function test_search($test_enc, $str, $look_for, $opt, $in_enc = 'EUC-JP')
{
mb_regex_encoding($test_enc);
$str = mb_convert_encoding($str, $test_enc, $in_enc);
$look_for = mb_convert_encoding($look_for, $test_enc, $in_enc);
mb_ereg_search_init($str, $look_for, $opt);
while (mb_ereg_search_pos()) {
$regs = mb_ereg_search_getregs();
array_shift($regs);
printf("(%s) (%d) %s\n", $test_enc, mb_ereg_search_getpos(), mb_convert_encoding(is_array($regs) ? implode('-', $regs) : '', $in_enc, $test_enc));
}
}
/**
* Regular expression split and return all parts.
*
* @param string $pattern Pattern
* @param string $subject Subject
* @param int $limit Limit
* @param string $option Option
* @return string[] Array of split parts, array with original string otherwise
* @throws MbRegexException When compilation error occurs
* @link http://php.net/function.mb-split.php
*/
public static function split($pattern, $subject, $option = '', $limit = -1)
{
static::setUp($pattern);
$position = 0;
$lastPosition = 0;
$res = array();
$subjectLen = \mb_strlen($subject);
do {
\mb_ereg_search_init($subject, $pattern, $option);
\mb_ereg_search_setpos($position);
$matches = \mb_ereg_search_regs();
if ($matches === false) {
break;
}
$position = \mb_ereg_search_getpos();
$res[] = \mb_substr($subject, $lastPosition, $position - \mb_strlen($matches[0]) - $lastPosition);
$lastPosition = $position;
} while ($position < $subjectLen && --$limit !== 1);
if ($lastPosition <= $subjectLen) {
$res[] = \mb_substr($subject, $lastPosition);
}
static::tearDown();
return $res;
}
$num = 4;
$str = "This string has four words.";
$str = mb_ereg_replace("four", $num, $str);
var_dump($str);
$test = "http://test.com/test";
$test = mb_ereg_replace("[[:alpha:]]+://[^<>[:space:]]+[[:alnum:]/]", "<a href=\"\\0\">\\0</a>", $test);
var_dump($test);
$str = "PrÜÝ" . "fung abc pÜ";
$reg = "\\w+";
mb_regex_encoding("UTF-8");
mb_ereg_search_init($str, $reg);
$r = mb_ereg_search();
$r = mb_ereg_search_getregs();
// get first result
var_dump($r === array("PrÜÝ" . "fung"));
var_dump(mb_ereg_search_getpos());
$str = "PrÜÝ" . "fung abc pÜ";
$reg = "\\w+";
mb_regex_encoding("UTF-8");
mb_ereg_search_init($str, $reg);
$r = mb_ereg_search();
$r = mb_ereg_search_getregs();
// get first result
var_dump($r == array("PrÜÝ" . "fung"));
$str = "PrÜÝ" . "fung abc pÜ";
$reg = "\\w+";
mb_regex_encoding("UTF-8");
mb_ereg_search_init($str, $reg);
$r = mb_ereg_search();
$r = mb_ereg_search_getregs();
// get first result
/**
* split text to search tokens
*
* @access private
* @param string $text 'UTF-8' encoded search text
* @return array array of search text token
*/
function _split_to_tokens($text)
{
$tokens = array();
// set search token patterns
// 1. double quoted phrase
// 2. single byte word contains html entities and latin1 letters
// 3. multi byte word
// 4. symbol - !#$%&'()*+,-./:;<=>?@[\]~_`{|}~ and latin1 supplement symbol
$pattern = sprintf('%s|%s|%s|%s', $this->_regex_patterns['phrase'], $this->_regex_patterns['sbword'], $this->_regex_patterns['mbword'], $this->_regex_patterns['symbol']);
mb_ereg_search_init($text, $pattern);
$len = strlen($text);
for ($i = 0; $i < $len; $i = mb_ereg_search_getpos()) {
mb_ereg_search_setpos($i);
$regs = mb_ereg_search_regs();
if ($regs === false) {
break;
}
// put back token encoding if changed to 'UTF-8'
$tokens[] = $regs[0];
}
return $tokens;
}