• 首页
  • vue
  • TypeScript
  • JavaScript
  • scss
  • css3
  • html5
  • php
  • MySQL
  • redis
  • jQuery
  • preg_match()

    (PHP 4, PHP 5, PHP 7)

    执行匹配正则表达式

    说明

    preg_match(string $pattern,string $subject[,array &$matches[,int $flags= 0[,int $offset= 0]]]): int

    搜索$subject$pattern给定的正则表达式的一个匹配.

    参数

    $pattern

    要搜索的模式,字符串类型。

    $subject

    输入字符串。

    $matches

    如果提供了参数$matches,它将被填充为搜索结果。$matches[0]将包含完整模式匹配到的文本,$matches[1]将包含第一个捕获子组匹配到的文本,以此类推。

    $flags

    $flags可以被设置为以下标记值:

    PREG_OFFSET_CAPTURE

    如果传递了这个标记,对于每一个出现的匹配返回时会附加字符串偏移量(相对于目标字符串的)。注意:这会改变填充到$matches参数的数组,使其每个元素成为一个由第0个元素是匹配到的字符串,第1个元素是该匹配字符串在目标字符串$subject中的偏移量。

    <?php
    preg_match('/(foo)(bar)(baz)/', 'foobarbaz', $matches, PREG_OFFSET_CAPTURE);
    print_r($matches);
    ?>
    

    以上例程会输出:

    Array
    (
        [0] => Array
            (
                [0] => foobarbaz
                [1] => 0
            )
        [1] => Array
            (
                [0] => foo
                [1] => 0
            )
        [2] => Array
            (
                [0] => bar
                [1] => 3
            )
        [3] => Array
            (
                [0] => baz
                [1] => 6
            )
    )
    
    $offset

    通常,搜索从目标字符串的开始位置开始。可选参数$offset用于指定从目标字符串的某个位置开始搜索(单位是字节)。

    Note:

    使用$offset参数不同于向preg_match()传递按照位置通过substr($subject,$offset)截取目标字符串结果,因为$pattern可以包含断言比如^$或者(?<=x)。比较:<?php $subject = "abcdef"; $pattern = '/^def/'; preg_match($pattern, $subject, $matches, PREG_OFFSET_CAPTURE, 3); print_r($matches); ?>

    以上例程会输出:

    Array ( )

    当这个示例使用截取后传递时

    <?php $subject = "abcdef"; $pattern = '/^def/'; preg_match($pattern, substr($subject,3), $matches, PREG_OFFSET_CAPTURE); print_r($matches); ?>

    将会产生匹配

    Array ( [0] => Array ( [0] => def [1] => 0 ) )

    返回值

    preg_match()返回$pattern的匹配次数。它的值将是0次(不匹配)或1次,因为preg_match()在第一次匹配后将会停止搜索。preg_match_all()不同于此,它会一直搜索$subject直到到达结尾。如果发生错误preg_match()返回FALSE

    更新日志

    版本说明
    5.3.6如果$offset$subject的长度还要大则返回FALSE
    5.2.2命名子组可以接受(?<name>)(?'name')以及(?P<name>)语法。之前版本仅接受(?P<name>)语法。

    范例

    查找文本字符串"php"

    <?php
    //模式分隔符后的"i"标记这是一个大小写不敏感的搜索
    if (preg_match("/php/i", "PHP is the web scripting language of choice.")) {
        echo "A match was found.";
    } else {
        echo "A match was not found.";
    }
    ?>
    

    查找单词"word"

    <?php
    /* 模式中的\b标记一个单词边界,所以只有独立的单词"web"会被匹配,而不会匹配
     * 单词的部分内容比如"webbing" 或 "cobweb" */
    if (preg_match("/\bweb\b/i", "PHP is the web scripting language of choice.")) {
        echo "A match was found.";
    } else {
        echo "A match was not found.";
    }
    if (preg_match("/\bweb\b/i", "PHP is the website scripting language of choice.")) {
        echo "A match was found.";
    } else {
        echo "A match was not found.";
    }
    ?>
    

    获取URL中的域名

    <?php
    //从URL中获取主机名称
    preg_match('@^(?:http://)?([^/]+)@i',
        "http://www.php.net/index", $matches);
    $host = $matches[1];
    //获取主机名称的后面两部分
    preg_match('/[^.]+\.[^.]+$/', $host, $matches);
    echo "domain name is: {$matches[0]}\n";
    ?>
    

    以上例程会输出:

    domain name is: php.net
    

    使用命名子组

    <?php
    $str = 'foobar: 2008';
    preg_match('/(?P<name>\w+): (?P<digit>\d+)/', $str, $matches);
    /* 下面例子在php 5.2.2(pcre 7.0)或更新版本下工作, 然而, 为了后向兼容, 上面的方式是推荐写法. */
    // preg_match('/(?<name>\w+): (?<digit>\d+)/', $str, $matches);
    print_r($matches);
    ?>
    

    以上例程会输出:

    Array
    (
        [0] => foobar: 2008
        [name] => foobar
        [1] => foobar
        [digit] => 2008
        [2] => 2008
    )
    

    注释

    Tip

    如果你仅仅想要检查某个字符串是否包含另外一个字符串,不要使用preg_match()。使用strpos()会更快。

    参见

    Simple regex
    Regex quick reference
    [abc]   A single character: a, b or c
    [^abc]   Any single character but a, b, or c
    [a-z]   Any single character in the range a-z
    [a-zA-Z]   Any single character in the range a-z or A-Z
    ^   Start of line
    $   End of line
    \A   Start of string
    \z   End of string
    .   Any single character
    \s   Any whitespace character
    \S   Any non-whitespace character
    \d   Any digit
    \D   Any non-digit
    \w   Any word character (letter, number, underscore)
    \W   Any non-word character
    \b   Any word boundary character
    (...)   Capture everything enclosed
    (a|b)   a or b
    a?   Zero or one of a
    a*   Zero or more of a
    a+   One or more of a
    a{3}   Exactly 3 of a
    a{3,}   3 or more of a
    a{3,6}   Between 3 and 6 of a
    options: i case insensitive m make dot match newlines x ignore whitespace in regex o perform #{...} substitutions only once
    Was working on a site that needed japanese and alphabetic letters and needed to 
    validate input using preg_match, I tried using \p{script} but didn't work:
    <?php
    $pattern ='/^([-a-zA-Z0-9_\p{Katakana}\p{Hiragana}\p{Han}]*)$/u'; // Didn't work
    ?>
    So I tried with ranges and it worked:
    <?php
    $pattern ='/^[-a-zA-Z0-9_\x{30A0}-\x{30FF}'
         .'\x{3040}-\x{309F}\x{4E00}-\x{9FBF}\s]*$/u';
    $match_string = '印刷最安 ニキビ跡除去 ゲームボーイ';
    if (preg_match($pattern, $match_string)) {
      echo "Found - pattern $pattern";
    } else {
      echo "Not found - pattern $pattern";
    }
    ?>
    U+4E00–U+9FBF Kanji
    U+3040–U+309F Hiragana
    U+30A0–U+30FF Katakana
    Hope its useful, it took me several hours to figure it out.
    Attention! PREG_OFFSET_CAPTURE not UTF-8 aware when using u modifier
    and it's not a but, it's a feature:
    https://bugs.php.net/bug.php?id=37391
    Possible workaround: Use mb_strpos to get the correct offset, instead of the flag.
     UTF-8 support would be nice.
    Sometimes its useful to negate a string. The first method which comes to mind to do this is: [^(string)] but this of course won't work. There is a solution, but it is not very well known. This is the simple piece of code on how a negation of a string is done:
    (?:(?!string).)
    ?: makes a subpattern (see http://www.php.net/manual/en/regexp.reference.subpatterns.php) and ?! is a negative look ahead. You put the negative look ahead in front of the dot because you want the regex engine to first check if there is an occurrence of the string you are negating. Only if it is not there, you want to match an arbitrary character.
    Hope this helps some ppl.
    I noticed that in order to deal with UTF-8 texts, without having to recompile php with the PCRE UTF-8 flag enabled, you can just add the following sequence at the start of your pattern: (*UTF8)
    for instance : '#(*UTF8)[[:alnum:]]#' will return TRUE for 'é' where '#[[:alnum:]]#' will return FALSE
    found this very very useful tip after hours of research over the web directly in pcre website right here : http://www.pcre.org/pcre.txt
    there are many further informations about UTF-8 support in the lib
    hop that will help!
    --
    cedric
    For those who search for a unicode regular expression example using preg_match here it is:
    Check for Persian digits
    preg_match( "/[^\x{06F0}-\x{06F9}\x]+/u" , '۱۲۳۴۵۶۷۸۹۰' );
    This sample regexp may be useful if you are working with DB field types. 
    (?P<type>\w+)($|\((?P<length>(\d+|(.*)))\))
    For example, if you are have a such type as "varchar(255)" or "text", the next fragment
    <?php
      $type = 'varchar(255)'; // type of field
      preg_match('/(?P<type>\w+)($|\((?P<length>(\d+|(.*)))\))/', $type, $field);
      print_r($field);
    ?>
    will output something like this:
    Array ( [0] => varchar(255) [type] => varchar [1] => varchar [2] => (255) [length] => 255 [3] => 255 [4] => 255 )
    When trying to check a file path that could be windows or unix it took me quite a few tries to get the escape characters right.
    The Unix directory separator must be escaped once and the windows directory separator must be escaped twice.
    This will match path/to/file and path\to\file.exe
    preg_match('/^[a-z0-9_.\/\\\]*$/i', $file_string);
    This sample is for checking persian character:
    <?php
      preg_match("/[\x{0600}-\x{06FF}\x]{1,32}/u", 'محمد');
    ?>
    
    There does not seem to be any mention of the PHP version of switches that can be used with regular expressions.
    preg_match_all('/regular expr/sim',$text).
    The s i m being the location for and available switches (I know about)
    The i is to ignore letter cases (this is commonly known - I think)
    The s tells the code NOT TO stop searching when it encounters \n (line break) - this is important with multi-line entries for example text from an editor that needs search.
    The m tells the code it is a multi-line entry, but importantly allows the use of ^ and $ to work when showing start and end.
    I am hoping this will save someone from the 4 hours of torture that I endured, trying to workout this issue.
    I just learned about named groups from a Python friend today and was curious if PHP supported them, guess what -- it does!!!
    http://www.regular-expressions.info/named.html
    <?php
      preg_match("/(?P<foo>abc)(.*)(?P<bar>xyz)/",
                'abcdefghijklmnopqrstuvwxyz',
                $matches);
      print_r($matches);
    ?>
    will produce: 
    Array
    (
      [0] => abcdefghijklmnopqrstuvwxyz
      [foo] => abc
      [1] => abc
      [2] => defghijklmnopqrstuvw
      [bar] => xyz
      [3] => xyz
    )
    Note that you actually get the named group as well as the numerical key
    value too, so if you do use them, and you're counting array elements, be
    aware that your array might be bigger than you initially expect it to be.
    To validate directorys on Windows i used this:
    if( preg_match("#^([a-z]{1}\:{1})?[\\\/]?([\-\w]+[\\\/]?)*$#i",$_GET['path'],$matches) !== 1 ){
      echo("Invalid value");
    }else{
      echo("Valid value");
    }
    The parts are:
    #^ and $i      Make the string matches at all the pattern, from start to end for ensure a complete match.
    ([a-z]{1}\:{1})?    The string may starts with one letter and a colon, but only 1 character for eachone, this is for the drive letter (C:)
    [\\\/]?      The string may contain, but not require 1 slash or backslash after the drive letter, (\/)
    ([\-\w]+[\\\/]?)*  The string must have 1 or more of any character like hyphen, letter, number, underscore, and may contain a slash or back slash at the end, to have a directory like ("/" or "folderName" or "folderName/"), this may be repeated one or more times.
    Bugs of preg_match (PHP-version 5.2.5)
    In most cases, the following example will show one of two PHP-bugs discovered with preg_match depending on your PHP-version and configuration.
    <?php
    $text = "test=";
    // creates a rather long text
    for ($i = 0; $i++ < 100000;)
      $text .= "%AB";
    // a typical URL_query validity-checker (the pattern's function does not matter for this example)
    $pattern  = '/^(?:[;\/?:@&=+$,]|(?:[^\W_]|[-_.!~*\()\[\] ])|(?:%[\da-fA-F]{2}))*$/';
      
    var_dump( preg_match( $pattern, $text ) );
    ?>
    Possible bug (1):
    =============
    On one of our Linux-Servers the above example crashes PHP-execution with a C(?) Segmentation Fault(!). This seems to be a known bug (see http://bugs.php.net/bug.php?id=40909), but I don't know if it has been fixed, yet.
    If you are looking for a work-around, the following code-snippet is what I found helpful. It wraps the possibly crashing preg_match call by decreasing the PCRE recursion limit in order to result in a Reg-Exp error instead of a PHP-crash.
    <?php
    [...]
    // decrease the PCRE recursion limit for the (possibly dangerous) preg_match call
    $former_recursion_limit = ini_set( "pcre.recursion_limit", 10000 );
    // the wrapped preg_match call
    $result = preg_match( $pattern, $text );
    // reset the PCRE recursion limit to its original value
    ini_set( "pcre.recursion_limit", $former_recursion_limit );
    // if the reg-exp fails due to the decreased recursion limit we may not make any statement, but PHP-execution continues
    if ( PREG_RECURSION_LIMIT_ERROR === preg_last_error() )
    {
      // react on the failed regular expression here
      $result = [...];
      
      // do logging or email-sending here
      [...]
    } //if
    ?>
    Possible bug (2):
    =============
    On one of our Windows-Servers the above example does not crash PHP, but (directly) hits the recursion-limit. Here, the problem is that preg_match does not return boolean(false) as expected by the description / manual of above.
    In short, preg_match seems to return an int(0) instead of the expected boolean(false) if the regular expression could not be executed due to the PCRE recursion-limit. So, if preg_match results in int(0) you seem to have to check preg_last_error() if maybe an error occurred.
    Be aware of bug https://bugs.php.net/bug.php?id=50887 when using sub patterns: Un-matched optional sub patterns at the end won't show up in $matches.
    Here is a workaround: Assign a name to all subpatterns you are interested in, and merge $match afterwards with an constant array containing some reasonable default values:
    <?php
    if (preg_match('/^(?P<lang>[^;*][^;]*){1}(?:;q=(?P<qval>[0-9.]+))?$/u', 'de', $match))
    {
     $match = array_merge(array('lang' => '', 'qval' => ''), $match);
     print_r($match);
    }
    ?>
    This outputs:
    Array
    (
      [lang] => de
      [qval] => 
      [0] => de
      [1] => de
    )
    Instead of:
    Array
    (
      [0] => de
      [lang] => de
      [1] => de
    )
    Matching a backslash character can be confusing, because double escaping is needed in the pattern: first for PHP, second for the regex engine
    <?php
    //match newline control character:
    preg_match('/\n/','\n');  //pattern matches and is stored as control character 0x0A in the pattern string
    preg_match('/\\\n/','\n'); //very same match, but is stored escaped as 0x5C,0x6E in the pattern string
    //trying to match "\'" (2 characters) in a text file, '\\\'' as PHP string:
    $subject = file_get_contents('myfile.txt');
    preg_match('/\\\'/',$subject);  //DOESN'T MATCH!!! stored as 0x5C,0x27 (escaped apostrophe), this only matches apostrophe
    preg_match('/\\\\\'/',$subject); //matches, stored as 0x5C,0x5C,0x27 (escaped backslash and unescaped apostrophe)
    preg_match('/\\\\\\\/',$subject); //also matches, stored as 0x5C,0x5C,0x5C,0x27 (escaped backslash and escaped apostrophe)
    //matching "\n" (2 characters):
    preg_match('/\\\\n/','\\n');
    preg_match('/\\\n/','\\n'); //same match - 3 backslashes are interpreted as 2 in PHP, if the following character is not escapeable
    ?>
    
    Because making a truly correct email validation function is harder than one may think, consider using this one which comes with PHP through the filter_var function (http://www.php.net/manual/en/function.filter-var.php):
    <?php
    $email = "someone@domain .local";
    if(!filter_var($email, FILTER_VALIDATE_EMAIL)) {
      echo "E-mail is not valid";
    } else {
      echo "E-mail is valid";
    }
    ?>
    
    If you want to validate an email in one line, use filter_var() function !
    http://fr.php.net/manual/en/function.filter-var.php
    easy use, as described in the document example :
    var_dump(filter_var('bob@example.com', FILTER_VALIDATE_EMAIL));
    Some times a Hacker use a php file or shell as a image to hack your website. so if you try to use move_uploaded_file() function as in example to allow for users to upload files, you must check if this file contains a bad codes or not so we use this function. preg match
    in this function we use
    unlink() - http://php.net/unlink
    after you upload file check a file with below function. 
    <?php
    /**
     * A simple function to check file from bad codes.
     *
     * @param (string) $file - file path.
     * @author Yousef Ismaeil - Cliprz[at]gmail[dot]com.
     */
    function is_clean_file ($file)
    {
      if (file_exists($file))
      {
        $contents = file_get_contents($file);
      }
      else
      {
        exit($file." Not exists.");
      }
      if (preg_match('/(base64_|eval|system|shell_|exec|php_)/i',$contents))
      {
        return true;
      }
      else if (preg_match("#&\#x([0-9a-f]+);#i", $contents))
      {
        return true;
      }
      elseif (preg_match('#&\#([0-9]+);#i', $contents))
      {
        return true;
      }
      elseif (preg_match("#([a-z]*)=([\`\'\"]*)script:#iU", $contents))
      {
        return true;
      }
      elseif (preg_match("#([a-z]*)=([\`\'\"]*)javascript:#iU", $contents))
      {
        return true;
      }
      elseif (preg_match("#([a-z]*)=([\'\"]*)vbscript:#iU", $contents))
      {
        return true;
      }
      elseif (preg_match("#(<[^>]+)style=([\`\'\"]*).*expression\([^>]*>#iU", $contents))
      {
        return true;
      }
      elseif (preg_match("#(<[^>]+)style=([\`\'\"]*).*behaviour\([^>]*>#iU", $contents))
      {
        return true;
      }
      elseif (preg_match("#</*(applet|link|style|script|iframe|frame|frameset|html|body|title|div|p|form)[^>]*>#i", $contents))
      {
        return true;
      }
      else
      {
        return false;
      }
    }
    ?>
    Use
    <?php
    // If image contains a bad codes
    $image  = "simpleimage.png";
    if (is_clean_file($image))
    {
      echo "Bad codes this is not image";
      unlink($image);
    }
    else
    {
      echo "This is a real image.";
    }
    ?>
    
    Simple function to return a sub-string following the preg convention. Kind of expensive, and some might say lazy but it has saved me time.
    # preg_substr($pattern,$subject,[$offset]) function
    # @author  aer0s
    # return a specific sub-string in a string using 
    #  a regular expression 
    # @param  $pattern  regular expression pattern to match
    # @param  $subject  string to search
    # @param  [$offset]  zero based match occurrence to return
    #               
    # [$offset] is 0 by default which returns the first occurrence,
    # if [$offset] is -1 it will return the last occurrence 
    function preg_substr($pattern,$subject,$offset=0){
      preg_match_all($pattern,$subject,$matches,PREG_PATTERN_ORDER);
      return $offset==-1?array_pop($matches[0]):$matches[0][$offset];
    } 
    example: 
           $pattern = "/model(\s|-)[a-z0-9]/i";
           $subject = "Is there something wrong with model 654, Model 732, and model 43xl or is Model aj45B the preferred choice?";
           echo preg_substr($pattern,$subject);
           echo preg_substr($pattern,$subject,1);
           echo preg_substr($pattern,$subject,-1); 
    Returns something like:
           model 654
           Model 732
           Model aj45B
    When using a 'bad words reject string' filter, preg_match is MUCH faster than strpos / stripos. Because in the other cases, you would need to do a foreach for each word. With efficient programming, the foreach is ONLY faster when the first word in the ban-list is found.
    (for 12 words, 100,000 iterations, no word found)
    stripos - Taken 1.4876 seconds.
    strpos - Taken 1.4207 seconds.
    preg_match - Taken 0.189 seconds.
    Interesting fact:
    With long words ('averylongwordtospitepreg'), the difference is only much less. Only about a 2/3rd of the time instead of 1/6th
    <?php
    $words = array('word1', 'word2', 'word3', 'word4', 'word5', 'word6', 'word7', 'word8', 'word9', 'word10', 'word11', 'word12' );
    $teststring = 'ThIs Is A tEsTsTrInG fOr TeStInG.';
    $count = 100000;
    $find = 0;
    $start = microtime(TRUE);
    for ($i = 0; $i < $count; $i++) {
      foreach ($words as $word) {
        if (stripos($teststring, $word) !== FALSE) {
          $find++;
          break;
        }
      }
    }
    echo 'stripos - Taken ' . round(microtime(TRUE) - $start, 4) . ' seconds.' . PHP_EOL;
    $start = microtime(TRUE);
    for ($i = 0; $i < $count; $i++) {
      foreach ($words as $word) {
        if (strpos($teststring, $word) !== FALSE) {
          $find++;
          break;
        }
      }
    }
    echo 'strpos - Taken ' . round(microtime(TRUE) - $start, 4) . ' seconds.' . PHP_EOL;
    $start = microtime(TRUE);
    $pattern = '/';
    $div = '';
    foreach ($words as $word) {
      $pattern .= $div . preg_quote($word);
      $div = '|';
    }
    $pattern .= '/i';
    //Pattern could easily be done somewhere else if words are static.
    for ($i = 0; $i < $count; $i++) {
      if (preg_match($pattern, $teststring)) {
        $find++;
      }
    }
    $end = microtime(TRUE);
    echo 'preg_match - Taken ' . round($end - $start, 4) . ' seconds.' . PHP_EOL;
    ?>
    
    As I wasted lots of time finding a REAL regex for URLs and resulted in building it on my own, I now have found one, that seems to work for all kinds of urls:
    <?php
      $regex = "((https?|ftp)\:\/\/)?"; // SCHEME
      $regex .= "([a-z0-9+!*(),;?&=\$_.-]+(\:[a-z0-9+!*(),;?&=\$_.-]+)?@)?"; // User and Pass
      $regex .= "([a-z0-9-.]*)\.([a-z]{2,3})"; // Host or IP
      $regex .= "(\:[0-9]{2,5})?"; // Port
      $regex .= "(\/([a-z0-9+\$_-]\.?)+)*\/?"; // Path
      $regex .= "(\?[a-z+&\$_.-][a-z0-9;:@&%=+\/\$_.-]*)?"; // GET Query
      $regex .= "(#[a-z_.-][a-z0-9+\$_.-]*)?"; // Anchor
    ?>
    Then, the correct way to check against the regex ist as follows:
    <?php
        if(preg_match("/^$regex$/", $url))
        {
            return true;
        }
    ?>
    
    I see a lot of people trying to put together phone regex's and struggling (hey, no worries...they're complicated). Here's one that we use that's pretty nifty. It's not perfect, but it should work for most non-idealists.
    *** Note: Only matches U.S. phone numbers. ***
    <?php
    // all on one line...
    $regex = '/^(?:1(?:[. -])?)?(?:\((?=\d{3}\)))?([2-9]\d{2})(?:(?<=\(\d{3})\))? ?(?:(?<=\d{3})[.-])?([2-9]\d{2})[. -]?(\d{4})(?: (?i:ext)\.? ?(\d{1,5}))?$/';
    // or broken up
    $regex = '/^(?:1(?:[. -])?)?(?:\((?=\d{3}\)))?([2-9]\d{2})'
        .'(?:(?<=\(\d{3})\))? ?(?:(?<=\d{3})[.-])?([2-9]\d{2})'
        .'[. -]?(\d{4})(?: (?i:ext)\.? ?(\d{1,5}))?$/';
    ?>
    If you're wondering why all the non-capturing subpatterns (which look like this "(?:", it's so that we can do this:
    <?php
    $formatted = preg_replace($regex, '($1) $2-$3 ext. $4', $phoneNumber);
    // or, provided you use the $matches argument in preg_match
    $formatted = "($matches[1]) $matches[2]-$matches[3]";
    if ($matches[4]) $formatted .= " $matches[4]";
    ?>
    *** Results: ***
    520-555-5542 :: MATCH
    520.555.5542 :: MATCH
    5205555542 :: MATCH
    520 555 5542 :: MATCH
    520) 555-5542 :: FAIL
    (520 555-5542 :: FAIL
    (520)555-5542 :: MATCH
    (520) 555-5542 :: MATCH
    (520) 555 5542 :: MATCH
    520-555.5542 :: MATCH
    520 555-0555 :: MATCH
    (520)5555542 :: MATCH
    520.555-4523 :: MATCH
    19991114444 :: FAIL
    19995554444 :: MATCH
    514 555 1231 :: MATCH
    1 555 555 5555 :: MATCH
    1.555.555.5555 :: MATCH
    1-555-555-5555 :: MATCH
    520-555-5542 ext.123 :: MATCH
    520.555.5542 EXT 123 :: MATCH
    5205555542 Ext. 7712 :: MATCH
    520 555 5542 ext 5 :: MATCH
    520) 555-5542 :: FAIL
    (520 555-5542 :: FAIL
    (520)555-5542 ext .4 :: FAIL
    (512) 555-1234 ext. 123 :: MATCH
    1(555)555-5555 :: MATCH
    Workaround for getting the offset in UTF-8
    (in some cases mb_strpos might be an option as well)
    <?php
    if(preg_match($pattern,$haystack,$out,PREG_OFFSET_CAPTURE)) {
      $offset = strlen(utf8_decode(substr($haystack,0,$out[0][1])));
    }
    ?>
    
    here is a small tool for someone learning to use regular expressions. it's very basic, and allows you to try different patterns and combinations. I made it to help me, because I like to try different things, to get a good understanding of how things work.
    <?php
    $search = isset($_POST['search'])?$_POST['search']:"//";
    $match = isset($_POST['match'])?$_POST['match']:"<>";
    echo '<form method="post">';
    echo 's: <input style="width:400px;" name="search" type="text" value="'.$search.'" /><br />';
    echo 'm:<input style="width:400px;" name="match" type="text" value="'.$match.'" /><input type="submit" value="go" /></form><br />';
    if (preg_match($search, $match)){echo "matches";}else{echo "no match";}
    ?>
    
    Here is a function that decreases the numbers inside a string (useful to convert DOM object into simplexml object)
    e.g.: decremente_chaine("somenode->anode[2]->achildnode[3]") will return "somenode->anode[1]->achildnode[2]"
    the numbering of the nodes in simplexml starts from zero, but from 1 in DOM xpath objects
    <?php
    function decremente_chaine($chaine)
      {
        //récupérer toutes les occurrences de nombres et leurs indices
        preg_match_all("/[0-9]+/",$chaine,$out,PREG_OFFSET_CAPTURE);
          //parcourir les occurrences 
          for($i=0;$i<sizeof($out[0]);$i++)
          {
            $longueurnombre = strlen((string)$out[0][$i][0]);
            $taillechaine = strlen($chaine);
            // découper la chaine en 3 morceaux
            $debut = substr($chaine,0,$out[0][$i][1]);
            $milieu = ($out[0][$i][0])-1;
            $fin = substr($chaine,$out[0][$i][1]+$longueurnombre,$taillechaine);
             // si c'est 10,100,1000 etc. on décale tout de 1 car le résultat comporte un chiffre de moins
             if(preg_match('#[1][0]+$#', $out[0][$i][0]))
             {
              for($j = $i+1;$j<sizeof($out[0]);$j++)
              {
                $out[0][$j][1] = $out[0][$j][1] -1;
              }
             }
            $chaine = $debut.$milieu.$fin;
          }
        return $chaine;
      }
    ?>
    
    Just an interesting note. Was just updating code to replace ereg() with strpos() and preg_match and the thought occured that preg_match() could be optimized to quit early when only searching if a string begins with something, for example
    <?php
    if(preg_match("/^http/", $url))
    {
     //do something
    }
    ?>
     vs 
    <?php 
    if(strpos($url, "http") === 0)
    {
    //do something
    }
    ?>
    As I guessed, strpos() is always faster (about 2x) for short strings like a URL but for very long strings of several paragraphs (e.g. a block of XML) when the string doesn't start with the needle preg_match as twice as fast as strpos() as it doesn't scan the entire string.
    So, if you are searching long strings and expect it to normally be true (e.g. validating XML), strpos() is a much faster BUT if you expect if to often fail, preg_match is the better choice.
    This is a function that uses regular expressions to match against the various VAT formats required across the EU.
    <?php
    /**
     * @param integer $country Country name
     * @param integer $vat_number VAT number to test e.g. GB123 4567 89
     * @return integer -1 if country not included OR 1 if the VAT Num matches for the country OR 0 if no match
    */
    function checkVatNumber( $country, $vat_number ) {
      switch($country) {
        case 'Austria':
          $regex = '/^(AT){0,1}U[0-9]{8}$/i';
          break;
        case 'Belgium':
          $regex = '/^(BE){0,1}[0]{0,1}[0-9]{9}$/i';
          break;
        case 'Bulgaria':
          $regex = '/^(BG){0,1}[0-9]{9,10}$/i';
          break;
        case 'Cyprus':
          $regex = '/^(CY){0,1}[0-9]{8}[A-Z]$/i';
          break;
        case 'Czech Republic':
          $regex = '/^(CZ){0,1}[0-9]{8,10}$/i';
          break;
        case 'Denmark':
          $regex = '/^(DK){0,1}([0-9]{2}[\ ]{0,1}){3}[0-9]{2}$/i';
          break;
        case 'Estonia':
        case 'Germany':
        case 'Greece':
        case 'Portugal':
          $regex = '/^(EE|EL|DE|PT){0,1}[0-9]{9}$/i';
          break;
        case 'France':
          $regex = '/^(FR){0,1}[0-9A-Z]{2}[\ ]{0,1}[0-9]{9}$/i';
          break;
        case 'Finland':
        case 'Hungary':
        case 'Luxembourg':
        case 'Malta':
        case 'Slovenia':
          $regex = '/^(FI|HU|LU|MT|SI){0,1}[0-9]{8}$/i';
          break;
        case 'Ireland':
          $regex = '/^(IE){0,1}[0-9][0-9A-Z\+\*][0-9]{5}[A-Z]$/i';
          break;
        case 'Italy':
        case 'Latvia':
          $regex = '/^(IT|LV){0,1}[0-9]{11}$/i';
          break;
        case 'Lithuania':
          $regex = '/^(LT){0,1}([0-9]{9}|[0-9]{12})$/i';
          break;
        case 'Netherlands':
          $regex = '/^(NL){0,1}[0-9]{9}B[0-9]{2}$/i';
          break;
        case 'Poland':
        case 'Slovakia':
          $regex = '/^(PL|SK){0,1}[0-9]{10}$/i';
          break;
        case 'Romania':
          $regex = '/^(RO){0,1}[0-9]{2,10}$/i';
          break;
        case 'Sweden':
          $regex = '/^(SE){0,1}[0-9]{12}$/i';
          break;
        case 'Spain':
          $regex = '/^(ES){0,1}([0-9A-Z][0-9]{7}[A-Z])|([A-Z][0-9]{7}[0-9A-Z])$/i';
          break;
        case 'United Kingdom':
          $regex = '/^(GB){0,1}([1-9][0-9]{2}[\ ]{0,1}[0-9]{4}[\ ]{0,1}[0-9]{2})|([1-9][0-9]{2}[\ ]{0,1}[0-9]{4}[\ ]{0,1}[0-9]{2}[\ ]{0,1}[0-9]{3})|((GD|HA)[0-9]{3})$/i';
          break;
        default:
          return -1;
          break;
      }
      
      return preg_match($regex, $vat_number);
    }
    ?>
    
    highlight Search Words 
    <?php
    function highlight($word, $subject) {
      
      $split_subject = explode(" ", $subject);
      $split_word = explode(" ", $word);
      foreach ($split_subject as $k => $v){
          foreach ($split_word as $k2 => $v2){
            if($v2 == $v){
              
              $split_subject[$k] = "<span class='highlight'>".$v."</span>";
            
            }
          }
       }
       
       return implode(' ', $split_subject);
    }
    ?>
    
    for those coming over from ereg, preg_match can be quite intimidating. to get started here is a migration tip.
    <?php
    if(ereg('[^0-9A-Za-z]',$test_string)) // will be true if characters arnt 0-9, A-Z or a-z.
    if(preg_match('/[^0-9A-Za-z]/',$test_string)) // this is the preg_match version. the /'s are now required.
    ?>
    
    I recently encountered a problem trying to capture multiple instances of named subpatterns from filenames.
    Therefore, I came up with this function.
    The function allows you to pass through flags (in this version it applies to all expressions tested), and generates an array of search results.
    Enjoy!
    <?php
    /**
     * Allows multiple expressions to be tested on one string.
     * This will return a boolean, however you may want to alter this.
     *
     * @author William Jaspers, IV <wjaspers4@gmail.com>
     * @created 2009-02-27 17:00:00 +6:00:00 GMT
     * @access public
     *
     * @param array $patterns An array of expressions to be tested.
     * @param String $subject The data to test.
     * @param array $findings Optional argument to store our results.
     * @param mixed $flags Pass-thru argument to allow normal flags to apply to all tested expressions.
     * @param array $errors A storage bin for errors
     *
     * @returns bool Whether or not errors occurred.
     */
    function preg_match_multiple( 
     array $patterns=array(), 
     $subject=null,
     &$findings=array(),
     $flags=false,
     &$errors=array()
    ) {
     foreach( $patterns as $name => $pattern )
     {
      if( 1 <= preg_match_all( $pattern, $subject, $found, $flags ) )
      {
       $findings[$name] = $found;
      } else 
      {
       if( PREG_NO_ERROR !== ( $code = preg_last_error() ))
       {
        $errors[$name] = $code;
       } else $findings[$name] = array();
      }
     }
     return (0===sizeof($errors));
    }
    ?>
    
    I have been working on a email system that will automatically generate a text email from a given HTML email by using strip_tags(). 
    The only issue I ran into, for my needs, were that the anchors would not keep their links. 
    I search for a little while and could not find anything to strip the links from the tags so I generated my own little snippet. 
    I am posting it here in hopes that others may find it useful and for later reference.
    A note to keep in mind:
    I was primarily concerned with valid HTML so if attributes do no use ' or " to contain the values then this will need to be tweaked.
    If you can edit this to work better, please let me know.
    <?php
    /**
     * Replaces anchor tags with text
     * - Will search string and replace all anchor tags with text (case insensitive)
     * 
     * How it works:
     * - Searches string for an anchor tag, checks to make sure it matches the criteria
     *     Anchor search criteria:
     *       - 1 - <a (must have the start of the anchor tag )
     *       - 2 - Can have any number of spaces or other attributes before and after the href attribute
     *       - 3 - Must close the anchor tag
     * 
     * - Once the check has passed it will then replace the anchor tag with the string replacement
     * - The string replacement can be customized
     * 
     * Know issue:
     * - This will not work for anchors that do not use a ' or " to contain the attributes. 
     *     (i.e.- <a href=http: //php.net>PHP.net</a> will not be replaced)
     */
    function replaceAnchorsWithText($data) {
      /**
       * Had to modify $regex so it could post to the site... so I broke it into 6 parts.
       */
      $regex = '/(<a\s*'; // Start of anchor tag
      $regex .= '(.*?)\s*'; // Any attributes or spaces that may or may not exist
      $regex .= 'href=[\'"]+?\s*(?P<link>\S+)\s*[\'"]+?'; // Grab the link
      $regex .= '\s*(.*?)\s*>\s*'; // Any attributes or spaces that may or may not exist before closing tag 
      $regex .= '(?P<name>\S+)'; // Grab the name
      $regex .= '\s*<\/a>)/i'; // Any number of spaces between the closing anchor tag (case insensitive)
      
      if (is_array($data)) {
        // This is what will replace the link (modify to you liking)
        $data = "{$data['name']}({$data['link']})";
      }
      return preg_replace_callback($regex, 'replaceAnchorsWithText', $data);
    }
    $input = 'Test 1: <a href="http: //php.net1">PHP.NET1</a>.<br />';
    $input .= 'Test 2: <A name="test" HREF=\'HTTP: //PHP.NET2\' target="_blank">PHP.NET2</A>.<BR />';
    $input .= 'Test 3: <a hRef=http: //php.net3>php.net3</a><br />';
    $input .= 'This last line had nothing to do with any of this';
    echo replaceAnchorsWithText($input).'<hr/>';
    ?>
    Will output:
    Test 1: PHP.NET1(http: //php.net1).
    Test 2: PHP.NET2(HTTP: //PHP.NET2).
    Test 3: php.net3 (is still an anchor)
    This last line had nothing to do with any of this
    Posting to this site is painful...
    Had to break up the regex and had to break the test links since it was being flagged as spam...
    When trying to match accented characters, such as those found in Spanish, there seems to be a different internal interpretation when using character classes. So the best way is to add the u option (for unicode) after the delimiters.
    <?php
    //echoes 1 (adding u would not alter the result)
    echo preg_match('/^áéíóúñ$/', 'áéíóúñ');
    //echoes 0 (unless with [ó]+ or [ó]* or adding u)
    echo preg_match('/^áéí[ó]úñ$/', 'áéíóúñ');
    //so to match 'espana' or 'españa', add u or this won't match
    //echoes 1
    echo preg_match('/^espa[nñ]a$/u', 'españa');
    ?>
    
    If you need to check for .com.br and .com.au and .uk and all the other crazy domain endings i found the following expression works well if you want to validate an email address. Its quite generous in what it will allow
    <?php
        $email_address = "phil.taylor@a_domain.tv";
      if (preg_match("/^[^@]*@[^@]*\.[^@]*$/", $email_address)) {
        return "E-mail address";    
      }
        
    ?>
    
    The regular expression for breaking-down a URI reference into its components:
       ^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))?
        12      3 4     5    6 7    8 9
    Source: ietf.org/rfc/rfc2396.txt
    The function will return false and raise a warning if the input $subject is too long :
    [PhpWarning] preg_match(): Subject is too long
    I believe the limit is 1 or 2 GB because I was using a 2.2GB string.
    While a parameter might exist to alter this limit, in my case it was possible and wiser to use <500MB strings instead.
    To extract scheme, host, path, ect. simply use 
    <?php
     $url = 'http://name:pass@';
     $url .= 'example.com:10000';
     $url .= '/path/to/file.php?a=1&amp;b=2#anchor';
     $url_data = parse_url ( $url );
     print_r ( $url_data );
    ?>
    ___
    prints out something like:
    Array
    (
      [scheme] => http
      [host] => wild.subdomain.orgy.domain.co.uk
      [port] => 10000
      [user] => name
      [pass] => pass
      [path] => /path/to/file.php
      [query] => a=1&b=2
      [fragment] => anchor
    )
    In my tests parse_url is up to 15x faster than preg_match(_all)!
    Preg_match returns empty result trying to validate $subject with carriege returns (/n/r).
    To solve it one need to use /s modifier in $pattern string.
    <?php
    $pattern='/.*/s';
    $valid=preg_match($pattern, $subject, $match);
    ?>
    
    If someone is from a country that accepts decimal numbers in format 9.00 and 9,00 (point or comma), number validation would be like that:
    <?php
    $number_check = "9,99";
    if (preg_match( '/^[\-+]?[0-9]*\.*\,?[0-9]+$/', $number_check)) {
      return TRUE; 
    }
    ?>
    However, if the number will be written in the database, most probably this comma needs to be replaced with a dot. 
    This can be done with use of str_replace, i.e :
    <?php
    $number_database = str_replace("," , "." , $number_check);
    ?>
    
    You can use the following code to detect non-latin (Cyrilic, Arabic, Greek...) characters:
    <?php
    preg_match("/^[a-zA-Z\p{Cyrillic}0-9\s\-]+$/u", "ABC abc 1234 АБВ абв");
    ?>
    
    Here's a regex to validate against the schema for common MySQL 
     identifiers:
    <?php
    $string = "$table_name";
    if (preg_match("/[^\\d\\sa-zA-Z$_]/", $string))
     echo "Failed validation";
    ?>
    
    I spent a while replacing all my ereg() calls to preg_match(), since ereg() is now deprecated and will not be supported as of v 6.0.
    Just a warning regarding the conversion, the two functions behave very similarly, but not exactly alike. Obviously, you will need to delimit your pattern with '/' or '|' characters.
    The difference that stumped me was that preg_replace overwrites the $matches array regardless if a match was found. If no match was found, $matches is simply empty.
    ereg(), however, would leave $matches alone if a match was not found. In my code, I had repeated calls to ereg, and was populating $matches with each match. I was only interested in the last match. However, with preg_match, if the very last call to the function did not result in a match, the $matches array would be overwritten with a blank value.
    Here is an example code snippet to illustrate:
    <?php
    $test = array('yes','no','yes','no','yes','no');
    foreach ($test as $key=>$value) {
     ereg("yes",$value,$matches1);
     preg_match("|yes|",$value,$matches2);
    }
     print "ereg result: $matches1[0]<br>";
     print "preg_match result: $matches2[0]<br>";
    ?>
    The output is:
    ereg result: yes
    preg_match result: 
    ($matches2[0] in this case is empty)
    I believe the preg_match behavior is cleaner. I just thought I would report this to hopefully save others some time.
    When using accented characters and "ñ" (áéíóúñ), preg_match does not work. It is a charset problem, use utf8_decode/decode to fix.
    To support large Unicode ranges (ie: [\x{E000}-\x{FFFD}] or \x{10FFFFF}) you must use the modifier '/u' at the end of your expression.
    When you use preg_match() for security purpose or huge data processing,
    mayby you should make consideration for backtrack_limit and recursion_limit.
    http://www.php.net/manual/en/pcre.configuration.php
    These limits may bring wrong matching result.
    You can verify whether you hit these limits by checking preg_last_error().
    http://www.php.net/manual/en/function.preg-last-error.php
    Note that if a parenthesed group is not matched, its key may or may not be present in $matches. For instance,
    <?php preg_match('/(foo)?(bar)?(baz)?/', 'bar', $matches);
    print_r($matches); 
    // outputs
    // Array
    // (
    //   [0] => bar
    //   [1] => 
    //   [2] => bar
    // )
    ?>
    Note that there is no element with key '3' in $matches, but an element with key '1' (the empty string). This inconsistent behavior also applies to named groups.
    Basic test for invalid UTF-8 that can hi-jack IE:
    <?php
    $valid = (preg_match('/^./us', $text) == 1);
    ?>
    See http://api.drupal.org/api/drupal/includes--bootstrap.inc/function/drupal_validate_utf8/7 for details.
    ---
    Test for valid UTF-8 and XML/XHTML character range compatibility:
    <?php
    $invalid = preg_match('@[^\x9\xA\xD\x20-\x{D7FF}\x{E000}-\x{FFFD}\x{10000}-\x{10FFFF}]@u', $text)
    ?>
    Ref: http://www.w3.org/TR/2000/REC-xml-20001006#charsets

    上篇:preg_match_all()

    下篇:preg_quote()