网上的方法有不少，都是一样的原理，按照需求，做了一下版本的class类文件，主要功能是：功能明确，易于修改维护和扩展； 英文的字串：不变返回(包括数字)；中文字符串：返回拼音首字符； 中英混合串: 返回拼音首字符和英文。该算法采用了二分法查找，修复了之前字母Z读取成Y的错误。好东西要收藏，故在此留下印记，以供后人考证！
复制代码 代码如下:
<?php
/**
* Modified by http://iulog.com @ 2013-05-07
* 修复二分法查找方法
* 汉字拼音首字母工具类
* 注： 英文的字串：不变返回(包括数字) eg .abc123 => abc123
* 中文字符串：返回拼音首字符 eg. 测试字符串 => CSZFC
* 中英混合串: 返回拼音首字符和英文 eg. 我i我j => WIWJ
* eg.
* $py = new str2PY();
* $result = $py->getInitials('啊吧才的饿飞就好i就看了吗你哦平去人是他uv我想一在');
*/
class str2PY
{
private $_pinyins = array(
176161 => 'A',
176197 => 'B',
178193 => 'C',
180238 => 'D',
182234 => 'E',
183162 => 'F',
184193 => 'G',
185254 => 'H',
187247 => 'J',
191166 => 'K',
192172 => 'L',
194232 => 'M',
196195 => 'N',
197182 => 'O',
197190 => 'P',
198218 => 'Q',
200187 => 'R',
200246 => 'S',
203250 => 'T',
205218 => 'W',
206244 => 'X',
209185 => 'Y',
212209 => 'Z',
);
private $_charset = null;
/**
* 构造函数, 指定需要的编码 default: utf-8
* 支持utf-8, gb2312
*
* @param unknown_type $charset
*/
public function __construct( $charset = 'utf-8' )
{
$this->_charset = $charset;
}
/**
* 中文字符串 substr
*
* @param string $str
* @param int $start
* @param int $len
* @return string
*/
private function _msubstr ($str, $start, $len)
{
$start = $start * 2;
$len = $len * 2;
$strlen = strlen($str);
$result = '';
for ( $i = 0; $i < $strlen; $i++ ) {
if ( $i >= $start && $i < ($start + $len) ) {
if ( ord(substr($str, $i, 1)) > 129 ) $result .= substr($str, $i, 2);
else $result .= substr($str, $i, 1);
}
if ( ord(substr($str, $i, 1)) > 129 ) $i++;
}
return $result;
}
/**
* 字符串切分为数组 (汉字或者一个字符为单位)
*
* @param string $str
* @return array
*/
private function _cutWord( $str )
{
$words = array();
while ( $str != "" )
{
if ( $this->_isAscii($str) ) {
$words[] = $str[0];
$str = substr( $str, strlen($str[0]) );
}else{
$word = $this->_msubstr( $str, 0, 1 );
$words[] = $word;
$str = substr( $str, strlen($word) );
}
}
return $words;
}
/**
* 判断字符是否是ascii字符
*
* @param string $char
* @return bool
*/
private function _isAscii( $char )
{
return ( ord( substr($char,0,1) ) < 160 );
}
/**
* 判断字符串前3个字符是否是ascii字符
*
* @param string $str
* @return bool
*/
private function _isAsciis( $str )
{
$len = strlen($str) >= 3 ? 3: 2;
$chars = array();
for( $i = 1; $i < $len -1; $i++ ){
$chars[] = $this->_isAscii( $str[$i] ) ? 'yes':'no';
}
$result = array_count_values( $chars );
if ( empty($result['no']) ){
return true;
}
return false;
}
/**
* 获取中文字串的拼音首字符
*
* @param string $str
* @return string
*/
public function getInitials( $str )
{
if ( empty($str) ) return '';
if ( $this->_isAscii($str[0]) && $this->_isAsciis( $str )){
return $str;
}
$result = array();
if ( $this->_charset == 'utf-8' ){
$str = iconv( 'utf-8', 'gb2312', $str );
}
$words = $this->_cutWord( $str );
foreach ( $words as $word )
{
if ( $this->_isAscii($word) ) {
$result[] = $word;
continue;
}
$code = ord( substr($word,0,1) ) * 1000 + ord( substr($word,1,1) );

if ( ($i = $this->_search($code)) != -1 ){
$result[] = $this->_pinyins[$i];
}
}
return strtoupper(implode('',$result));
}
private function _getChar( $ascii )
{
if ( $ascii >= 48 && $ascii <= 57){
return chr($ascii);
}elseif ( $ascii>=65 && $ascii<=90 ){
return chr($ascii);
}elseif ($ascii>=97 && $ascii<=122){
return chr($ascii-32);
}else{
return '-';
}
}

/**
* 查找需要的汉字内码(gb2312) 对应的拼音字符( 二分法 )
*
* @param int $code
* @return int
*/
private function _search( $code )
{
$data = array_keys($this->_pinyins);
$lower = 0;
$upper = sizeof($data)-1;
$middle = (int) round(($lower + $upper) / 2);
if ( $code < $data[0] ) return -1;
for (;;) {
if ( $lower > $upper ){
return $data[$lower-1];
}
$tmp = (int) round(($lower + $upper) / 2);
if ( !isset($data[$tmp]) ){
return $data[$middle];
}else{
$middle = $tmp;
}
if ( $data[$middle] < $code ){
$lower = (int)$middle + 1;
}else if ( $data[$middle] == $code ) {
return $data[$middle];
}else{
$upper = (int)$middle - 1;
}
}
}
}
?>