Note that if you use <span><strong class="command">myisamchk</strong></span> to perform an
        operation that modifies table indexes (such as repair or
        analyze), the <code class="literal">FULLTEXT</code> indexes are rebuilt
        using the <span class="emphasis"><em>default</em></span> full-text parameter
        values for minimum word length, maximum word length, and
        stopword file unless you specify otherwise. This can result in
        queries failing.
      </p><p>
        The problem occurs because these parameters are known only by
        the server. They are not stored in <code class="literal">MyISAM</code>
        index files. To avoid the problem if you have modified the
        minimum or maximum word length or stopword file values used by
        the server, specify the same <code class="literal">ft_min_word_len</code>,
        <code class="literal">ft_max_word_len</code>, and
        <code class="literal">ft_stopword_file</code> values to
        <span><strong class="command">myisamchk</strong></span> that you use for
        <span><strong class="command">mysqld</strong></span>. For example, if you have set the
        minimum word length to 3, you can repair a table with
        <span><strong class="command">myisamchk</strong></span> like this:
      </p><pre class="programlisting">shell&gt; <strong class="userinput"><code>myisamchk --recover --ft_min_word_len=3 <em class="replaceable"><code>tbl_name</code></em>.MYI</code></strong>
</pre><p>
        To ensure that <span><strong class="command">myisamchk</strong></span> and the server use
        the same values for full-text parameters, place each one in both
        the <code class="literal">[mysqld]</code> and
        <code class="literal">[myisamchk]</code> sections of an option file:
      </p><pre class="programlisting">[mysqld]
ft_min_word_len=3

[myisamchk]
ft_min_word_len=3
</pre><p>
        An alternative to using <span><strong class="command">myisamchk</strong></span> is to use
        the <code class="literal">REPAIR TABLE</code>, <code class="literal">ANALYZE
        TABLE</code>, <code class="literal">OPTIMIZE TABLE</code>, or
        <code class="literal">ALTER TABLE</code> statements. These statements are
        performed by the server, which knows the proper full-text
        parameter values to use.
      </p></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="cast-functions"></a>Cast Functions and Operators</h2></div></div></div><a id="id2846471" class="indexterm"></a><a id="id2846478" class="indexterm"></a><a id="id2846485" class="indexterm"></a><a id="id2846495" class="indexterm"></a><a id="id2846502" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#operator_binary"><code class="literal">BINARY</code></a></td><td>Cast a string to a binary string</td></tr><tr><td><a href="mysqlqb_functions.html#function_cast"><code class="literal">CAST()</code></a></td><td>Cast a value as a certain type</td></tr></tbody></table></div><div class="itemizedlist"><ul type="disc"><li><p><a id="operator_binary"></a>
          <a id="id2846593" class="indexterm"></a>

          <code class="literal">BINARY</code>
        </p><p>
          The <code class="literal">BINARY</code> operator casts the string
          following it to a binary string. This is an easy way to force
          a column comparison to be done byte by byte rather than
          character by character. This causes the comparison to be case
          sensitive even if the column isn't defined as
          <code class="literal">BINARY</code> or <code class="literal">BLOB</code>.
          <code class="literal">BINARY</code> also causes trailing spaces to be
          significant.
        </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 'a' = 'A';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT BINARY 'a' = 'A';</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 'a' = 'a ';</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT BINARY 'a' = 'a ';</code></strong>
        -&gt; 0
</pre><p>
          In a comparison, <code class="literal">BINARY</code> affects the entire
          operation; it can be given before either operand with the same
          result.
        </p><p>
          <code class="literal">BINARY <em class="replaceable"><code>str</code></em></code> is
          shorthand for <code class="literal">CAST(<em class="replaceable"><code>str</code></em> AS
          BINARY)</code>.
        </p><p>
          Note that in some contexts, if you cast an indexed column to
          <code class="literal">BINARY</code>, MySQL is not able to use the index
          efficiently.
        </p></li><li><p><a id="function_cast"></a>
          <a id="id2846730" class="indexterm"></a>

          <a id="id2846736" class="indexterm"></a>

          <code class="literal">CAST(<em class="replaceable"><code>expr</code></em> AS
          <em class="replaceable"><code>type</code></em>)</code>,
          <code class="literal">CONVERT(<em class="replaceable"><code>expr</code></em>,<em class="replaceable"><code>type</code></em>)</code>,
          <code class="literal">CONVERT(<em class="replaceable"><code>expr</code></em> USING
          <em class="replaceable"><code>transcoding_name</code></em>)</code>
        </p><p>
          The <code class="literal">CAST()</code> and <code class="literal">CONVERT()</code>
          functions take a value of one type and produce a value of
          another type.
        </p><p>
          The <em class="replaceable"><code>type</code></em> can be one of the
          following values:
        </p><div class="itemizedlist"><ul type="circle"><li><p>
              <code class="literal">BINARY[(<em class="replaceable"><code>N</code></em>)]</code>
            </p></li><li><p>
              <code class="literal">CHAR[(<em class="replaceable"><code>N</code></em>)]</code>
            </p></li><li><p>
              <code class="literal">DATE</code>
            </p></li><li><p>
              <code class="literal">DATETIME</code>
            </p></li><li><p>
              <code class="literal">DECIMAL</code>
            </p></li><li><p>
              <code class="literal">SIGNED [INTEGER]</code>
            </p></li><li><p>
              <code class="literal">TIME</code>
            </p></li><li><p>
              <code class="literal">UNSIGNED [INTEGER]</code>
            </p></li></ul></div><p>
          <code class="literal">BINARY</code> produces a string with the
          <code class="literal">BINARY</code> data type. See
          <a href="http://dev.mysql.com/doc/refman/5.0/en/binary-varbinary.html" target="_top">The <code class="literal">BINARY</code> and <code class="literal">VARBINARY</code> Types</a> for a description of how
          this affects comparisons. If the optional length
          <em class="replaceable"><code>N</code></em> is given,
          <code class="literal">BINARY(<em class="replaceable"><code>N</code></em>)</code> causes
          the cast to use no more than <em class="replaceable"><code>N</code></em>
          bytes of the argument. As of MySQL 5.0.17, values shorter than
          <em class="replaceable"><code>N</code></em> bytes are padded with
          <code class="literal">0x00</code> bytes to a length of
          <em class="replaceable"><code>N</code></em>.
        </p><p>
          <code class="literal">CHAR(<em class="replaceable"><code>N</code></em>)</code> causes
          the cast to use no more than <em class="replaceable"><code>N</code></em>
          characters of the argument.
        </p><p>
          The <code class="literal">DECIMAL</code> type is available as of MySQL
          5.0.8.
        </p><p>
          <code class="literal">CAST()</code> and <code class="literal">CONVERT(... USING
          ...)</code> are standard SQL syntax. The
          non-<code class="literal">USING</code> form of
          <code class="literal">CONVERT()</code> is ODBC syntax.
        </p><p>
          <code class="literal">CONVERT()</code> with <code class="literal">USING</code> is
          used to convert data between different character sets. In
          MySQL, transcoding names are the same as the corresponding
          character set names. For example, this statement converts the
          string <code class="literal">'abc'</code> in the default character set
          to the corresponding string in the <code class="literal">utf8</code>
          character set:
        </p><pre class="programlisting">SELECT CONVERT('abc' USING utf8);
</pre></li></ul></div><p>
      Normally, you cannot compare a <code class="literal">BLOB</code> value or
      other binary string in case-insensitive fashion because binary
      strings have no character set, and thus no concept of lettercase.
      To perform a case-insensitive comparison, use the
      <code class="literal">CONVERT()</code> function to convert the value to a
      non-binary string. If the character set of the result has a
      case-insensitive collation, the <code class="literal">LIKE</code> operation
      is not case sensitive:
    </p><pre class="programlisting">SELECT 'A' LIKE CONVERT(<em class="replaceable"><code>blob_col</code></em> USING latin1) FROM <em class="replaceable"><code>tbl_name</code></em>;
</pre><p>
      To use a different character set, substitute its name for
      <code class="literal">latin1</code> in the preceding statement. To ensure
      that a case-insensitive collation is used, specify a
      <code class="literal">COLLATE</code> clause following the
      <code class="literal">CONVERT()</code> call.
    </p><p>
      <code class="literal">CONVERT()</code> can be used more generally for
      comparing strings that are represented in different character
      sets.
    </p><p>
      The cast functions are useful when you want to create a column
      with a specific type in a <code class="literal">CREATE ... SELECT</code>
      statement:
    </p><pre class="programlisting">CREATE TABLE new_table SELECT CAST('2000-01-01' AS DATE);
</pre><p>
      The functions also can be useful for sorting
      <code class="literal">ENUM</code> columns in lexical order. Normally,
      sorting of <code class="literal">ENUM</code> columns occurs using the
      internal numeric values. Casting the values to
      <code class="literal">CHAR</code> results in a lexical sort:
    </p><pre class="programlisting">SELECT <em class="replaceable"><code>enum_col</code></em> FROM <em class="replaceable"><code>tbl_name</code></em> ORDER BY CAST(<em class="replaceable"><code>enum_col</code></em> AS CHAR);
</pre><p>
      <code class="literal">CAST(<em class="replaceable"><code>str</code></em> AS BINARY)</code>
      is the same thing as <code class="literal">BINARY
      <em class="replaceable"><code>str</code></em></code>.
      <code class="literal">CAST(<em class="replaceable"><code>expr</code></em> AS CHAR)</code>
      treats the expression as a string with the default character set.
    </p><p>
      <code class="literal">CAST()</code> also changes the result if you use it as
      part of a more complex expression such as <code class="literal">CONCAT('Date:
      ',CAST(NOW() AS DATE))</code>.
    </p><p>
      You should not use <code class="literal">CAST()</code> to extract data in
      different formats but instead use string functions like
      <code class="literal">LEFT()</code> or <code class="literal">EXTRACT()</code>. See
      <a href="mysqlqb_functions.html#date-and-time-functions" title="Date and Time Functions">Date and Time Functions</a>.
    </p><p>
      To cast a string to a numeric value in numeric context, you
      normally do not have to do anything other than to use the string
      value as though it were a number:
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1+'1';</code></strong>
       -&gt; 2
</pre><p>
      If you use a number in string context, the number automatically is
      converted to a <code class="literal">BINARY</code> string.
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CONCAT('hello you ',2);</code></strong>
        -&gt; 'hello you 2'
</pre><p>
      MySQL supports arithmetic with both signed and unsigned 64-bit
      values. If you are using numeric operators (such as
      <code class="literal">+</code> or <code class="literal">-</code>) and one of the
      operands is an unsigned integer, the result is unsigned. You can
      override this by using the <code class="literal">SIGNED</code> and
      <code class="literal">UNSIGNED</code> cast operators to cast the operation
      to a signed or unsigned 64-bit integer, respectively.
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CAST(1-2 AS UNSIGNED)</code></strong>
        -&gt; 18446744073709551615
mysql&gt; <strong class="userinput"><code>SELECT CAST(CAST(1-2 AS UNSIGNED) AS SIGNED);</code></strong>
        -&gt; -1
</pre><p>
      Note that if either operand is a floating-point value, the result
      is a floating-point value and is not affected by the preceding
      rule. (In this context, <code class="literal">DECIMAL</code> column values
      are regarded as floating-point values.)
    </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CAST(1 AS UNSIGNED) - 2.0;</code></strong>
        -&gt; -1.0
</pre><p>
      If you are using a string in an arithmetic operation, this is
      converted to a floating-point number.
    </p><p>
      If you convert a “<span class="quote">zero</span>” date string to a date,
      <code class="literal">CONVERT()</code> and <code class="literal">CAST()</code> return
      <code class="literal">NULL</code> when the <code class="literal">NO_ZERO_DATE</code>
      SQL mode is enabled. As of MySQL 5.0.4, they also produce a
      warning.
    </p></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="other-functions"></a>Other Functions</h2></div></div></div><div class="toc"><dl><dt><span class="section"><a href="mysqlqb_functions.html#bit-functions">Bit Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#encryption-functions">Encryption and Compression Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#information-functions">Information Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#miscellaneous-functions">Miscellaneous Functions</a></span></dt></dl></div><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_aes-encrypt"><code class="literal">AES_DECRYPT()</code></a></td><td>Decrypt using AES</td></tr><tr><td><a href="mysqlqb_functions.html#function_aes-encrypt"><code class="literal">AES_ENCRYPT()</code></a></td><td>Encrypt using AES</td></tr><tr><td><a href="mysqlqb_functions.html#function_benchmark"><code class="literal">BENCHMARK()</code></a></td><td>Repeatedly execute an expression</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-and"><code class="literal">BIT_AND()</code></a></td><td>Return bitwise and</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-count"><code class="literal">BIT_COUNT()</code></a></td><td>Return the number of bits that are set</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-or"><code class="literal">BIT_OR()</code></a></td><td>Return bitwise or</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-xor"><code class="literal">BIT_XOR()</code></a></td><td>Return bitwise xor</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-and"><code class="literal">&amp;</code></a></td><td>Bitwise AND</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-or"><code class="literal">|</code></a></td><td>Bitwise OR</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-xor"><code class="literal">^</code></a></td><td>Bitwise XOR</td></tr><tr><td><a href="mysqlqb_functions.html#function_charset"><code class="literal">CHARSET()</code></a></td><td>Return the character set of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_coercibility"><code class="literal">COERCIBILITY()</code></a></td><td>Return the collation coercibility value of the string  argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_collation"><code class="literal">COLLATION()</code></a></td><td>Return the collation of the string argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_compress"><code class="literal">COMPRESS()</code></a></td><td>Return result as a binary string</td></tr><tr><td><a href="mysqlqb_functions.html#function_connection-id"><code class="literal">CONNECTION_ID()</code></a></td><td>Return the connection ID (thread ID)   for the connection</td></tr><tr><td><a href="mysqlqb_functions.html#function_count"><code class="literal">COUNT()</code></a></td><td>Return a count of the number of rows returned</td></tr><tr><td><a href="mysqlqb_functions.html#function_current-user"><code class="literal">CURRENT_USER()</code>, <code class="literal">CURRENT_USER</code></a></td><td>Return the username and hostname combination</td></tr><tr><td><a href="mysqlqb_functions.html#function_database"><code class="literal">DATABASE()</code></a></td><td>Return the default (current) database name</td></tr><tr><td><a href="mysqlqb_functions.html#function_decode"><code class="literal">DECODE()</code></a></td><td>Decodes a string encrypted using ENCODE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_default"><code class="literal">DEFAULT()</code></a></td><td>Return the default value for a table column</td></tr><tr><td><a href="mysqlqb_functions.html#function_des-decrypt"><code class="literal">DES_DECRYPT()</code></a></td><td>Decrypt a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_des-encrypt"><code class="literal">DES_ENCRYPT()</code></a></td><td>Decrypt a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_encode"><code class="literal">ENCODE()</code></a></td><td>Encode a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_encrypt"><code class="literal">ENCRYPT()</code></a></td><td>Encrypt a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_format"><code class="literal">FORMAT()</code></a></td><td>Return a number formatted to specified number of decimal places</td></tr><tr><td><a href="mysqlqb_functions.html#function_found-rows"><code class="literal">FOUND_ROWS()</code></a></td><td>For a SELECT with a LIMIT clause, the  number of rows that would be returned were there no LIMIT clause</td></tr><tr><td><a href="mysqlqb_functions.html#function_get-lock"><code class="literal">GET_LOCK()</code></a></td><td>Get a named lock</td></tr><tr><td><a href="mysqlqb_functions.html#function_inet-aton"><code class="literal">INET_ATON()</code></a></td><td>Return the numeric value of an IP address</td></tr><tr><td><a href="mysqlqb_functions.html#function_inet-ntoa"><code class="literal">INET_NTOA()</code></a></td><td>Return the IP address from a numeric value</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-free-lock"><code class="literal">IS_FREE_LOCK()</code></a></td><td>Checks whether the named lock  is free</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-used-lock"><code class="literal">IS_USED_LOCK()</code></a></td><td>Checks whether the named lock is in use.  Return connection identifier if true.</td></tr><tr><td><a href="mysqlqb_functions.html#function_last-insert-id"><code class="literal">LAST_INSERT_ID()</code></a></td><td>Value of the AUTOINCREMENT column   for the last INSERT</td></tr><tr><td><a href="mysqlqb_functions.html#operator_left-shift"><code class="literal">&lt;&lt;</code></a></td><td>Left shift</td></tr><tr><td><a href="mysqlqb_functions.html#function_master-pos-wait"><code class="literal">MASTER_POS_WAIT()</code></a></td><td>Block until the slave has read and  applied all updates up to the specified position</td></tr><tr><td><a href="mysqlqb_functions.html#function_md5"><code class="literal">MD5()</code></a></td><td>Calculate MD5 checksum</td></tr><tr><td><a href="mysqlqb_functions.html#function_name-const"><code class="literal">NAME_CONST()</code></a>(v5.0.12)</td><td>Causes the column to have the given name</td></tr><tr><td><a href="mysqlqb_functions.html#function_old-password"><code class="literal">OLD_PASSWORD()</code></a></td><td>Return the value of the  old (pre-4.1) implementation of PASSWORD</td></tr><tr><td><a href="mysqlqb_functions.html#function_password"><code class="literal">PASSWORD()</code></a></td><td>Calculate and return a password string</td></tr><tr><td><a href="mysqlqb_functions.html#function_release-lock"><code class="literal">RELEASE_LOCK()</code></a></td><td>Releases the named lock</td></tr><tr><td><a href="mysqlqb_functions.html#operator_right-shift"><code class="literal">&gt;&gt;</code></a></td><td>Right shift</td></tr><tr><td><a href="mysqlqb_functions.html#function_row-count"><code class="literal">ROW_COUNT()</code></a></td><td>The number of rows  updated</td></tr><tr><td><a href="mysqlqb_functions.html#function_schema"><code class="literal">SCHEMA()</code></a></td><td>A synonym for DATABASE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_session-user"><code class="literal">SESSION_USER()</code></a></td><td>Synonym for USER()</td></tr><tr><td><a href="mysqlqb_functions.html#function_sha1"><code class="literal">SHA1()</code>, <code class="literal">SHA()</code></a></td><td>Calculate an SHA-1 160-bit checksum</td></tr><tr><td><a href="mysqlqb_functions.html#function_sleep"><code class="literal">SLEEP()</code></a></td><td>Sleep for a number of seconds</td></tr><tr><td><a href="mysqlqb_functions.html#function_system-user"><code class="literal">SYSTEM_USER()</code></a></td><td>Synonym for USER()</td></tr><tr><td><a href="mysqlqb_functions.html#operator_tilde"><code class="literal">~</code></a></td><td>Invert bits</td></tr><tr><td><a href="mysqlqb_functions.html#function_uncompress"><code class="literal">UNCOMPRESS()</code></a></td><td>Uncompress a string compressed</td></tr><tr><td><a href="mysqlqb_functions.html#function_uncompressed-length"><code class="literal">UNCOMPRESSED_LENGTH()</code></a></td><td>Return the length of a string before compression</td></tr><tr><td><a href="mysqlqb_functions.html#function_user"><code class="literal">USER()</code></a></td><td>Return the current username and hostname</td></tr><tr><td><a href="mysqlqb_functions.html#function_uuid"><code class="literal">UUID()</code></a></td><td>Return a Universal Unique Identifier (UUID)</td></tr><tr><td><a href="mysqlqb_functions.html#function_values"><code class="literal">VALUES()</code></a></td><td>Defines the values to be used during an INSERT</td></tr></tbody></table></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="bit-functions"></a>Bit Functions</h3></div></div></div><a id="id2847948" class="indexterm"></a><a id="id2847955" class="indexterm"></a><a id="id2847966" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_bit-and"><code class="literal">BIT_AND()</code></a></td><td>Return bitwise and</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-count"><code class="literal">BIT_COUNT()</code></a></td><td>Return the number of bits that are set</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-or"><code class="literal">BIT_OR()</code></a></td><td>Return bitwise or</td></tr><tr><td><a href="mysqlqb_functions.html#function_bit-xor"><code class="literal">BIT_XOR()</code></a></td><td>Return bitwise xor</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-and"><code class="literal">&amp;</code></a></td><td>Bitwise AND</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-or"><code class="literal">|</code></a></td><td>Bitwise OR</td></tr><tr><td><a href="mysqlqb_functions.html#operator_bitwise-xor"><code class="literal">^</code></a></td><td>Bitwise XOR</td></tr><tr><td><a href="mysqlqb_functions.html#operator_left-shift"><code class="literal">&lt;&lt;</code></a></td><td>Left shift</td></tr><tr><td><a href="mysqlqb_functions.html#operator_right-shift"><code class="literal">&gt;&gt;</code></a></td><td>Right shift</td></tr><tr><td><a href="mysqlqb_functions.html#operator_tilde"><code class="literal">~</code></a></td><td>Invert bits</td></tr></tbody></table></div><p>
        MySQL uses <code class="literal">BIGINT</code> (64-bit) arithmetic for bit
        operations, so these operators have a maximum range of 64 bits.
      </p><div class="itemizedlist"><ul type="disc"><li><p><a id="operator_bitwise-or"></a>
            <a id="id2848156" class="indexterm"></a>

            <a id="id2848162" class="indexterm"></a>

            <code class="literal">|</code>
          </p><p>
            Bitwise OR:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 29 | 15;</code></strong>
        -&gt; 31
</pre><p>
            The result is an unsigned 64-bit integer.
          </p></li><li><p><a id="operator_bitwise-and"></a>
            <a id="id2848230" class="indexterm"></a>

            <a id="id2848237" class="indexterm"></a>

            <code class="literal">&amp;</code>
          </p><p>
            Bitwise AND:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 29 &amp; 15;</code></strong>
        -&gt; 13
</pre><p>
            The result is an unsigned 64-bit integer.
          </p></li><li><p><a id="operator_bitwise-xor"></a>
            <a id="id2848305" class="indexterm"></a>

            <a id="id2848312" class="indexterm"></a>

            <code class="literal">^</code>
          </p><p>
            Bitwise XOR:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 ^ 1;</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT 1 ^ 0;</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT 11 ^ 3;</code></strong>
        -&gt; 8
</pre><p>
            The result is an unsigned 64-bit integer.
          </p></li><li><p><a id="operator_left-shift"></a>
            <a id="id2848388" class="indexterm"></a>

            <code class="literal">&lt;&lt;</code>
          </p><p>
            Shifts a longlong (<code class="literal">BIGINT</code>) number to the
            left.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 1 &lt;&lt; 2;</code></strong>
        -&gt; 4
</pre><p>
            The result is an unsigned 64-bit integer.
          </p></li><li><p><a id="operator_right-shift"></a>
            <a id="id2848458" class="indexterm"></a>

            <code class="literal">&gt;&gt;</code>
          </p><p>
            Shifts a longlong (<code class="literal">BIGINT</code>) number to the
            right.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 4 &gt;&gt; 2;</code></strong>
        -&gt; 1
</pre><p>
            The result is an unsigned 64-bit integer.
          </p></li><li><p><a id="operator_tilde"></a>
            <a id="id2848527" class="indexterm"></a>

            <code class="literal">~</code>
          </p><p>
            Invert all bits.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT 5 &amp; ~1;</code></strong>
        -&gt; 4
</pre><p>
            The result is an unsigned 64-bit integer.
          </p></li><li><p><a id="function_bit-count"></a>
            <a id="id2848591" class="indexterm"></a>

            <code class="literal">BIT_COUNT(<em class="replaceable"><code>N</code></em>)</code>
          </p><p>
            Returns the number of bits that are set in the argument
            <em class="replaceable"><code>N</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT BIT_COUNT(29), BIT_COUNT(b'101010');</code></strong>
        -&gt; 4, 3
</pre></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="encryption-functions"></a>Encryption and Compression Functions</h3></div></div></div><a id="id2848652" class="indexterm"></a><a id="id2848659" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_aes-encrypt"><code class="literal">AES_DECRYPT()</code></a></td><td>Decrypt using AES</td></tr><tr><td><a href="mysqlqb_functions.html#function_aes-encrypt"><code class="literal">AES_ENCRYPT()</code></a></td><td>Encrypt using AES</td></tr><tr><td><a href="mysqlqb_functions.html#function_compress"><code class="literal">COMPRESS()</code></a></td><td>Return result as a binary string</td></tr><tr><td><a href="mysqlqb_functions.html#function_decode"><code class="literal">DECODE()</code></a></td><td>Decodes a string encrypted using ENCODE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_des-decrypt"><code class="literal">DES_DECRYPT()</code></a></td><td>Decrypt a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_des-encrypt"><code class="literal">DES_ENCRYPT()</code></a></td><td>Decrypt a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_encode"><code class="literal">ENCODE()</code></a></td><td>Encode a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_encrypt"><code class="literal">ENCRYPT()</code></a></td><td>Encrypt a string</td></tr><tr><td><a href="mysqlqb_functions.html#function_md5"><code class="literal">MD5()</code></a></td><td>Calculate MD5 checksum</td></tr><tr><td><a href="mysqlqb_functions.html#function_old-password"><code class="literal">OLD_PASSWORD()</code></a></td><td>Return the value of the  old (pre-4.1) implementation of PASSWORD</td></tr><tr><td><a href="mysqlqb_functions.html#function_password"><code class="literal">PASSWORD()</code></a></td><td>Calculate and return a password string</td></tr><tr><td><a href="mysqlqb_functions.html#function_sha1"><code class="literal">SHA1()</code>, <code class="literal">SHA()</code></a></td><td>Calculate an SHA-1 160-bit checksum</td></tr><tr><td><a href="mysqlqb_functions.html#function_uncompress"><code class="literal">UNCOMPRESS()</code></a></td><td>Uncompress a string compressed</td></tr><tr><td><a href="mysqlqb_functions.html#function_uncompressed-length"><code class="literal">UNCOMPRESSED_LENGTH()</code></a></td><td>Return the length of a string before compression</td></tr></tbody></table></div><p>
        The functions in this section perform encryption and decryption,
        and compression and uncompression:
      </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>Compression or encryption</strong></span></td><td><span class="bold"><strong>Uncompression or decryption</strong></span></td></tr><tr><td><a href="mysqlqb_functions.html#function_aes-encrypt">AES_ENCRYT()</a></td><td><a href="mysqlqb_functions.html#function_aes-encrypt">AES_DECRYPT()</a></td></tr><tr><td><a href="mysqlqb_functions.html#function_compress">COMPRESS()</a></td><td><a href="mysqlqb_functions.html#function_uncompress">UNCOMPRESS()</a></td></tr><tr><td><a href="mysqlqb_functions.html#function_encode">ENCODE()</a></td><td><a href="mysqlqb_functions.html#function_decode">DECODE()</a></td></tr><tr><td><a href="mysqlqb_functions.html#function_des-encrypt">DES_ENCRYPT()</a></td><td><a href="mysqlqb_functions.html#function_des-decrypt">DES_DECRYPT()</a></td></tr><tr><td><a href="mysqlqb_functions.html#function_encrypt">ENCRYPT()</a></td><td>Not available</td></tr><tr><td><a href="mysqlqb_functions.html#function_md5">MD5()</a></td><td>Not available</td></tr><tr><td><a href="mysqlqb_functions.html#function_old-password">OLD_PASSWORD()</a></td><td>Not available</td></tr><tr><td><a href="mysqlqb_functions.html#function_password">PASSWORD()</a></td><td>Not available</td></tr><tr><td><a href="mysqlqb_functions.html#function_sha1">SHA() or SHA1()</a></td><td>Not available</td></tr><tr><td>Not available</td><td><a href="mysqlqb_functions.html#function_uncompressed-length">UNCOMPRESSED_LENGTH()</a></td></tr></tbody></table></div><p>
        <span class="bold"><strong>Note</strong></span>: The encryption and
        compression functions return binary strings. For many of these
        functions, the result might contain arbitrary byte values. If
        you want to store these results, use a <code class="literal">BLOB</code>
        column rather than a <code class="literal">CHAR</code> or (before MySQL
        5.0.3) <code class="literal">VARCHAR</code> column to avoid potential
        problems with trailing space removal that would change data
        values.
      </p><p>
        <span class="bold"><strong>Note</strong></span>: Exploits for the MD5 and
        SHA-1 algorithms have become known. You may wish to consider
        using one of the other encryption functions described in this
        section instead.
      </p><div class="itemizedlist"><ul type="disc"><li><p><a id="function_aes-encrypt"></a>
            <a id="id2849133" class="indexterm"></a>

            <a id="id2849140" class="indexterm"></a>

            <code class="literal">AES_ENCRYPT(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>key_str</code></em>)</code>,
            <code class="literal">AES_DECRYPT(<em class="replaceable"><code>crypt_str</code></em>,<em class="replaceable"><code>key_str</code></em>)</code>
          </p><p>
            These functions allow encryption and decryption of data
            using the official AES (Advanced Encryption Standard)
            algorithm, previously known as “<span class="quote">Rijndael.</span>”
            Encoding with a 128-bit key length is used, but you can
            extend it up to 256 bits by modifying the source. We chose
            128 bits because it is much faster and it is secure enough
            for most purposes.
          </p><p>
            <code class="literal">AES_ENCRYPT()</code> encrypts a string and
            returns a binary string. <code class="literal">AES_DECRYPT()</code>
            decrypts the encrypted string and returns the original
            string. The input arguments may be any length. If either
            argument is <code class="literal">NULL</code>, the result of this
            function is also <code class="literal">NULL</code>.
          </p><p>
            Because AES is a block-level algorithm, padding is used to
            encode uneven length strings and so the result string length
            may be calculated using this formula:
          </p><pre class="programlisting">16 × (trunc(<em class="replaceable"><code>string_length</code></em> / 16) + 1)
</pre><p>
            If <code class="literal">AES_DECRYPT()</code> detects invalid data or
            incorrect padding, it returns <code class="literal">NULL</code>.
            However, it is possible for <code class="literal">AES_DECRYPT()</code>
            to return a non-<code class="literal">NULL</code> value (possibly
            garbage) if the input data or the key is invalid.
          </p><p>
            You can use the AES functions to store data in an encrypted
            form by modifying your queries:
          </p><pre class="programlisting">INSERT INTO t VALUES (1,AES_ENCRYPT('text','password'));
</pre><p>
            <code class="literal">AES_ENCRYPT()</code> and
            <code class="literal">AES_DECRYPT()</code> can be considered the most
            cryptographically secure encryption functions currently
            available in MySQL.
          </p></li><li><p><a id="function_compress"></a>
            <a id="id2849302" class="indexterm"></a>

            <code class="literal">COMPRESS(<em class="replaceable"><code>string_to_compress</code></em>)</code>
          </p><p>
            Compresses a string and returns the result as a binary
            string. This function requires MySQL to have been compiled
            with a compression library such as <code class="literal">zlib</code>.
            Otherwise, the return value is always
            <code class="literal">NULL</code>. The compressed string can be
            uncompressed with <code class="literal">UNCOMPRESS()</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LENGTH(COMPRESS(REPEAT('a',1000)));</code></strong>
        -&gt; 21
mysql&gt; <strong class="userinput"><code>SELECT LENGTH(COMPRESS(''));</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT LENGTH(COMPRESS('a'));</code></strong>
        -&gt; 13
mysql&gt; <strong class="userinput"><code>SELECT LENGTH(COMPRESS(REPEAT('a',16)));</code></strong>
        -&gt; 15
</pre><p>
            The compressed string contents are stored the following way:
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                Empty strings are stored as empty strings.
              </p></li><li><p>
                Non-empty strings are stored as a four-byte length of
                the uncompressed string (low byte first), followed by
                the compressed string. If the string ends with space, an
                extra ‘<code class="literal">.</code>’ character is
                added to avoid problems with endspace trimming should
                the result be stored in a <code class="literal">CHAR</code> or
                <code class="literal">VARCHAR</code> column. (Use of
                <code class="literal">CHAR</code> or <code class="literal">VARCHAR</code> to
                store compressed strings is not recommended. It is
                better to use a <code class="literal">BLOB</code> column instead.)
              </p></li></ul></div></li><li><p><a id="function_decode"></a>
            <a id="id2849453" class="indexterm"></a>

            <code class="literal">DECODE(<em class="replaceable"><code>crypt_str</code></em>,<em class="replaceable"><code>pass_str</code></em>)</code>
          </p><p>
            Decrypts the encrypted string
            <em class="replaceable"><code>crypt_str</code></em> using
            <em class="replaceable"><code>pass_str</code></em> as the password.
            <em class="replaceable"><code>crypt_str</code></em> should be a string
            returned from <code class="literal">ENCODE()</code>.
          </p></li><li><p><a id="function_encode"></a>
            <a id="id2849524" class="indexterm"></a>

            <code class="literal">ENCODE(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>pass_str</code></em>)</code>
          </p><p>
            Encrypt <em class="replaceable"><code>str</code></em> using
            <em class="replaceable"><code>pass_str</code></em> as the password. To
            decrypt the result, use <code class="literal">DECODE()</code>.
          </p><p>
            The result is a binary string of the same length as
            <em class="replaceable"><code>str</code></em>.
          </p><p>
            The strength of the encryption is based on how good the
            random generator is. It should suffice for short strings.
          </p></li><li><p><a id="function_des-decrypt"></a>
            <a id="id2849605" class="indexterm"></a>

            <code class="literal">DES_DECRYPT(<em class="replaceable"><code>crypt_str</code></em>[,<em class="replaceable"><code>key_str</code></em>])</code>
          </p><p>
            Decrypts a string encrypted with
            <code class="literal">DES_ENCRYPT()</code>. If an error occurs, this
            function returns <code class="literal">NULL</code>.
          </p><p>
            Note that this function works only if MySQL has been
            configured with SSL support. See
            <a href="http://dev.mysql.com/doc/refman/5.0/en/secure-connections.html" target="_top">Using Secure Connections</a>.
          </p><p>
            If no <em class="replaceable"><code>key_str</code></em> argument is given,
            <code class="literal">DES_DECRYPT()</code> examines the first byte of
            the encrypted string to determine the DES key number that
            was used to encrypt the original string, and then reads the
            key from the DES key file to decrypt the message. For this
            to work, the user must have the <code class="literal">SUPER</code>
            privilege. The key file can be specified with the
            <code class="option">--des-key-file</code> server option.
          </p><p>
            If you pass this function a
            <em class="replaceable"><code>key_str</code></em> argument, that string is
            used as the key for decrypting the message.
          </p><p>
            If the <em class="replaceable"><code>crypt_str</code></em> argument does
            not appear to be an encrypted string, MySQL returns the
            given <em class="replaceable"><code>crypt_str</code></em>.
          </p></li><li><p><a id="function_des-encrypt"></a>
            <a id="id2849729" class="indexterm"></a>

            <code class="literal">DES_ENCRYPT(<em class="replaceable"><code>str</code></em>[,{<em class="replaceable"><code>key_num</code></em>|<em class="replaceable"><code>key_str</code></em>}])</code>
          </p><p>
            Encrypts the string with the given key using the Triple-DES
            algorithm.
          </p><p>
            Note that this function works only if MySQL has been
            configured with SSL support. See
            <a href="http://dev.mysql.com/doc/refman/5.0/en/secure-connections.html" target="_top">Using Secure Connections</a>.
          </p><p>
            The encryption key to use is chosen based on the second
            argument to <code class="literal">DES_ENCRYPT()</code>, if one was
            given:
          </p><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>Argument</strong></span></td><td><span class="bold"><strong>Description</strong></span></td></tr><tr><td>No argument</td><td>The first key from the DES key file is used.</td></tr><tr><td><em class="replaceable"><code>key_num</code></em></td><td>The given key number (0-9) from the DES key file is used.</td></tr><tr><td><em class="replaceable"><code>key_str</code></em></td><td>The given key string is used to encrypt <em class="replaceable"><code>str</code></em>.</td></tr></tbody></table></div><p>
            The key file can be specified with the
            <code class="option">--des-key-file</code> server option.
          </p><p>
            The return string is a binary string where the first
            character is <code class="literal">CHAR(128 |
            <em class="replaceable"><code>key_num</code></em>)</code>. If an error
            occurs, <code class="literal">DES_ENCRYPT()</code> returns
            <code class="literal">NULL</code>.
          </p><p>
            The 128 is added to make it easier to recognize an encrypted
            key. If you use a string key,
            <em class="replaceable"><code>key_num</code></em> is 127.
          </p><p>
            The string length for the result is given by this formula:
          </p><pre class="programlisting"><em class="replaceable"><code>new_len</code></em> = <em class="replaceable"><code>orig_len</code></em> + (8 - (<em class="replaceable"><code>orig_len</code></em> % 8)) + 1
</pre><p>
            Each line in the DES key file has the following format:
          </p><pre class="programlisting"><em class="replaceable"><code>key_num</code></em> <em class="replaceable"><code>des_key_str</code></em>
</pre><p>
            Each <em class="replaceable"><code>key_num</code></em> value must be a
            number in the range from <code class="literal">0</code> to
            <code class="literal">9</code>. Lines in the file may be in any order.
            <em class="replaceable"><code>des_key_str</code></em> is the string that is
            used to encrypt the message. There should be at least one
            space between the number and the key. The first key is the
            default key that is used if you do not specify any key
            argument to <code class="literal">DES_ENCRYPT()</code>.
          </p><p>
            You can tell MySQL to read new key values from the key file
            with the <code class="literal">FLUSH DES_KEY_FILE</code> statement.
            This requires the <code class="literal">RELOAD</code> privilege.
          </p><p>
            One benefit of having a set of default keys is that it gives
            applications a way to check for the existence of encrypted
            column values, without giving the end user the right to
            decrypt those values.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT customer_address FROM customer_table </code></strong>
     &gt; <strong class="userinput"><code>WHERE crypted_credit_card = DES_ENCRYPT('credit_card_number');</code></strong>
</pre></li><li><p><a id="function_encrypt"></a>
            <a id="id2850022" class="indexterm"></a>

            <code class="literal">ENCRYPT(<em class="replaceable"><code>str</code></em>[,<em class="replaceable"><code>salt</code></em>])</code>
          </p><p>
            Encrypts <em class="replaceable"><code>str</code></em> using the Unix
            <code class="literal">crypt()</code> system call and returns a binary
            string. The <em class="replaceable"><code>salt</code></em> argument should
            be a string with at least two characters. If no
            <em class="replaceable"><code>salt</code></em> argument is given, a random
            value is used.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT ENCRYPT('hello');</code></strong>
        -&gt; 'VxuFAJXVARROc'
</pre><p>
            <code class="literal">ENCRYPT()</code> ignores all but the first eight
            characters of <em class="replaceable"><code>str</code></em>, at least on
            some systems. This behavior is determined by the
            implementation of the underlying <code class="literal">crypt()</code>
            system call.
          </p><p>
            The use of <code class="literal">ENCYPT()</code> with multi-byte
            character sets other than <code class="literal">utf8</code> is not
            recommended because the system call expects a string
            terminated by a zero byte.
          </p><p>
            If <code class="literal">crypt()</code> is not available on your
            system (as is the case with Windows),
            <code class="literal">ENCRYPT()</code> always returns
            <code class="literal">NULL</code>.
          </p></li><li><p><a id="function_md5"></a>
            <a id="id2850156" class="indexterm"></a>

            <code class="literal">MD5(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            Calculates an MD5 128-bit checksum for the string. The value
            is returned as a binary string of 32 hex digits, or
            <code class="literal">NULL</code> if the argument was
            <code class="literal">NULL</code>. The return value can, for example,
            be used as a hash key.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT MD5('testing');</code></strong>
        -&gt; 'ae2b1fca515949e5d54fb22b8ed95575'
</pre><p>
            This is the “<span class="quote">RSA Data Security, Inc. MD5
            Message-Digest Algorithm.</span>”
          </p><p>
            If you want to convert the value to uppercase, see the
            description of binary string conversion given in the entry
            for the <code class="literal">BINARY</code> operator in
            <a href="mysqlqb_functions.html#cast-functions" title="Cast Functions and Operators">Cast Functions and Operators</a>.
          </p><p>
            See the note regarding the MD5 algorithm at the beginning
            this section.
          </p></li><li><p><a id="function_old-password"></a>
            <a id="id2850260" class="indexterm"></a>

            <code class="literal">OLD_PASSWORD(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            <code class="literal">OLD_PASSWORD()</code> was added to MySQL when
            the implementation of <code class="literal">PASSWORD()</code> was
            changed to improve security.
            <code class="literal">OLD_PASSWORD()</code> returns the value of the
            old (pre-4.1) implementation of
            <code class="literal">PASSWORD()</code> as a binary string, and is
            intended to permit you to reset passwords for any pre-4.1
            clients that need to connect to your version
            5.0 MySQL server without locking them out. See
            <a href="http://dev.mysql.com/doc/refman/5.0/en/password-hashing.html" target="_top">Password Hashing as of MySQL 4.1</a>.
          </p></li><li><p><a id="function_password"></a>
            <a id="id2850340" class="indexterm"></a>

            <code class="literal">PASSWORD(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            Calculates and returns a password string from the plaintext
            password <em class="replaceable"><code>str</code></em> and returns a binary
            string, or <code class="literal">NULL</code> if the argument was
            <code class="literal">NULL</code>. This is the function that is used
            for encrypting MySQL passwords for storage in the
            <code class="literal">Password</code> column of the
            <code class="literal">user</code> grant table.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT PASSWORD('badpwd');</code></strong>
        -&gt; '*AAB3E285149C0135D51A520E1940DD3263DC008C'
</pre><p>
            <a id="id2850411" class="indexterm"></a>

            <code class="literal">PASSWORD()</code> encryption is one-way (not
            reversible).
          </p><p>
            <code class="literal">PASSWORD()</code> does not perform password
            encryption in the same way that Unix passwords are
            encrypted. See <code class="literal">ENCRYPT()</code>.
          </p><p>
            <span class="bold"><strong>Note</strong></span>: The
            <code class="literal">PASSWORD()</code> function is used by the
            authentication system in MySQL Server; you should
            <span class="emphasis"><em>not</em></span> use it in your own applications.
            For that purpose, consider <code class="literal">MD5()</code> or
            <code class="literal">SHA1()</code> instead. Also see
            <a href="http://rfc.net/rfc2195.html" target="_top">RFC 2195, section 2
            (Challenge-Response Authentication Mechanism
            (CRAM))</a>, for more information about handling
            passwords and authentication securely in your applications.
          </p></li><li><p><a id="function_sha1"></a>
            <a id="id2850502" class="indexterm"></a>

            <a id="id2850508" class="indexterm"></a>

            <code class="literal">SHA1(<em class="replaceable"><code>str</code></em>)</code>,
            <code class="literal">SHA(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            Calculates an SHA-1 160-bit checksum for the string, as
            described in RFC 3174 (Secure Hash Algorithm). The value is
            returned as a binary string of 40 hex digits, or
            <code class="literal">NULL</code> if the argument was
            <code class="literal">NULL</code>. One of the possible uses for this
            function is as a hash key. You can also use it as a
            cryptographic function for storing passwords.
            <code class="literal">SHA()</code> is synonymous with
            <code class="literal">SHA1()</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SHA1('abc');</code></strong>
        -&gt; 'a9993e364706816aba3e25717850c26c9cd0d89d'
</pre><p>
            <code class="literal">SHA1()</code> can be considered a
            cryptographically more secure equivalent of
            <code class="literal">MD5()</code>. However, see the note regarding
            the MD5 and SHA-1 algorithms at the beginning this section.
          </p></li><li><p><a id="function_uncompress"></a>
            <a id="id2850616" class="indexterm"></a>

            <code class="literal">UNCOMPRESS(<em class="replaceable"><code>string_to_uncompress</code></em>)</code>
          </p><p>
            Uncompresses a string compressed by the
            <code class="literal">COMPRESS()</code> function. If the argument is
            not a compressed value, the result is
            <code class="literal">NULL</code>. This function requires MySQL to
            have been compiled with a compression library such as
            <code class="literal">zlib</code>. Otherwise, the return value is
            always <code class="literal">NULL</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UNCOMPRESS(COMPRESS('any string'));</code></strong>
        -&gt; 'any string'
mysql&gt; <strong class="userinput"><code>SELECT UNCOMPRESS('any string');</code></strong>
        -&gt; NULL
</pre></li><li><p><a id="function_uncompressed-length"></a>
            <a id="id2850706" class="indexterm"></a>

            <code class="literal">UNCOMPRESSED_LENGTH(<em class="replaceable"><code>compressed_string</code></em>)</code>
          </p><p>
            Returns the length that the compressed string had before
            being compressed.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UNCOMPRESSED_LENGTH(COMPRESS(REPEAT('a',30)));</code></strong>
        -&gt; 30
</pre></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="information-functions"></a>Information Functions</h3></div></div></div><a id="id2850765" class="indexterm"></a><a id="id2850772" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_benchmark"><code class="literal">BENCHMARK()</code></a></td><td>Repeatedly execute an expression</td></tr><tr><td><a href="mysqlqb_functions.html#function_charset"><code class="literal">CHARSET()</code></a></td><td>Return the character set of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_coercibility"><code class="literal">COERCIBILITY()</code></a></td><td>Return the collation coercibility value of the string  argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_collation"><code class="literal">COLLATION()</code></a></td><td>Return the collation of the string argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_connection-id"><code class="literal">CONNECTION_ID()</code></a></td><td>Return the connection ID (thread ID)   for the connection</td></tr><tr><td><a href="mysqlqb_functions.html#function_current-user"><code class="literal">CURRENT_USER()</code>, <code class="literal">CURRENT_USER</code></a></td><td>Return the username and hostname combination</td></tr><tr><td><a href="mysqlqb_functions.html#function_database"><code class="literal">DATABASE()</code></a></td><td>Return the default (current) database name</td></tr><tr><td><a href="mysqlqb_functions.html#function_found-rows"><code class="literal">FOUND_ROWS()</code></a></td><td>For a SELECT with a LIMIT clause, the  number of rows that would be returned were there no LIMIT clause</td></tr><tr><td><a href="mysqlqb_functions.html#function_last-insert-id"><code class="literal">LAST_INSERT_ID()</code></a></td><td>Value of the AUTOINCREMENT column   for the last INSERT</td></tr><tr><td><a href="mysqlqb_functions.html#function_row-count"><code class="literal">ROW_COUNT()</code></a></td><td>The number of rows  updated</td></tr><tr><td><a href="mysqlqb_functions.html#function_schema"><code class="literal">SCHEMA()</code></a></td><td>A synonym for DATABASE()</td></tr><tr><td><a href="mysqlqb_functions.html#function_session-user"><code class="literal">SESSION_USER()</code></a></td><td>Synonym for USER()</td></tr><tr><td><a href="mysqlqb_functions.html#function_system-user"><code class="literal">SYSTEM_USER()</code></a></td><td>Synonym for USER()</td></tr><tr><td><a href="mysqlqb_functions.html#function_user"><code class="literal">USER()</code></a></td><td>Return the current username and hostname</td></tr></tbody></table></div><div class="itemizedlist"><ul type="disc"><li><p><a id="function_benchmark"></a>
            <a id="id2851008" class="indexterm"></a>

            <code class="literal">BENCHMARK(<em class="replaceable"><code>count</code></em>,<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            The <code class="literal">BENCHMARK()</code> function executes the
            expression <em class="replaceable"><code>expr</code></em> repeatedly
            <em class="replaceable"><code>count</code></em> times. It may be used to
            time how quickly MySQL processes the expression. The result
            value is always <code class="literal">0</code>. The intended use is
            from within the <span><strong class="command">mysql</strong></span> client, which
            reports query execution times:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT BENCHMARK(1000000,ENCODE('hello','goodbye'));</code></strong>
+----------------------------------------------+
| BENCHMARK(1000000,ENCODE('hello','goodbye')) |
+----------------------------------------------+
|                                            0 |
+----------------------------------------------+
1 row in set (4.74 sec)
</pre><p>
            The time reported is elapsed time on the client end, not CPU
            time on the server end. It is advisable to execute
            <code class="literal">BENCHMARK()</code> several times, and to
            interpret the result with regard to how heavily loaded the
            server machine is.
          </p><p>
            <code class="literal">BENCHMARK()</code> is intended for measuring the
            runtime performance of scalar expressions, which has some
            significant implications for the way that you use it and
            interpret the results:
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                Only scalar expressions can be used. Although the
                expression can be a subquery, it must return a single
                column and at most a single row. For example,
                <code class="literal">BENCHMARK(10, (SELECT * FROM t))</code> will
                fail if the table <code class="literal">t</code> has more than one
                column or more than one row.
              </p></li><li><p>
                Executing a <code class="literal">SELECT
                <em class="replaceable"><code>expr</code></em></code> statement
                <em class="replaceable"><code>N</code></em> times differs from
                executing <code class="literal">SELECT
                BENCHMARK(<em class="replaceable"><code>N</code></em>,
                <em class="replaceable"><code>expr</code></em>)</code> in terms of
                the amount of overhead involved. The two have very
                different execution profiles and you should not expect
                them to take the same amount of time. The former
                involves the parser, optimizer, table locking, and
                runtime evaluation <em class="replaceable"><code>N</code></em> times
                each. The latter involves only runtime evaluation
                <em class="replaceable"><code>N</code></em> times, and all the other
                components just once. Memory structures already
                allocated are reused, and runtime optimizations such as
                local cacheing of results already evaluated for
                aggregate functions can alter the results. Use of
                <code class="literal">BENCHMARK()</code> thus measures performance
                of the runtime component by giving more weight to that
                component and removing the “<span class="quote">noise</span>”
                introduced by the network, parser, optimizer, and so
                forth.
              </p></li></ul></div></li><li><p><a id="function_charset"></a>
            <a id="id2814531" class="indexterm"></a>

            <code class="literal">CHARSET(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            Returns the character set of the string argument.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CHARSET('abc');</code></strong>
        -&gt; 'latin1'
mysql&gt; <strong class="userinput"><code>SELECT CHARSET(CONVERT('abc' USING utf8));</code></strong>
        -&gt; 'utf8'
mysql&gt; <strong class="userinput"><code>SELECT CHARSET(USER());</code></strong>
        -&gt; 'utf8'
</pre></li><li><p><a id="function_coercibility"></a>
            <a id="id2814604" class="indexterm"></a>

            <code class="literal">COERCIBILITY(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            Returns the collation coercibility value of the string
            argument.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT COERCIBILITY('abc' COLLATE latin1_swedish_ci);</code></strong>
        -&gt; 0
mysql&gt; <strong class="userinput"><code>SELECT COERCIBILITY(USER());</code></strong>
        -&gt; 3
mysql&gt; <strong class="userinput"><code>SELECT COERCIBILITY('abc');</code></strong>
        -&gt; 4
</pre><p>
            The return values have the meanings shown in the following
            table. Lower values have higher precedence.
          </p><div class="informaltable"><table border="1"><colgroup><col /><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>Coercibility</strong></span></td><td><span class="bold"><strong>Meaning</strong></span></td><td><span class="bold"><strong>Example</strong></span></td></tr><tr><td><code class="literal">0</code></td><td>Explicit collation</td><td>Value with <code class="literal">COLLATE</code> clause</td></tr><tr><td><code class="literal">1</code></td><td>No collation</td><td>Concatenation of strings with different collations</td></tr><tr><td><code class="literal">2</code></td><td>Implicit collation</td><td>Column value</td></tr><tr><td><code class="literal">3</code></td><td>System constant</td><td><code class="literal">USER()</code> return value</td></tr><tr><td><code class="literal">4</code></td><td>Coercible</td><td>Literal string</td></tr><tr><td><code class="literal">5</code></td><td>Ignorable</td><td><code class="literal">NULL</code> or an expression derived from
                    <code class="literal">NULL</code></td></tr></tbody></table></div><p>
            Before MySQL 5.0.3, the return values are shown as follows,
            and functions such as <code class="literal">USER()</code> have a
            coercibility of 2:
          </p><div class="informaltable"><table border="1"><colgroup><col /><col /><col /></colgroup><tbody><tr><td><span class="bold"><strong>Coercibility</strong></span></td><td><span class="bold"><strong>Meaning</strong></span></td><td><span class="bold"><strong>Example</strong></span></td></tr><tr><td><code class="literal">0</code></td><td>Explicit collation</td><td>Value with <code class="literal">COLLATE</code> clause</td></tr><tr><td><code class="literal">1</code></td><td>No collation</td><td>Concatenation of strings with different collations</td></tr><tr><td><code class="literal">2</code></td><td>Implicit collation</td><td>Column value, stored routine parameter or local variable</td></tr><tr><td><code class="literal">3</code></td><td>Coercible</td><td>Literal string</td></tr></tbody></table></div></li><li><p><a id="function_collation"></a>
            <a id="id2851346" class="indexterm"></a>

            <code class="literal">COLLATION(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            Returns the collation of the string argument.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT COLLATION('abc');</code></strong>
        -&gt; 'latin1_swedish_ci'
mysql&gt; <strong class="userinput"><code>SELECT COLLATION(_utf8'abc');</code></strong>
        -&gt; 'utf8_general_ci'
</pre></li><li><p><a id="function_connection-id"></a>
            <a id="id2851415" class="indexterm"></a>

            <code class="literal">CONNECTION_ID()</code>
          </p><p>
            Returns the connection ID (thread ID) for the connection.
            Every connection has an ID that is unique among the set of
            currently connected clients.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT CONNECTION_ID();</code></strong>
        -&gt; 23786
</pre></li><li><p><a id="function_current-user"></a>
            <a id="id2851476" class="indexterm"></a>

            <code class="literal">CURRENT_USER</code>,
            <code class="literal">CURRENT_USER()</code>
          </p><p>
            Returns the username and hostname combination for the MySQL
            account that the server used to authenticate the current
            client. This account determines your access privileges. As
            of MySQL 5.0.10, within a stored routine that is defined
            with the <code class="literal">SQL SECURITY DEFINER</code>
            characteristic, <code class="literal">CURRENT_USER()</code> returns
            the creator of the routine. The return value is a string in
            the <code class="literal">utf8</code> character set.
          </p><p>
            The value of <code class="literal">CURRENT_USER()</code> can differ
            from the value of <code class="literal">USER()</code>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT USER();</code></strong>
        -&gt; 'davida@localhost'
mysql&gt; <strong class="userinput"><code>SELECT * FROM mysql.user;</code></strong>
ERROR 1044: Access denied for user ''@'localhost' to
database 'mysql'
mysql&gt; <strong class="userinput"><code>SELECT CURRENT_USER();</code></strong>
        -&gt; '@localhost'
</pre><p>
            The example illustrates that although the client specified a
            username of <code class="literal">davida</code> (as indicated by the
            value of the <code class="literal">USER()</code> function), the server
            authenticated the client using an anonymous user account (as
            seen by the empty username part of the
            <code class="literal">CURRENT_USER()</code> value). One way this might
            occur is that there is no account listed in the grant tables
            for <code class="literal">davida</code>.
          </p></li><li><p><a id="function_database"></a>
            <a id="id2851603" class="indexterm"></a>

            <code class="literal">DATABASE()</code>
          </p><p>
            Returns the default (current) database name as a string in
            the <code class="literal">utf8</code> character set. If there is no
            default database, <code class="literal">DATABASE()</code> returns
            <code class="literal">NULL</code>. Within a stored routine, the
            default database is the database that the routine is
            associated with, which is not necessarily the same as the
            database that is the default in the calling context.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT DATABASE();</code></strong>
        -&gt; 'test'
</pre></li><li><p><a id="function_found-rows"></a>
            <a id="id2851678" class="indexterm"></a>

            <a id="id2851685" class="indexterm"></a>

            <code class="literal">FOUND_ROWS()</code>
          </p><p>
            A <code class="literal">SELECT</code> statement may include a
            <code class="literal">LIMIT</code> clause to restrict the number of
            rows the server returns to the client. In some cases, it is
            desirable to know how many rows the statement would have
            returned without the <code class="literal">LIMIT</code>, but without
            running the statement again. To obtain this row count,
            include a <code class="literal">SQL_CALC_FOUND_ROWS</code> option in
            the <code class="literal">SELECT</code> statement, and then invoke
            <code class="literal">FOUND_ROWS()</code> afterward:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SQL_CALC_FOUND_ROWS * FROM <em class="replaceable"><code>tbl_name</code></em></code></strong>
    -&gt; <strong class="userinput"><code>WHERE id &gt; 100 LIMIT 10;</code></strong>
mysql&gt; <strong class="userinput"><code>SELECT FOUND_ROWS();</code></strong>
</pre><p>
            The second <code class="literal">SELECT</code> returns a number
            indicating how many rows the first <code class="literal">SELECT</code>
            would have returned had it been written without the
            <code class="literal">LIMIT</code> clause.
          </p><p>
            In the absence of the <code class="literal">SQL_CALC_FOUND_ROWS</code>
            option in the most recent <code class="literal">SELECT</code>
            statement, <code class="literal">FOUND_ROWS()</code> returns the
            number of rows in the result set returned by that statement.
          </p><p>
            The row count available through
            <code class="literal">FOUND_ROWS()</code> is transient and not
            intended to be available past the statement following the
            <code class="literal">SELECT SQL_CALC_FOUND_ROWS</code> statement. If
            you need to refer to the value later, save it:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SQL_CALC_FOUND_ROWS * FROM ... ;</code></strong>
mysql&gt; <strong class="userinput"><code>SET @rows = FOUND_ROWS();</code></strong>
</pre><p>
            If you are using <code class="literal">SELECT
            SQL_CALC_FOUND_ROWS</code>, MySQL must calculate how many
            rows are in the full result set. However, this is faster
            than running the query again without
            <code class="literal">LIMIT</code>, because the result set need not be
            sent to the client.
          </p><p>
            <code class="literal">SQL_CALC_FOUND_ROWS</code> and
            <code class="literal">FOUND_ROWS()</code> can be useful in situations
            when you want to restrict the number of rows that a query
            returns, but also determine the number of rows in the full
            result set without running the query again. An example is a
            Web script that presents a paged display containing links to
            the pages that show other sections of a search result. Using
            <code class="literal">FOUND_ROWS()</code> allows you to determine how
            many other pages are needed for the rest of the result.
          </p><p>
            The use of <code class="literal">SQL_CALC_FOUND_ROWS</code> and
            <code class="literal">FOUND_ROWS()</code> is more complex for
            <code class="literal">UNION</code> statements than for simple
            <code class="literal">SELECT</code> statements, because
            <code class="literal">LIMIT</code> may occur at multiple places in a
            <code class="literal">UNION</code>. It may be applied to individual
            <code class="literal">SELECT</code> statements in the
            <code class="literal">UNION</code>, or global to the
            <code class="literal">UNION</code> result as a whole.
          </p><p>
            The intent of <code class="literal">SQL_CALC_FOUND_ROWS</code> for
            <code class="literal">UNION</code> is that it should return the row
            count that would be returned without a global
            <code class="literal">LIMIT</code>. The conditions for use of
            <code class="literal">SQL_CALC_FOUND_ROWS</code> with
            <code class="literal">UNION</code> are:
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                The <code class="literal">SQL_CALC_FOUND_ROWS</code> keyword must
                appear in the first <code class="literal">SELECT</code> of the
                <code class="literal">UNION</code>.
              </p></li><li><p>
                The value of <code class="literal">FOUND_ROWS()</code> is exact
                only if <code class="literal">UNION ALL</code> is used. If
                <code class="literal">UNION</code> without <code class="literal">ALL</code>
                is used, duplicate removal occurs and the value of
                <code class="literal">FOUND_ROWS()</code> is only approximate.
              </p></li><li><p>
                If no <code class="literal">LIMIT</code> is present in the
                <code class="literal">UNION</code>,
                <code class="literal">SQL_CALC_FOUND_ROWS</code> is ignored and
                returns the number of rows in the temporary table that
                is created to process the <code class="literal">UNION</code>.
              </p></li></ul></div></li><li><p><a id="function_last-insert-id"></a>
            <a id="id2852016" class="indexterm"></a>

            <code class="literal">LAST_INSERT_ID()</code>,
            <code class="literal">LAST_INSERT_ID(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            <code class="literal">LAST_INSERT_ID()</code> (with no argument)
            returns the <span class="emphasis"><em>first</em></span> automatically
            generated value that was set for an
            <code class="literal">AUTO_INCREMENT</code> column by the
            <span class="emphasis"><em>most recently executed</em></span>
            <code class="literal">INSERT</code> statement to affect such a column.
            For example, after inserting a row that generates an
            <code class="literal">AUTO_INCREMENT</code> value, you can get the
            value like this:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT LAST_INSERT_ID();</code></strong>
        -&gt; 195
</pre><p>
            The currently executing statement does not affect the value
            of <code class="literal">LAST_INSERT_ID()</code>. Suppose that you
            generate an <code class="literal">AUTO_INCREMENT</code> value with one
            statement, and then refer to
            <code class="literal">LAST_INSERT_ID()</code> in a multiple-row
            <code class="literal">INSERT</code> statement that inserts rows into a
            table with its own <code class="literal">AUTO_INCREMENT</code> column.
            The value of <code class="literal">LAST_INSERT_ID()</code> will remain
            stable in the second statement; its value for the second and
            later rows is not affected by the earlier row insertions.
            (However, if you mix references to
            <code class="literal">LAST_INSERT_ID()</code> and
            <code class="literal">LAST_INSERT_ID(<em class="replaceable"><code>expr</code></em>)</code>,
            the effect is undefined.)
          </p><p>
            If the previous statement returned an error, the value of
            <code class="literal">LAST_INSERT_ID()</code> is undefined. For
            transactional tables, if the statement is rolled back due to
            an error, the value of <code class="literal">LAST_INSERT_ID()</code>
            is left undefined. For manual <code class="literal">ROLLBACK</code>,
            the value of <code class="literal">LAST_INSERT_ID()</code> is not
            restored to that before the transaction; it remains as it
            was at the point of the <code class="literal">ROLLBACK</code>.
          </p><p>
            Within the body of a stored routine (procedure or function)
            or a trigger, the value of
            <code class="literal">LAST_INSERT_ID()</code> changes the same way as
            for statements executed outside the body of these kinds of
            objects. The effect of a stored routine or trigger upon the
            value of <code class="literal">LAST_INSERT_ID()</code> that is seen by
            following statements depends on the kind of routine:
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                If a stored procedure executes statements that change
                the value of <code class="literal">LAST_INSERT_ID()</code>, the
                changed value will be seen by statements that follow the
                procedure call.
              </p></li><li><p>
                For stored functions and triggers that change the value,
                the value is restored when the function or trigger ends,
                so following statements will not see a changed value.
              </p></li></ul></div><p>
            The ID that was generated is maintained in the server on a
            <span class="emphasis"><em>per-connection basis</em></span>. This means that
            the value returned by the function to a given client is the
            first <code class="literal">AUTO_INCREMENT</code> value generated for
            most recent statement affecting an
            <code class="literal">AUTO_INCREMENT</code> column <span class="emphasis"><em>by that
            client</em></span>. This value cannot be affected by other
            clients, even if they generate
            <code class="literal">AUTO_INCREMENT</code> values of their own. This
            behavior ensures that each client can retrieve its own ID
            without concern for the activity of other clients, and
            without the need for locks or transactions.
          </p><p>
            The value of <code class="literal">LAST_INSERT_ID()</code> is not
            changed if you set the <code class="literal">AUTO_INCREMENT</code>
            column of a row to a non-“<span class="quote">magic</span>” value (that
            is, a value that is not <code class="literal">NULL</code> and not
            <code class="literal">0</code>).
          </p><p>
            <span class="bold"><strong>Important</strong></span>: If you insert
            multiple rows using a single <code class="literal">INSERT</code>
            statement, <code class="literal">LAST_INSERT_ID()</code> returns the
            value generated for the <span class="emphasis"><em>first</em></span> inserted
            row <span class="emphasis"><em>only</em></span>. The reason for this is to
            make it possible to reproduce easily the same
            <code class="literal">INSERT</code> statement against some other
            server.
          </p><p>
            For example:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>USE test;</code></strong>
Database changed
mysql&gt; <strong class="userinput"><code>CREATE TABLE t (</code></strong>
    -&gt;   <strong class="userinput"><code>id INT AUTO_INCREMENT NOT NULL PRIMARY KEY,</code></strong>
    -&gt;   <strong class="userinput"><code>name VARCHAR(10) NOT NULL</code></strong>
    -&gt; <strong class="userinput"><code>);</code></strong>
Query OK, 0 rows affected (0.09 sec)

mysql&gt; <strong class="userinput"><code>INSERT INTO t VALUES (NULL, 'Bob');</code></strong>
Query OK, 1 row affected (0.01 sec)

mysql&gt; <strong class="userinput"><code>SELECT * FROM t;</code></strong>
+----+------+
| id | name |
+----+------+
|  1 | Bob  |
+----+------+
1 row in set (0.01 sec)

mysql&gt; <strong class="userinput"><code>SELECT LAST_INSERT_ID();</code></strong>
+------------------+
| LAST_INSERT_ID() |
+------------------+
|                1 |
+------------------+
1 row in set (0.00 sec)

mysql&gt; <strong class="userinput"><code>INSERT INTO t VALUES</code></strong>
    -&gt; <strong class="userinput"><code>(NULL, 'Mary'), (NULL, 'Jane'), (NULL, 'Lisa');</code></strong>
Query OK, 3 rows affected (0.00 sec)
Records: 3  Duplicates: 0  Warnings: 0

mysql&gt; SELECT * FROM t;
+----+------+
| id | name |
+----+------+
|  1 | Bob  |
|  2 | Mary |
|  3 | Jane |
|  4 | Lisa |
+----+------+
4 rows in set (0.01 sec)

mysql&gt; <strong class="userinput"><code>SELECT LAST_INSERT_ID();</code></strong>
+------------------+
| LAST_INSERT_ID() |
+------------------+
|                2 |
+------------------+
1 row in set (0.00 sec)
</pre><p>
            Although the second <code class="literal">INSERT</code> statement
            inserted three new rows into <code class="literal">t</code>, the ID
            generated for the first of these rows was
            <code class="literal">2</code>, and it is this value that is returned
            by <code class="literal">LAST_INSERT_ID()</code> for the following
            <code class="literal">SELECT</code> statement.
          </p><p>
            If you use <code class="literal">INSERT IGNORE</code> and the row is
            ignored, the <code class="literal">AUTO_INCREMENT</code> counter is
            not incremented and <code class="literal">LAST_INSERT_ID()</code>
            returns <code class="literal">0</code>, which reflects that no row was
            inserted.
          </p><p>
            <a id="id2852422" class="indexterm"></a>

            If <em class="replaceable"><code>expr</code></em> is given as an argument
            to <code class="literal">LAST_INSERT_ID()</code>, the value of the
            argument is returned by the function and is remembered as
            the next value to be returned by
            <code class="literal">LAST_INSERT_ID()</code>. This can be used to
            simulate sequences:
          </p><div class="orderedlist"><ol type="1"><li><p>
                Create a table to hold the sequence counter and
                initialize it:
              </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>CREATE TABLE sequence (id INT NOT NULL);</code></strong>
mysql&gt; <strong class="userinput"><code>INSERT INTO sequence VALUES (0);</code></strong>
</pre></li><li><p>
                Use the table to generate sequence numbers like this:
              </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>UPDATE sequence SET id=LAST_INSERT_ID(id+1);</code></strong>
mysql&gt; <strong class="userinput"><code>SELECT LAST_INSERT_ID();</code></strong>
</pre><p>
                The <code class="literal">UPDATE</code> statement increments the
                sequence counter and causes the next call to
                <code class="literal">LAST_INSERT_ID()</code> to return the
                updated value. The <code class="literal">SELECT</code> statement
                retrieves that value. The
                <code class="literal">mysql_insert_id()</code> C API function can
                also be used to get the value. See
                <a href="http://dev.mysql.com/doc/refman/5.0/en/mysql-insert-id.html" target="_top"><code class="literal">mysql_insert_id()</code></a>.
              </p></li></ol></div><p>
            You can generate sequences without calling
            <code class="literal">LAST_INSERT_ID()</code>, but the utility of
            using the function this way is that the ID value is
            maintained in the server as the last automatically generated
            value. It is multi-user safe because multiple clients can
            issue the <code class="literal">UPDATE</code> statement and get their
            own sequence value with the <code class="literal">SELECT</code>
            statement (or <code class="literal">mysql_insert_id()</code>), without
            affecting or being affected by other clients that generate
            their own sequence values.
          </p><p>
            Note that <code class="literal">mysql_insert_id()</code> is only
            updated after <code class="literal">INSERT</code> and
            <code class="literal">UPDATE</code> statements, so you cannot use the
            C API function to retrieve the value for
            <code class="literal">LAST_INSERT_ID(<em class="replaceable"><code>expr</code></em>)</code>
            after executing other SQL statements like
            <code class="literal">SELECT</code> or <code class="literal">SET</code>.
          </p></li><li><p><a id="function_row-count"></a>
            <a id="id2852599" class="indexterm"></a>

            <code class="literal">ROW_COUNT()</code>
          </p><p>
            <code class="literal">ROW_COUNT()</code> returns the number of rows
            updated, inserted, or deleted by the preceding statement.
            This is the same as the row count that the
            <span><strong class="command">mysql</strong></span> client displays and the value from
            the <code class="literal">mysql_affected_rows()</code> C API function.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>INSERT INTO t VALUES(1),(2),(3);</code></strong>
Query OK, 3 rows affected (0.00 sec)
Records: 3  Duplicates: 0  Warnings: 0

mysql&gt; <strong class="userinput"><code>SELECT ROW_COUNT();</code></strong>
+-------------+
| ROW_COUNT() |
+-------------+
|           3 |
+-------------+
1 row in set (0.00 sec)

mysql&gt; <strong class="userinput"><code>DELETE FROM t WHERE i IN(1,2);</code></strong>
Query OK, 2 rows affected (0.00 sec)

mysql&gt; <strong class="userinput"><code>SELECT ROW_COUNT();</code></strong>
+-------------+
| ROW_COUNT() |
+-------------+
|           2 |
+-------------+
1 row in set (0.00 sec)
</pre><p>
            <code class="literal">ROW_COUNT()</code> was added in MySQL 5.0.1.
          </p></li><li><p><a id="function_schema"></a>
            <a id="id2852702" class="indexterm"></a>

            <code class="literal">SCHEMA()</code>
          </p><p>
            This function is a synonym for
            <code class="literal">DATABASE()</code>. It was added in MySQL 5.0.2.
          </p></li><li><p><a id="function_session-user"></a>
            <a id="id2852755" class="indexterm"></a>

            <code class="literal">SESSION_USER()</code>
          </p><p>
            <code class="literal">SESSION_USER()</code> is a synonym for
            <code class="literal">USER()</code>.
          </p></li><li><p><a id="function_system-user"></a>
            <a id="id2852811" class="indexterm"></a>

            <code class="literal">SYSTEM_USER()</code>
          </p><p>
            <code class="literal">SYSTEM_USER()</code> is a synonym for
            <code class="literal">USER()</code>.
          </p></li><li><p><a id="function_user"></a>
            <a id="id2852866" class="indexterm"></a>

            <code class="literal">USER()</code>
          </p><p>
            Returns the current MySQL username and hostname as a string
            in the <code class="literal">utf8</code> character set.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT USER();</code></strong>
        -&gt; 'davida@localhost'
</pre><p>
            The value indicates the username you specified when
            connecting to the server, and the client host from which you
            connected. The value can be different from that of
            <code class="literal">CURRENT_USER()</code>.
          </p><p>
            You can extract only the username part like this:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT SUBSTRING_INDEX(USER(),'@',1);</code></strong>
        -&gt; 'davida'
</pre></li><li><p><a id="function_version"></a>
            <a id="id2852956" class="indexterm"></a>

            <code class="literal">VERSION()</code>
          </p><p>
            Returns a string that indicates the MySQL server version.
            The string uses the <code class="literal">utf8</code> character set.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT VERSION();</code></strong>
        -&gt; '5.0.38-standard'
</pre><p>
            Note that if your version string ends with
            <code class="literal">-log</code> this means that logging is enabled.
          </p></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="miscellaneous-functions"></a>Miscellaneous Functions</h3></div></div></div><a id="id2853024" class="indexterm"></a><a id="id2853031" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_count"><code class="literal">COUNT()</code></a></td><td>Return a count of the number of rows returned</td></tr><tr><td><a href="mysqlqb_functions.html#function_default"><code class="literal">DEFAULT()</code></a></td><td>Return the default value for a table column</td></tr><tr><td><a href="mysqlqb_functions.html#function_format"><code class="literal">FORMAT()</code></a></td><td>Return a number formatted to specified number of decimal places</td></tr><tr><td><a href="mysqlqb_functions.html#function_get-lock"><code class="literal">GET_LOCK()</code></a></td><td>Get a named lock</td></tr><tr><td><a href="mysqlqb_functions.html#function_inet-aton"><code class="literal">INET_ATON()</code></a></td><td>Return the numeric value of an IP address</td></tr><tr><td><a href="mysqlqb_functions.html#function_inet-ntoa"><code class="literal">INET_NTOA()</code></a></td><td>Return the IP address from a numeric value</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-free-lock"><code class="literal">IS_FREE_LOCK()</code></a></td><td>Checks whether the named lock  is free</td></tr><tr><td><a href="mysqlqb_functions.html#function_is-used-lock"><code class="literal">IS_USED_LOCK()</code></a></td><td>Checks whether the named lock is in use.  Return connection identifier if true.</td></tr><tr><td><a href="mysqlqb_functions.html#function_master-pos-wait"><code class="literal">MASTER_POS_WAIT()</code></a></td><td>Block until the slave has read and  applied all updates up to the specified position</td></tr><tr><td><a href="mysqlqb_functions.html#function_name-const"><code class="literal">NAME_CONST()</code></a>(v5.0.12)</td><td>Causes the column to have the given name</td></tr><tr><td><a href="mysqlqb_functions.html#function_release-lock"><code class="literal">RELEASE_LOCK()</code></a></td><td>Releases the named lock</td></tr><tr><td><a href="mysqlqb_functions.html#function_sleep"><code class="literal">SLEEP()</code></a></td><td>Sleep for a number of seconds</td></tr><tr><td><a href="mysqlqb_functions.html#function_uuid"><code class="literal">UUID()</code></a></td><td>Return a Universal Unique Identifier (UUID)</td></tr><tr><td><a href="mysqlqb_functions.html#function_values"><code class="literal">VALUES()</code></a></td><td>Defines the values to be used during an INSERT</td></tr></tbody></table></div><div class="itemizedlist"><ul type="disc"><li><p><a id="function_default"></a>
            <a id="id2853265" class="indexterm"></a>

            <code class="literal">DEFAULT(<em class="replaceable"><code>col_name</code></em>)</code>
          </p><p>
            Returns the default value for a table column. Starting with
            MySQL 5.0.2, an error results if the column has no default
            value.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>UPDATE t SET i = DEFAULT(i)+1 WHERE id &lt; 100;</code></strong>
</pre></li><li><p>
            <code class="literal">FORMAT(<em class="replaceable"><code>X</code></em>,<em class="replaceable"><code>D</code></em>)</code>
          </p><p>
            Formats the number <em class="replaceable"><code>X</code></em> to a format
            like <code class="literal">'#,###,###.##'</code>, rounded to
            <em class="replaceable"><code>D</code></em> decimal places, and returns the
            result as a string. For details, see
            <a href="mysqlqb_functions.html#string-functions" title="String Functions">String Functions</a>.
          </p></li><li><p><a id="function_get-lock"></a>
            <a id="id2853378" class="indexterm"></a>

            <a id="id2853385" class="indexterm"></a>

            <code class="literal">GET_LOCK(<em class="replaceable"><code>str</code></em>,<em class="replaceable"><code>timeout</code></em>)</code>
          </p><p>
            Tries to obtain a lock with a name given by the string
            <em class="replaceable"><code>str</code></em>, using a timeout of
            <em class="replaceable"><code>timeout</code></em> seconds. Returns
            <code class="literal">1</code> if the lock was obtained successfully,
            <code class="literal">0</code> if the attempt timed out (for example,
            because another client has previously locked the name), or
            <code class="literal">NULL</code> if an error occurred (such as
            running out of memory or the thread was killed with
            <span><strong class="command">mysqladmin kill</strong></span>). If you have a lock
            obtained with <code class="literal">GET_LOCK()</code>, it is released
            when you execute <code class="literal">RELEASE_LOCK()</code>, execute
            a new <code class="literal">GET_LOCK()</code>, or your connection
            terminates (either normally or abnormally). Locks obtained
            with <code class="literal">GET_LOCK()</code> do not interact with
            transactions. That is, committing a transaction does not
            release any such locks obtained during the transaction.
          </p><p>
            This function can be used to implement application locks or
            to simulate record locks. Names are locked on a server-wide
            basis. If a name has been locked by one client,
            <code class="literal">GET_LOCK()</code> blocks any request by another
            client for a lock with the same name. This allows clients
            that agree on a given lock name to use the name to perform
            cooperative advisory locking. But be aware that it also
            allows a client that is not among the set of cooperating
            clients to lock a name, either inadvertently or
            deliberately, and thus prevent any of the cooperating
            clients from locking that name. One way to reduce the
            likelihood of this is to use lock names that are
            database-specific or application-specific. For example, use
            lock names of the form
            <em class="replaceable"><code>db_name.str</code></em> or
            <em class="replaceable"><code>app_name.str</code></em>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT GET_LOCK('lock1',10);</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT IS_FREE_LOCK('lock2');</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT GET_LOCK('lock2',10);</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT RELEASE_LOCK('lock2');</code></strong>
        -&gt; 1
mysql&gt; <strong class="userinput"><code>SELECT RELEASE_LOCK('lock1');</code></strong>
        -&gt; NULL
</pre><p>
            The second <code class="literal">RELEASE_LOCK()</code> call returns
            <code class="literal">NULL</code> because the lock
            <code class="literal">'lock1'</code> was automatically released by the
            second <code class="literal">GET_LOCK()</code> call.
          </p><p>
            Note: If a client attempts to acquire a lock that is already
            held by another client, it blocks according to the
            <em class="replaceable"><code>timeout</code></em> argument. If the blocked
            client terminates, its thread does not die until the lock
            request times out. This is a known bug.
          </p></li><li><p><a id="function_inet-aton"></a>
            <a id="id2853580" class="indexterm"></a>

            <code class="literal">INET_ATON(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Given the dotted-quad representation of a network address as
            a string, returns an integer that represents the numeric
            value of the address. Addresses may be 4- or 8-byte
            addresses.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT INET_ATON('209.207.224.40');</code></strong>
        -&gt; 3520061480
</pre><p>
            The generated number is always in network byte order. For
            the example just shown, the number is calculated as
            209×256<sup>3</sup> +
            207×256<sup>2</sup> + 224×256 +
            40.
          </p><p>
            <code class="literal">INET_ATON()</code> also understands short-form
            IP addresses:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT INET_ATON('127.0.0.1'), INET_ATON('127.1');</code></strong>
        -&gt; 2130706433, 2130706433
</pre><p>
            <span class="bold"><strong>Note</strong></span>: When storing values
            generated by <code class="literal">INET_ATON()</code>, it is
            recommended that you use an <code class="literal">INT UNSIGNED</code>
            column. If you use a (signed) <code class="literal">INT</code> column,
            values corresponding to IP addresses for which the first
            octet is greater than 127 cannot be stored correctly. See
            <a href="http://dev.mysql.com/doc/refman/5.0/en/numeric-types.html" target="_top">Numeric Types</a>.
          </p></li><li><p><a id="function_inet-ntoa"></a>
            <a id="id2853714" class="indexterm"></a>

            <code class="literal">INET_NTOA(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Given a numeric network address (4 or 8 byte), returns the
            dotted-quad representation of the address as a string.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT INET_NTOA(3520061480);</code></strong>
        -&gt; '209.207.224.40'
</pre></li><li><p><a id="function_is-free-lock"></a>
            <a id="id2853781" class="indexterm"></a>

            <code class="literal">IS_FREE_LOCK(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            Checks whether the lock named <em class="replaceable"><code>str</code></em>
            is free to use (that is, not locked). Returns
            <code class="literal">1</code> if the lock is free (no one is using
            the lock), <code class="literal">0</code> if the lock is in use, and
            <code class="literal">NULL</code> if an error occurs (such as an
            incorrect argument).
          </p></li><li><p><a id="function_is-used-lock"></a>
            <a id="id2853851" class="indexterm"></a>

            <code class="literal">IS_USED_LOCK(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            Checks whether the lock named <em class="replaceable"><code>str</code></em>
            is in use (that is, locked). If so, it returns the
            connection identifier of the client that holds the lock.
            Otherwise, it returns <code class="literal">NULL</code>.
          </p></li><li><p><a id="function_master-pos-wait"></a>
            <a id="id2853912" class="indexterm"></a>

            <code class="literal">MASTER_POS_WAIT(<em class="replaceable"><code>log_name</code></em>,<em class="replaceable"><code>log_pos</code></em>[,<em class="replaceable"><code>timeout</code></em>])</code>
          </p><p>
            This function is useful for control of master/slave
            synchronization. It blocks until the slave has read and
            applied all updates up to the specified position in the
            master log. The return value is the number of log events the
            slave had to wait for to advance to the specified position.
            The function returns <code class="literal">NULL</code> if the slave
            SQL thread is not started, the slave's master information is
            not initialized, the arguments are incorrect, or an error
            occurs. It returns <code class="literal">-1</code> if the timeout has
            been exceeded. If the slave SQL thread stops while
            <code class="literal">MASTER_POS_WAIT()</code> is waiting, the
            function returns <code class="literal">NULL</code>. If the slave is
            past the specified position, the function returns
            immediately.
          </p><p>
            If a <em class="replaceable"><code>timeout</code></em> value is specified,
            <code class="literal">MASTER_POS_WAIT()</code> stops waiting when
            <em class="replaceable"><code>timeout</code></em> seconds have elapsed.
            <em class="replaceable"><code>timeout</code></em> must be greater than 0; a
            zero or negative <em class="replaceable"><code>timeout</code></em> means no
            timeout.
          </p></li><li><p><a id="function_name-const"></a>
            <a id="id2854023" class="indexterm"></a>

            <code class="literal">NAME_CONST(<em class="replaceable"><code>name</code></em>,<em class="replaceable"><code>value</code></em>)</code>
          </p><p>
            Returns the given value. When used to produce a result set
            column, <code class="literal">NAME_CONST()</code> causes the column to
            have the given name.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT NAME_CONST('myname', 14);</code></strong>
+--------+
| myname |
+--------+
|     14 |
+--------+
</pre><p>
            This function was added in MySQL 5.0.12. It is for internal
            use only. The server uses it when writing statements from
            stored routines that contain references to local routine
            variables, as described in
            <a href="http://dev.mysql.com/doc/refman/5.0/en/stored-procedure-logging.html" target="_top">Binary Logging of Stored Routines and Triggers</a>, You might see
            this function in the output from
            <span><strong class="command">mysqlbinlog</strong></span>.
          </p></li><li><p><a id="function_release-lock"></a>
            <a id="id2854113" class="indexterm"></a>

            <code class="literal">RELEASE_LOCK(<em class="replaceable"><code>str</code></em>)</code>
          </p><p>
            Releases the lock named by the string
            <em class="replaceable"><code>str</code></em> that was obtained with
            <code class="literal">GET_LOCK()</code>. Returns <code class="literal">1</code>
            if the lock was released, <code class="literal">0</code> if the lock
            was not established by this thread (in which case the lock
            is not released), and <code class="literal">NULL</code> if the named
            lock did not exist. The lock does not exist if it was never
            obtained by a call to <code class="literal">GET_LOCK()</code> or if it
            has previously been released.
          </p><p>
            The <code class="literal">DO</code> statement is convenient to use
            with <code class="literal">RELEASE_LOCK()</code>. See
            <a href="mysqlqb_statements.html#do" title="DO Syntax"><code class="literal">DO</code> Syntax</a>.
          </p></li><li><p><a id="function_sleep"></a>
            <a id="id2854209" class="indexterm"></a>

            <code class="literal">SLEEP(<em class="replaceable"><code>duration</code></em>)</code>
          </p><p>
            Sleeps (pauses) for the number of seconds given by the
            <em class="replaceable"><code>duration</code></em> argument, then returns
            0. If <code class="literal">SLEEP()</code> is interrupted, it returns
            1. The duration may have a fractional part given in
            microseconds. This function was added in MySQL 5.0.12.
          </p></li><li><p><a id="function_uuid"></a>
            <a id="id2854272" class="indexterm"></a>

            <code class="literal">UUID()</code>
          </p><p>
            Returns a Universal Unique Identifier (UUID) generated
            according to “<span class="quote">DCE 1.1: Remote Procedure Call</span>”
            (Appendix A) CAE (Common Applications Environment)
            Specifications published by The Open Group in October 1997
            (Document Number C706,
            <a href="http://www.opengroup.org/public/pubs/catalog/c706.htm" target="_top">http://www.opengroup.org/public/pubs/catalog/c706.htm</a>).
          </p><p>
            A UUID is designed as a number that is globally unique in
            space and time. Two calls to <code class="literal">UUID()</code> are
            expected to generate two different values, even if these
            calls are performed on two separate computers that are not
            connected to each other.
          </p><p>
            A UUID is a 128-bit number represented by a string of five
            hexadecimal numbers in
            <code class="literal">aaaaaaaa-bbbb-cccc-dddd-eeeeeeeeeeee</code>
            format:
          </p><div class="itemizedlist"><ul type="circle"><li><p>
                The first three numbers are generated from a timestamp.
              </p></li><li><p>
                The fourth number preserves temporal uniqueness in case
                the timestamp value loses monotonicity (for example, due
                to daylight saving time).
              </p></li><li><p>
                The fifth number is an IEEE 802 node number that
                provides spatial uniqueness. A random number is
                substituted if the latter is not available (for example,
                because the host computer has no Ethernet card, or we do
                not know how to find the hardware address of an
                interface on your operating system). In this case,
                spatial uniqueness cannot be guaranteed. Nevertheless, a
                collision should have <span class="emphasis"><em>very</em></span> low
                probability.
              </p><p>
                Currently, the MAC address of an interface is taken into
                account only on FreeBSD and Linux. On other operating
                systems, MySQL uses a randomly generated 48-bit number.
              </p></li></ul></div><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT UUID();</code></strong>
        -&gt; '6ccd780c-baba-1026-9564-0040f4311e29'
</pre><p>
            Note that <code class="literal">UUID()</code> does not yet work with
            replication.
          </p></li><li><p><a id="function_values"></a>
            <a id="id2854425" class="indexterm"></a>

            <code class="literal">VALUES(<em class="replaceable"><code>col_name</code></em>)</code>
          </p><p>
            In an <code class="literal">INSERT ... ON DUPLICATE KEY UPDATE</code>
            statement, you can use the
            <code class="literal">VALUES(<em class="replaceable"><code>col_name</code></em>)</code>
            function in the <code class="literal">UPDATE</code> clause to refer to
            column values from the <code class="literal">INSERT</code> portion of
            the statement. In other words,
            <code class="literal">VALUES(<em class="replaceable"><code>col_name</code></em>)</code>
            in the <code class="literal">UPDATE</code> clause refers to the value
            of <em class="replaceable"><code>col_name</code></em> that would be
            inserted, had no duplicate-key conflict occurred. This
            function is especially useful in multiple-row inserts. The
            <code class="literal">VALUES()</code> function is meaningful only in
            <code class="literal">INSERT ... ON DUPLICATE KEY UPDATE</code>
            statements and returns <code class="literal">NULL</code> otherwise.
            <a href="mysqlqb_statements.html#insert-on-duplicate" title="INSERT ... ON DUPLICATE KEY UPDATE Syntax"><code class="literal">INSERT ... ON DUPLICATE KEY UPDATE</code> Syntax</a>.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>INSERT INTO table (a,b,c) VALUES (1,2,3),(4,5,6)</code></strong>
    -&gt; <strong class="userinput"><code>ON DUPLICATE KEY UPDATE c=VALUES(a)+VALUES(b);</code></strong>
</pre></li></ul></div></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h2 class="title" style="clear: both"><a id="group-by-functions-and-modifiers"></a>Functions and Modifiers for Use with <code class="literal">GROUP BY</code> Clauses</h2></div></div></div><div class="toc"><dl><dt><span class="section"><a href="mysqlqb_functions.html#group-by-functions"><code class="literal">GROUP BY</code> (Aggregate) Functions</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#group-by-modifiers"><code class="literal">GROUP BY</code> Modifiers</a></span></dt><dt><span class="section"><a href="mysqlqb_functions.html#group-by-hidden-fields"><code class="literal">GROUP BY</code> and <code class="literal">HAVING</code> with Hidden
        Fields</a></span></dt></dl></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="group-by-functions"></a><code class="literal">GROUP BY</code> (Aggregate) Functions</h3></div></div></div><a id="id2854560" class="indexterm"></a><a id="id2854566" class="indexterm"></a><div class="informaltable"><table border="1"><colgroup><col /><col /></colgroup><thead><tr><th><span class="bold"><strong>Name</strong></span></th><th><span class="bold"><strong>Description</strong></span></th></tr></thead><tbody><tr><td><a href="mysqlqb_functions.html#function_avg"><code class="literal">AVG()</code></a></td><td>Return the average value of the argument</td></tr><tr><td><a href="mysqlqb_functions.html#function_count-distinct"><code class="literal">COUNT(DISTINCT)</code></a></td><td>Return the count of a number of different values</td></tr><tr><td><a href="mysqlqb_functions.html#function_group-concat"><code class="literal">GROUP_CONCAT()</code></a></td><td>Return a concatenated string</td></tr><tr><td><a href="mysqlqb_functions.html#function_min"><code class="literal">MAX()</code></a></td><td>Return the maximum value</td></tr><tr><td><a href="mysqlqb_functions.html#function_min"><code class="literal">MIN()</code></a></td><td>Return the minimum value</td></tr><tr><td><a href="mysqlqb_functions.html#function_std"><code class="literal">STD()</code>, <code class="literal">STDDEV()</code></a></td><td>Return the population standard deviation</td></tr><tr><td><a href="mysqlqb_functions.html#function_stddev-pop"><code class="literal">STDDEV_POP()</code></a></td><td>Return the population standard deviation</td></tr><tr><td><a href="mysqlqb_functions.html#function_stddev-samp"><code class="literal">STDDEV_SAMP()</code></a></td><td>Return the sample standard deviation</td></tr><tr><td><a href="mysqlqb_functions.html#function_sum"><code class="literal">SUM()</code></a></td><td>Return the sum</td></tr><tr><td><a href="mysqlqb_functions.html#function_var-pop"><code class="literal">VAR_POP()</code></a></td><td>Return the population standard variance</td></tr><tr><td><a href="mysqlqb_functions.html#function_var-samp"><code class="literal">VAR_SAMP()</code></a></td><td>Return the sample variance</td></tr><tr><td><a href="mysqlqb_functions.html#function_variance"><code class="literal">VARIANCE()</code></a></td><td>Return the population standard variance</td></tr></tbody></table></div><p>
        This section describes group (aggregate) functions that operate
        on sets of values. Unless otherwise stated, group functions
        ignore <code class="literal">NULL</code> values.
      </p><p>
        If you use a group function in a statement containing no
        <code class="literal">GROUP BY</code> clause, it is equivalent to grouping
        on all rows.
      </p><p>
        For numeric arguments, the variance and standard deviation
        functions return a <code class="literal">DOUBLE</code> value. The
        <code class="literal">SUM()</code> and <code class="literal">AVG()</code> functions
        return a <code class="literal">DECIMAL</code> value for exact-value
        arguments (integer or <code class="literal">DECIMAL</code>), and a
        <code class="literal">DOUBLE</code> value for approximate-value arguments
        (<code class="literal">FLOAT</code> or <code class="literal">DOUBLE</code>). (Before
        MySQL 5.0.3, <code class="literal">SUM()</code> and
        <code class="literal">AVG()</code> return <code class="literal">DOUBLE</code> for
        all numeric arguments.)
      </p><p>
        The <code class="literal">SUM()</code> and <code class="literal">AVG()</code>
        aggregate functions do not work with temporal values. (They
        convert the values to numbers, losing everything after the first
        non-numeric character.) To work around this problem, you can
        convert to numeric units, perform the aggregate operation, and
        convert back to a temporal value. Examples:
      </p><pre class="programlisting">SELECT SEC_TO_TIME(SUM(TIME_TO_SEC(<em class="replaceable"><code>time_col</code></em>))) FROM <em class="replaceable"><code>tbl_name</code></em>;
SELECT FROM_DAYS(SUM(TO_DAYS(<em class="replaceable"><code>date_col</code></em>))) FROM <em class="replaceable"><code>tbl_name</code></em>;
</pre><div class="itemizedlist"><ul type="disc"><li><p><a id="function_avg"></a>
            <a id="id2854888" class="indexterm"></a>

            <a id="id2854895" class="indexterm"></a>

            <a id="id2854902" class="indexterm"></a>

            <code class="literal">AVG([DISTINCT]
            <em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the average value of
            <code class="literal"><em class="replaceable"><code>expr</code></em></code>. The
            <code class="literal">DISTINCT</code> option can be used as of MySQL
            5.0.3 to return the average of the distinct values of
            <em class="replaceable"><code>expr</code></em>.
          </p><p>
            <code class="literal">AVG()</code> returns <code class="literal">NULL</code> if
            there were no matching rows.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT student_name, AVG(test_score)</code></strong>
    -&gt;        <strong class="userinput"><code>FROM student</code></strong>
    -&gt;        <strong class="userinput"><code>GROUP BY student_name;</code></strong>
</pre></li><li><p><a id="function_bit-and"></a>
            <a id="id2855003" class="indexterm"></a>

            <code class="literal">BIT_AND(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the bitwise <code class="literal">AND</code> of all bits in
            <em class="replaceable"><code>expr</code></em>. The calculation is
            performed with 64-bit (<code class="literal">BIGINT</code>) precision.
          </p><p>
            This function returns
            <code class="literal">18446744073709551615</code> if there were no
            matching rows. (This is the value of an unsigned
            <code class="literal">BIGINT</code> value with all bits set to 1.)
          </p></li><li><p><a id="function_bit-or"></a>
            <a id="id2855081" class="indexterm"></a>

            <code class="literal">BIT_OR(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the bitwise <code class="literal">OR</code> of all bits in
            <em class="replaceable"><code>expr</code></em>. The calculation is
            performed with 64-bit (<code class="literal">BIGINT</code>) precision.
          </p><p>
            This function returns <code class="literal">0</code> if there were no
            matching rows.
          </p></li><li><p><a id="function_bit-xor"></a>
            <a id="id2855152" class="indexterm"></a>

            <code class="literal">BIT_XOR(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the bitwise <code class="literal">XOR</code> of all bits in
            <em class="replaceable"><code>expr</code></em>. The calculation is
            performed with 64-bit (<code class="literal">BIGINT</code>) precision.
          </p><p>
            This function returns <code class="literal">0</code> if there were no
            matching rows.
          </p></li><li><p><a id="function_count"></a>
            <a id="id2855224" class="indexterm"></a>

            <code class="literal">COUNT(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns a count of the number of non-<code class="literal">NULL</code>
            values in the rows retrieved by a <code class="literal">SELECT</code>
            statement. The result is a <code class="literal">BIGINT</code> value.
          </p><p>
            <code class="literal">COUNT()</code> returns <code class="literal">0</code> if
            there were no matching rows.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT student.student_name,COUNT(*)</code></strong>
    -&gt;        <strong class="userinput"><code>FROM student,course</code></strong>
    -&gt;        <strong class="userinput"><code>WHERE student.student_id=course.student_id</code></strong>
    -&gt;        <strong class="userinput"><code>GROUP BY student_name;</code></strong>
</pre><p>
            <code class="literal">COUNT(*)</code> is somewhat different in that it
            returns a count of the number of rows retrieved, whether or
            not they contain <code class="literal">NULL</code> values.
          </p><p>
            <code class="literal">COUNT(*)</code> is optimized to return very
            quickly if the <code class="literal">SELECT</code> retrieves from one
            table, no other columns are retrieved, and there is no
            <code class="literal">WHERE</code> clause. For example:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT COUNT(*) FROM student;</code></strong>
</pre><p>
            This optimization applies only to <code class="literal">MyISAM</code>
            tables only, because an exact row count is stored for this
            storage engine and can be accessed very quickly. For
            transactional storage engines such as
            <code class="literal">InnoDB</code> and <code class="literal">BDB</code>,
            storing an exact row count is more problematic because
            multiple transactions may be occurring, each of which may
            affect the count.
          </p></li><li><p><a id="function_count-distinct"></a>
            <a id="id2855390" class="indexterm"></a>

            <a id="id2855396" class="indexterm"></a>

            <code class="literal">COUNT(DISTINCT
            <em class="replaceable"><code>expr</code></em>,[<em class="replaceable"><code>expr</code></em>...])</code>
          </p><p>
            Returns a count of the number of different
            non-<code class="literal">NULL</code> values.
          </p><p>
            <code class="literal">COUNT(DISTINCT)</code> returns
            <code class="literal">0</code> if there were no matching rows.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT COUNT(DISTINCT results) FROM student;</code></strong>
</pre><p>
            In MySQL, you can obtain the number of distinct expression
            combinations that do not contain <code class="literal">NULL</code> by
            giving a list of expressions. In standard SQL, you would
            have to do a concatenation of all expressions inside
            <code class="literal">COUNT(DISTINCT ...)</code>.
          </p></li><li><p><a id="function_group-concat"></a>
            <a id="id2855506" class="indexterm"></a>

            <code class="literal">GROUP_CONCAT(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            This function returns a string result with the concatenated
            non-<code class="literal">NULL</code> values from a group. It returns
            <code class="literal">NULL</code> if there are no
            non-<code class="literal">NULL</code> values. The full syntax is as
            follows:
          </p><pre class="programlisting">GROUP_CONCAT([DISTINCT] <em class="replaceable"><code>expr</code></em> [,<em class="replaceable"><code>expr</code></em> ...]
             [ORDER BY {<em class="replaceable"><code>unsigned_integer</code></em> | <em class="replaceable"><code>col_name</code></em> | <em class="replaceable"><code>expr</code></em>}
                 [ASC | DESC] [,<em class="replaceable"><code>col_name</code></em> ...]]
             [SEPARATOR <em class="replaceable"><code>str_val</code></em>])
</pre><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT student_name,</code></strong>
    -&gt;     <strong class="userinput"><code>GROUP_CONCAT(test_score)</code></strong>
    -&gt;     <strong class="userinput"><code>FROM student</code></strong>
    -&gt;     <strong class="userinput"><code>GROUP BY student_name;</code></strong>
</pre><p>
            Or:
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT student_name,</code></strong>
    -&gt;     <strong class="userinput"><code>GROUP_CONCAT(DISTINCT test_score</code></strong>
    -&gt;               <strong class="userinput"><code>ORDER BY test_score DESC SEPARATOR ' ')</code></strong>
    -&gt;     <strong class="userinput"><code>FROM student</code></strong>
    -&gt;     <strong class="userinput"><code>GROUP BY student_name;</code></strong>
</pre><p>
            In MySQL, you can get the concatenated values of expression
            combinations. You can eliminate duplicate values by using
            <code class="literal">DISTINCT</code>. If you want to sort values in
            the result, you should use <code class="literal">ORDER BY</code>
            clause. To sort in reverse order, add the
            <code class="literal">DESC</code> (descending) keyword to the name of
            the column you are sorting by in the <code class="literal">ORDER
            BY</code> clause. The default is ascending order; this
            may be specified explicitly using the <code class="literal">ASC</code>
            keyword. <code class="literal">SEPARATOR</code> is followed by the
            string value that should be inserted between values of
            result. The default is a comma
            (‘<code class="literal">,</code>’). You can eliminate the
            separator altogether by specifying <code class="literal">SEPARATOR
            ''</code>.
          </p><p>
            You can set a maximum allowed length with the
            <code class="literal">group_concat_max_len</code> system variable.
            (The default value is 1024.) The syntax to do this at
            runtime is as follows, where <em class="replaceable"><code>val</code></em>
            is an unsigned integer:
          </p><pre class="programlisting">SET [SESSION | GLOBAL] group_concat_max_len = <em class="replaceable"><code>val</code></em>;
</pre><p>
            If a maximum length has been set, the result is truncated to
            this maximum length.
          </p><p>
            Beginning with MySQL 5.0.19, the type returned by
            <code class="literal">GROUP_CONCAT()</code> is always
            <code class="literal">VARCHAR</code> unless
            <code class="literal">group_concat_max_len</code> is greater than 512,
            in which case, it returns a <code class="literal">BLOB</code>.
            (Previously, it returned a <code class="literal">BLOB</code> with
            <code class="literal">group_concat_max_len</code> greater than 512
            only if the query included an <code class="literal">ORDER BY</code>
            clause.)
          </p><p>
            See also <code class="literal">CONCAT()</code> and
            <code class="literal">CONCAT_WS()</code>:
            <a href="mysqlqb_functions.html#string-functions" title="String Functions">String Functions</a>.
          </p></li><li><p><a id="function_min"></a>
            <a id="id2855794" class="indexterm"></a>

            <a id="id2855801" class="indexterm"></a>

            <a id="id2855808" class="indexterm"></a>

            <a id="id2855818" class="indexterm"></a>

            <a id="id2855824" class="indexterm"></a>

            <a id="id2855831" class="indexterm"></a>

            <code class="literal">MIN([DISTINCT]
            <em class="replaceable"><code>expr</code></em>)</code>,
            <code class="literal">MAX([DISTINCT]
            <em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the minimum or maximum value of
            <em class="replaceable"><code>expr</code></em>. <code class="literal">MIN()</code>
            and <code class="literal">MAX()</code> may take a string argument; in
            such cases they return the minimum or maximum string value.
            See <a href="http://dev.mysql.com/doc/refman/5.0/en/mysql-indexes.html" target="_top">How MySQL Uses Indexes</a>. The
            <code class="literal">DISTINCT</code> keyword can be used to find the
            minimum or maximum of the distinct values of
            <em class="replaceable"><code>expr</code></em>, however, this produces the
            same result as omitting <code class="literal">DISTINCT</code>.
          </p><p>
            <code class="literal">MIN()</code> and <code class="literal">MAX()</code> return
            <code class="literal">NULL</code> if there were no matching rows.
          </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT student_name, MIN(test_score), MAX(test_score)</code></strong>
    -&gt;        <strong class="userinput"><code>FROM student</code></strong>
    -&gt;        <strong class="userinput"><code>GROUP BY student_name;</code></strong>
</pre><p>
            For <code class="literal">MIN()</code>, <code class="literal">MAX()</code>, and
            other aggregate functions, MySQL currently compares
            <code class="literal">ENUM</code> and <code class="literal">SET</code> columns
            by their string value rather than by the string's relative
            position in the set. This differs from how <code class="literal">ORDER
            BY</code> compares them. This is expected to be rectified
            in a future MySQL release.
          </p></li><li><p><a id="function_std"></a>
            <a id="id2855997" class="indexterm"></a>

            <a id="id2856004" class="indexterm"></a>

            <a id="id2856010" class="indexterm"></a>

            <a id="id2856017" class="indexterm"></a>

            <code class="literal">STD(<em class="replaceable"><code>expr</code></em>)</code>
            <code class="literal">STDDEV(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the population standard deviation of
            <em class="replaceable"><code>expr</code></em>. This is an extension to
            standard SQL. The <code class="literal">STDDEV()</code> form of this
            function is provided for compatibility with Oracle. As of
            MySQL 5.0.3, the standard SQL function
            <code class="literal">STDDEV_POP()</code> can be used instead.
          </p><p>
            These functions return <code class="literal">NULL</code> if there were
            no matching rows.
          </p></li><li><p><a id="function_stddev-pop"></a>
            <a id="id2856103" class="indexterm"></a>

            <code class="literal">STDDEV_POP(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the population standard deviation of
            <em class="replaceable"><code>expr</code></em> (the square root of
            <code class="literal">VAR_POP()</code>). This function was added in
            MySQL 5.0.3. Before 5.0.3, you can use
            <code class="literal">STD()</code> or <code class="literal">STDDEV()</code>,
            which are equivalent but not standard SQL.
          </p><p>
            <code class="literal">STDDEV_POP()</code> returns
            <code class="literal">NULL</code> if there were no matching rows.
          </p></li><li><p><a id="function_stddev-samp"></a>
            <a id="id2856185" class="indexterm"></a>

            <code class="literal">STDDEV_SAMP(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the sample standard deviation of
            <em class="replaceable"><code>expr</code></em> (the square root of
            <code class="literal">VAR_SAMP()</code>. This function was added in
            MySQL 5.0.3.
          </p><p>
            <code class="literal">STDDEV_SAMP()</code> returns
            <code class="literal">NULL</code> if there were no matching rows.
          </p></li><li><p><a id="function_sum"></a>
            <a id="id2856263" class="indexterm"></a>

            <a id="id2856270" class="indexterm"></a>

            <a id="id2856277" class="indexterm"></a>

            <code class="literal">SUM([DISTINCT]
            <em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the sum of <em class="replaceable"><code>expr</code></em>. If the
            return set has no rows, <code class="literal">SUM()</code> returns
            <code class="literal">NULL</code>. The <code class="literal">DISTINCT</code>
            keyword can be used in MySQL 5.0 to sum only
            the distinct values of <em class="replaceable"><code>expr</code></em>.
          </p><p>
            <code class="literal">SUM()</code> returns <code class="literal">NULL</code> if
            there were no matching rows.
          </p></li><li><p><a id="function_var-pop"></a>
            <a id="id2856364" class="indexterm"></a>

            <code class="literal">VAR_POP(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the population standard variance of
            <em class="replaceable"><code>expr</code></em>. It considers rows as the
            whole population, not as a sample, so it has the number of
            rows as the denominator. This function was added in MySQL
            5.0.3. Before 5.0.3, you can use
            <code class="literal">VARIANCE()</code>, which is equivalent but is
            not standard SQL.
          </p><p>
            <code class="literal">VAR_POP()</code> returns <code class="literal">NULL</code>
            if there were no matching rows.
          </p></li><li><p><a id="function_var-samp"></a>
            <a id="id2856440" class="indexterm"></a>

            <code class="literal">VAR_SAMP(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the sample variance of
            <em class="replaceable"><code>expr</code></em>. That is, the denominator is
            the number of rows minus one. This function was added in
            MySQL 5.0.3.
          </p><p>
            <code class="literal">VAR_SAMP()</code> returns
            <code class="literal">NULL</code> if there were no matching rows.
          </p></li><li><p><a id="function_variance"></a>
            <a id="id2856509" class="indexterm"></a>

            <code class="literal">VARIANCE(<em class="replaceable"><code>expr</code></em>)</code>
          </p><p>
            Returns the population standard variance of
            <em class="replaceable"><code>expr</code></em>. This is an extension to
            standard SQL. As of MySQL 5.0.3, the standard SQL function
            <code class="literal">VAR_POP()</code> can be used instead.
          </p><p>
            <code class="literal">VARIANCE()</code> returns
            <code class="literal">NULL</code> if there were no matching rows.
          </p></li></ul></div></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="group-by-modifiers"></a><code class="literal">GROUP BY</code> Modifiers</h3></div></div></div><a id="id2856577" class="indexterm"></a><a id="id2856584" class="indexterm"></a><p>
        The <code class="literal">GROUP BY</code> clause allows a <code class="literal">WITH
        ROLLUP</code> modifier that causes extra rows to be added to
        the summary output. These rows represent higher-level (or
        super-aggregate) summary operations. <code class="literal">ROLLUP</code>
        thus allows you to answer questions at multiple levels of
        analysis with a single query. It can be used, for example, to
        provide support for OLAP (Online Analytical Processing)
        operations.
      </p><p>
        Suppose that a table named <code class="literal">sales</code> has
        <code class="literal">year</code>, <code class="literal">country</code>,
        <code class="literal">product</code>, and <code class="literal">profit</code>
        columns for recording sales profitability:
      </p><pre class="programlisting">CREATE TABLE sales
(
    year    INT NOT NULL,
    country VARCHAR(20) NOT NULL,
    product VARCHAR(32) NOT NULL,
    profit  INT
);
</pre><p>
        The table's contents can be summarized per year with a simple
        <code class="literal">GROUP BY</code> like this:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT year, SUM(profit) FROM sales GROUP BY year;</code></strong>
+------+-------------+
| year | SUM(profit) |
+------+-------------+
| 2000 |        4525 |
| 2001 |        3010 |
+------+-------------+
</pre><p>
        This output shows the total profit for each year, but if you
        also want to determine the total profit summed over all years,
        you must add up the individual values yourself or run an
        additional query.
      </p><p>
        Or you can use <code class="literal">ROLLUP</code>, which provides both
        levels of analysis with a single query. Adding a <code class="literal">WITH
        ROLLUP</code> modifier to the <code class="literal">GROUP BY</code>
        clause causes the query to produce another row that shows the
        grand total over all year values:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT year, SUM(profit) FROM sales GROUP BY year WITH ROLLUP;</code></strong>
+------+-------------+
| year | SUM(profit) |
+------+-------------+
| 2000 |        4525 |
| 2001 |        3010 |
| NULL |        7535 |
+------+-------------+
</pre><p>
        The grand total super-aggregate line is identified by the value
        <code class="literal">NULL</code> in the <code class="literal">year</code> column.
      </p><p>
        <code class="literal">ROLLUP</code> has a more complex effect when there
        are multiple <code class="literal">GROUP BY</code> columns. In this case,
        each time there is a “<span class="quote">break</span>” (change in value) in
        any but the last grouping column, the query produces an extra
        super-aggregate summary row.
      </p><p>
        For example, without <code class="literal">ROLLUP</code>, a summary on the
        <code class="literal">sales</code> table based on <code class="literal">year</code>,
        <code class="literal">country</code>, and <code class="literal">product</code> might
        look like this:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT year, country, product, SUM(profit)</code></strong>
    -&gt; <strong class="userinput"><code>FROM sales</code></strong>
    -&gt; <strong class="userinput"><code>GROUP BY year, country, product;</code></strong>
+------+---------+------------+-------------+
| year | country | product    | SUM(profit) |
+------+---------+------------+-------------+
| 2000 | Finland | Computer   |        1500 |
| 2000 | Finland | Phone      |         100 |
| 2000 | India   | Calculator |         150 |
| 2000 | India   | Computer   |        1200 |
| 2000 | USA     | Calculator |          75 |
| 2000 | USA     | Computer   |        1500 |
| 2001 | Finland | Phone      |          10 |
| 2001 | USA     | Calculator |          50 |
| 2001 | USA     | Computer   |        2700 |
| 2001 | USA     | TV         |         250 |
+------+---------+------------+-------------+
</pre><p>
        The output indicates summary values only at the
        year/country/product level of analysis. When
        <code class="literal">ROLLUP</code> is added, the query produces several
        extra rows:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT year, country, product, SUM(profit)</code></strong>
    -&gt; <strong class="userinput"><code>FROM sales</code></strong>
    -&gt; <strong class="userinput"><code>GROUP BY year, country, product WITH ROLLUP;</code></strong>
+------+---------+------------+-------------+
| year | country | product    | SUM(profit) |
+------+---------+------------+-------------+
| 2000 | Finland | Computer   |        1500 |
| 2000 | Finland | Phone      |         100 |
| 2000 | Finland | NULL       |        1600 |
| 2000 | India   | Calculator |         150 |
| 2000 | India   | Computer   |        1200 |
| 2000 | India   | NULL       |        1350 |
| 2000 | USA     | Calculator |          75 |
| 2000 | USA     | Computer   |        1500 |
| 2000 | USA     | NULL       |        1575 |
| 2000 | NULL    | NULL       |        4525 |
| 2001 | Finland | Phone      |          10 |
| 2001 | Finland | NULL       |          10 |
| 2001 | USA     | Calculator |          50 |
| 2001 | USA     | Computer   |        2700 |
| 2001 | USA     | TV         |         250 |
| 2001 | USA     | NULL       |        3000 |
| 2001 | NULL    | NULL       |        3010 |
| NULL | NULL    | NULL       |        7535 |
+------+---------+------------+-------------+
</pre><p>
        For this query, adding <code class="literal">ROLLUP</code> causes the
        output to include summary information at four levels of
        analysis, not just one. Here's how to interpret the
        <code class="literal">ROLLUP</code> output:
      </p><div class="itemizedlist"><ul type="disc"><li><p>
            Following each set of product rows for a given year and
            country, an extra summary row is produced showing the total
            for all products. These rows have the
            <code class="literal">product</code> column set to
            <code class="literal">NULL</code>.
          </p></li><li><p>
            Following each set of rows for a given year, an extra
            summary row is produced showing the total for all countries
            and products. These rows have the <code class="literal">country</code>
            and <code class="literal">products</code> columns set to
            <code class="literal">NULL</code>.
          </p></li><li><p>
            Finally, following all other rows, an extra summary row is
            produced showing the grand total for all years, countries,
            and products. This row has the <code class="literal">year</code>,
            <code class="literal">country</code>, and <code class="literal">products</code>
            columns set to <code class="literal">NULL</code>.
          </p></li></ul></div><p>
        <span class="bold"><strong>Other Considerations When using
        <code class="literal">ROLLUP</code></strong></span>
      </p><p>
        The following items list some behaviors specific to the MySQL
        implementation of <code class="literal">ROLLUP</code>:
      </p><p>
        When you use <code class="literal">ROLLUP</code>, you cannot also use an
        <code class="literal">ORDER BY</code> clause to sort the results. In other
        words, <code class="literal">ROLLUP</code> and <code class="literal">ORDER BY</code>
        are mutually exclusive. However, you still have some control
        over sort order. <code class="literal">GROUP BY</code> in MySQL sorts
        results, and you can use explicit <code class="literal">ASC</code> and
        <code class="literal">DESC</code> keywords with columns named in the
        <code class="literal">GROUP BY</code> list to specify sort order for
        individual columns. (The higher-level summary rows added by
        <code class="literal">ROLLUP</code> still appear after the rows from which
        they are calculated, regardless of the sort order.)
      </p><p>
        <code class="literal">LIMIT</code> can be used to restrict the number of
        rows returned to the client. <code class="literal">LIMIT</code> is applied
        after <code class="literal">ROLLUP</code>, so the limit applies against
        the extra rows added by <code class="literal">ROLLUP</code>. For example:
      </p><pre class="programlisting">mysql&gt; <strong class="userinput"><code>SELECT year, country, product, SUM(profit)</code></strong>
    -&gt; <strong class="userinput"><code>FROM sales</code></strong>
    -&gt; <strong class="userinput"><code>GROUP BY year, country, product WITH ROLLUP</code></strong>
    -&gt; <strong class="userinput"><code>LIMIT 5;</code></strong>
+------+---------+------------+-------------+
| year | country | product    | SUM(profit) |
+------+---------+------------+-------------+
| 2000 | Finland | Computer   |        1500 |
| 2000 | Finland | Phone      |         100 |
| 2000 | Finland | NULL       |        1600 |
| 2000 | India   | Calculator |         150 |
| 2000 | India   | Computer   |        1200 |
+------+---------+------------+-------------+
</pre><p>
        Using <code class="literal">LIMIT</code> with <code class="literal">ROLLUP</code>
        may produce results that are more difficult to interpret,
        because you have less context for understanding the
        super-aggregate rows.
      </p><p>
        The <code class="literal">NULL</code> indicators in each super-aggregate
        row are produced when the row is sent to the client. The server
        looks at the columns named in the <code class="literal">GROUP BY</code>
        clause following the leftmost one that has changed value. For
        any column in the result set with a name that is a lexical match
        to any of those names, its value is set to
        <code class="literal">NULL</code>. (If you specify grouping columns by
        column number, the server identifies which columns to set to
        <code class="literal">NULL</code> by number.)
      </p><p>
        Because the <code class="literal">NULL</code> values in the
        super-aggregate rows are placed into the result set at such a
        late stage in query processing, you cannot test them as
        <code class="literal">NULL</code> values within the query itself. For
        example, you cannot add <code class="literal">HAVING product IS
        NULL</code> to the query to eliminate from the output all but
        the super-aggregate rows.
      </p><p>
        On the other hand, the <code class="literal">NULL</code> values do appear
        as <code class="literal">NULL</code> on the client side and can be tested
        as such using any MySQL client programming interface.
      </p></div><div class="section" lang="en" xml:lang="en"><div class="titlepage"><div><div><h3 class="title"><a id="group-by-hidden-fields"></a><code class="literal">GROUP BY</code> and <code class="literal">HAVING</code> with Hidden
        Fields</h3></div></div></div><a id="id2857261" class="indexterm"></a><a id="id2857272" class="indexterm"></a><a id="id2857283" class="indexterm"></a><p>
        MySQL extends the use of <code class="literal">GROUP BY</code> so that you
        can use non-aggregated columns or calculations in the
        <code class="literal">SELECT</code> list that do not appear in the
        <code class="literal">GROUP BY</code> clause. You can use this feature to
        get better performance by avoiding unnecessary column sorting
        and grouping. For example, you do not need to group on
        <code class="literal">customer.name</code> in the following query:
      </p><pre class="programlisting">SELECT order.custid, customer.name, MAX(payments)
  FROM order,customer
  WHERE order.custid = customer.custid
  GROUP BY order.custid;
</pre><p>
        In standard SQL, you would have to add
        <code class="literal">customer.name</code> to the <code class="literal">GROUP
        BY</code> clause. In MySQL, the name is redundant.
      </p><p>
        Do <span class="emphasis"><em>not</em></span> use this feature if the columns you
        omit from the <code class="literal">GROUP BY</code> part are not constant
        in the group. The server is free to return any value from the
        group, so the results are indeterminate unless all values are
        the same.
      </p><p>
        A similar MySQL extension applies to the
        <code class="literal">HAVING</code> clause. The SQL standard does not
        allow the <code class="literal">HAVING</code> clause to name any column
        that is not found in the <code class="literal">GROUP BY</code> clause if
        it is not enclosed in an aggregate function. MySQL allows the
        use of such columns to simplify calculations. This extension
        assumes that the non-grouped columns will have the same
        group-wise values. Otherwise, the result is indeterminate.
      </p><p>
        If the <code class="literal">ONLY_FULL_GROUP_BY</code> SQL mode is
        enabled, the MySQL extension to <code class="literal">GROUP BY</code> does
        not apply. That is, columns not named in the <code class="literal">GROUP
        BY</code> clause cannot be used in the
        <code class="literal">SELECT</code> list or <code class="literal">HAVING</code>
        clause if not used in an aggregate function.
      </p><p>
        The select list extension also applies to <code class="literal">ORDER
        BY</code>. That is, you can use non-aggregated columns or
        calculations in the <code class="literal">ORDER BY</code> clause that do
        not appear in the <code class="literal">GROUP BY</code> clause. This
        extension does not apply if the
        <code class="literal">ONLY_FULL_GROUP_BY</code> SQL mode is enabled.
      </p><p>
        In some cases, you can use <code class="literal">MIN()</code> and
        <code class="literal">MAX()</code> to obtain a specific column value even
        if it isn't unique. The following gives the value of
        <code class="literal">column</code> from the row containing the smallest
        value in the <code class="literal">sort</code> column:
      </p><pre class="programlisting">SUBSTR(MIN(CONCAT(RPAD(sort,6,' '),column)),7)
</pre><p>
        See <a href="http://dev.mysql.com/doc/refman/5.0/en/example-maximum-column-group-row.html" target="_top">The Rows Holding the Group-wise Maximum of a Certain Field</a>.
      </p><a id="id2857459" class="indexterm"></a><a id="id2857470" class="indexterm"></a><a id="id2857484" class="indexterm"></a><a id="id2857491" class="indexterm"></a><p>
        Note that if you are trying to follow standard SQL, you can't
        use expressions in <code class="literal">GROUP BY</code> clauses. You can
        work around this limitation by using an alias for the
        expression:
      </p><pre class="programlisting">SELECT id,FLOOR(value/100) AS val
  FROM <em class="replaceable"><code>tbl_name</code></em>
  GROUP BY id, val;
</pre><p>
        MySQL does allow expressions in <code class="literal">GROUP BY</code>
        clauses. For example:
      </p><pre class="programlisting">SELECT id,FLOOR(value/100)
  FROM <em class="replaceable"><code>tbl_name</code></em>
  GROUP BY id, FLOOR(value/100);
</pre></div></div></div><div class="navfooter"><hr /><table width="100%" summary="Navigation footer"><tr><td width="40%" align="left"><a accesskey="p" href="index.html">Prev</a> </td><td width="20%" align="center"> </td><td width="40%" align="right"> <a accesskey="n" href="mysqlqb_statements.html">Next</a></td></tr><tr><td width="40%" align="left" valign="top">Query Browser Help </td><td width="20%" align="center"><a accesskey="h" href="index.html">Home</a></td><td width="40%" align="right" valign="top"> SQL Statement Syntax</td></tr></table></div></body></html>