Skip to content

Instantly share code, notes, and snippets.

@mingyun
Created December 24, 2013 03:39
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save mingyun/8108504 to your computer and use it in GitHub Desktop.
Save mingyun/8108504 to your computer and use it in GitHub Desktop.
PHP获取可以用GBK编码的汉字拼音首字母
<?php
/**
* 中文拼音
* 目前仅支持可以以GBK编码的20902个汉字提取拼音首字母
*/
class Spell{
//当前字符
protected $word='';
//资源
protected $source = array();
protected $pos = array();
//运行期间数据
protected $temp = array();
//数据暂存标识//在运行期间多次使用时有明显效果
protected $flag = true;
//输入输出字符集
protected $charset = array('in'=>'gbk','out'=>'utf-8');
//例外处理//当没有拼音首字母other===false的情况下,返回原字串否则设置为该值
protected $other = '!';//other=false;
public function __construct($flag=true){
$this->flag = $flag;
//加载资源
$this->source['gk221']=file_get_contents(DATA_PATH.'word/gk2-2-1.txt');
$this->source['gk31']=file_get_contents(DATA_PATH.'word/gk3-1.txt');
$this->source['gk41']=file_get_contents(DATA_PATH.'word/gk4-1.txt');
$this->pos=json_decode(file_get_contents(DATA_PATH.'word/pos.txt'),true);
}
/**
* 获得拼音首字母
* @param string/arr $data 数据,可以是字符串和数组
* @param string $in 数据的编码
* @param string $out 输出的编码
* @return arr 返回数组
*/
public function getInitial($data,$in='gbk',$out='utf-8'){
if(is_string($data)){
return self::getInitialByStr($data,$in,$out);
}elseif(is_array($data)){
return self::getInitialByArr($data,$in,$out);
}
}
/**
* 获得拼音首字母
* @param string $data 字符串数据
* @param string $in 数据的编码
* @param string $out 输出的编码
* @return arr 返回数组
*/
public function getInitialByStr($str,$in='gbk',$out='utf-8'){
$this->charset['in'] = strtolower($in);
$this->charset['out'] = strtolower($out);
if($this->flag!=true){
$this->temp['fws']= array();
}
switch ($this->charset['in']) {
case 'gbk':
return self::_getInitialInGBK($str);
break;
case 'utf-8':
return self::_getInitialInUTF8($str);
break;
default:
# code...
break;
}
//历史数据
if($this->flag!=true){
unset($this->temp['fws']);
}
}
/**
* 获得拼音首字母
* @param array $data 数组数据
* @param string $in 数据的编码
* @param string $out 输出的编码
* @return arr 返回数组
*/
public function getInitialByArr($arr,$in='gbk',$out='utf-8'){
$this->charset['in'] = strtolower($in);
$this->charset['out'] = strtolower($out);
if($this->flag!=true){
$this->temp['fws']= array();
}
switch ($this->charset['in']) {
case 'gbk':
return self::_getInitialInGBKArr($arr);
break;
case 'utf-8':
return self::_getInitialInUTF8Arr($arr);
break;
default:
# code...
break;
}
//历史数据
if($this->flag!=true){
unset($this->temp['fws']);
}
}
/**
* 处理gbk编码字符串的首字母
* @param string $str 字符串
* @return array 数组
*/
protected function _getInitialInGBK($str){
//存放字符串拼音
$w = array();
$i = 0;
$str_length = strlen($str); //字符串的字节数
while ($i<$str_length){
//得到字符串中第$i位字符的ASCII码
$ascnum = ord($str[$i]);
if($ascnum >= 0x81){//gbk区域
$nstr = substr($str, $i, 2);
$i = $i + 2;
}else{
$nstr = substr($str, $i, 1);
$i = $i + 1;
}
if(isset($this->temp['fws'][$nstr])){
$w[] = $this->temp['fws'][$nstr];
}else{
$w[] = self::_preGetInitial($nstr);
}
}
return $w;
}
/**
* 处理gbk编码数组的首字母
* @param array $arr 字符串单字数组
* @return array 数组
*/
protected function _getInitialInGBKArr($arr){
//存放字符串拼音
$w = array();
foreach ($arr as $key => $word) {
if(isset($this->temp['fws'][$word])){
$w[] = $this->temp['fws'][$word];
}else{
$w[] = self::_preGetInitial($word);
}
}
return $w;
}
/**
* 处理utf-8编码字符串的首字母
* @param string $str 字符串
* @return array 数组
*/
protected function _getInitialInUTF8($str){
//存放字符串拼音
$w = array();
$nstr = '';
$i = 0;
$str_length = strlen($str); //字符串的字节数
while ($i<$str_length){
$ascnum = ord($str[$i]); //得到字符串中第$i位字符的ASCII码
if ( $ascnum >= 252){//如果ASCII位高与252
$nstr = substr($str, $i, 6); //根据UTF-8编码规范,将6个连续的字符计为单个字符
$i = $i + 6; //实际Byte计为6
}elseif ( $ascnum >= 248 ){//如果ASCII位高与248
$nstr = substr($str, $i, 5); //根据UTF-8编码规范,将5个连续的字符计为单个字符
$i = $i + 5; //实际Byte计为5
}elseif ( $ascnum >= 240 ){//如果ASCII位高与240
$nstr = substr($str, $i, 4); //根据UTF-8编码规范,将4个连续的字符计为单个字符
$i = $i + 4; //实际Byte计为4
}elseif ( $ascnum >= 224 ){//如果ASCII位高与224
$nstr = substr($str, $i, 3); //根据UTF-8编码规范,将3个连续的字符计为单个字符
$i = $i + 3 ; //实际Byte计为3
}elseif ( $ascnum >= 192 ){//如果ASCII位高与192
$nstr = substr($str, $i, 2); //根据UTF-8编码规范,将2个连续的字符计为单个字符
$i = $i + 2; //实际Byte计为2
}else{//其他情况下,包括大写字母,小写字母和半角标点符号,%,&,@,m,w等
$nstr = substr($str, $i, 1);
$i = $i + 1; //实际的Byte数计1个
}
//编码转换至GBK
$nstr = iconv('utf-8','gbk',$nstr);
if(isset($this->temp['fws'][$nstr])){
$w[] = $this->temp['fws'][$nstr];
}else{
$w[] = self::_preGetInitial($nstr);
}
}
return $w;
}
/**
* 处理utf-8编码数组的首字母
* @param array $arr 字符串单字数组
* @return array 数组
*/
protected function _getInitialInUTF8Arr($arr){
//存放字符串拼音
$w = array();
foreach ($arr as $key => $word) {
$nword = iconv('utf-8','GBK',$word);
if(isset($this->temp['fws'][$nword])){
$w[] = $this->temp['fws'][$nword];
}else{
$w[] = self::_preGetInitial($nword);
}
}
return $w;
}
/**
* 对单字预处理
* @param string $word 单字,gbk编码
* @return string 拼音首字母,编码视$this->charset['out']
*/
protected function _preGetInitial($word){
$this->word = iconv('gbk', 'utf-8',$word);
$fw = self::_getInitial($word);//返回的utf-8编码数据首字母
if($fw!==false){
$nstr=$this->temp['fws'][$word]=iconv('utf-8',$this->charset['out'],$fw);
}else{
$nstr=$this->temp['fws'][$word]=iconv('gbk',$this->charset['out'],$word);
}
return $nstr;
}
/**
* 获得汉字拼音首字母的核心函数
* @param string $word 单字,gbk编码
* @return string 首字母,utf-8编码
*/
protected function _getInitial($word){
$high = ord($word{0});
$low = ord($word{1});
//对20902个汉字支持拼音首字母提取
$hexc = $high * 256 + $low;
//GBK/2:gb2312汉字表(拼音排序),低位a0开始
if($hexc >= 0xB0A1 and $hexc <= 0xD7F9 and $low>=0xA0){
//共3755个字
return self::_getInGBK21($hexc);
}
//GBK/2:gb2312汉字表,低位a0开始
if($hexc >= 0xD8A1 and $hexc <= 0xF7FE and $low>=0xA0){
//共3008个字
return self::_getInGBK('gk221');
}
//GBK/3:扩充汉字表(UCS 代码大小排列)
if($hexc >= 0x8140 and $hexc <= 0xA0FE){
//共6080个字
return self::_getInGBK('gk31');
}
//GBK/4:扩充汉字表(按康熙字典的页码/字位排列)
//低位00开始
if($hexc >= 0xAA40 and $hexc <= 0xFEA0){
//共8059个字
return self::_getInGBK('gk41');
}
//都没有
return $this->other;
}
/**
* 获取首字母
* GBK/2:gb2312汉字表(拼音序)
* 共3755个字
* @param int $hexc 单字GBK编码值
* @return string 首字母,utf-8编码
*/
protected function _getInGBK21($hexc){
//无i,u,v开始的拼音
$char = array("",//填充位置
"A","B","C","D","E","F",
"G","H","J","K","L","M",
"N","O","P","Q","R","S",
"T","W","X","Y","Z"
);
$hcs = array(
0xB0A1,0xb0c5,0xb2c1,0xb4ee,0xb6ea,0xb7a2,
0xb8c1,0xb9fe,0xbbf7,0xbfa6,0xc0ac,0xc2e8,
0xc4c3,0xc5b6,0xc5be,0xc6da,0xc8bb,0xc8f6,
0xcbfa,0xcdda,0xcef4,0xd1b9,0xd4d1
);
if($key=array_search($hexc,$hcs)){
return $char[$key];
}else{
$hcs[] = $hexc;
sort($hcs);
return $char[array_search($hexc,$hcs)];
}
}
/**
* 获取首字母
* @param string $type 单字所属GBK区域类型
* @return string 首字母,utf-8编码
*/
protected function _getInGBK($type){
//无i,u,v开始的拼音
$char = array("",//填充位置
"A","B","C","D","E","F",
"G","H","J","K","L","M",
"N","O","P","Q","R","S",
"T","W","X","Y","Z"
);
$str = str_replace("\r\n",'',$this->source[$type]);
$p = stripos($str,$this->word)+3;//居右//stripos($str,$word),居左
$str = '';
if($key=array_search($p,$this->pos[$type])){
return $char[$key];
}else{
$pos = $this->pos[$type];
$pos[] = $p;
sort($pos);
return $char[array_search($p,$pos)];
}
}
}
define('DATA_PATH',__DIR__.'/');
$t1 = microtime(true);
$py = new Spell();
$str =file_get_contents(DATA_PATH.'test/18.txt');
$r = $py->getInitial($str,'utf-8','utf-8');
$t2 = microtime(true);
echo $t2.'<br>';
echo intval(($t2-$t1)*1000),'毫秒';
exit;
define('DATA_PATH',__DIR__.'/');
$t1 = microtime(true);
$py = new Spell();
$str =file_get_contents(DATA_PATH.'test/18-gbk.txt');
$r = $py->getInitial($str,'gbk','utf-8');
$t2 = microtime(true);
echo $t2.'<br>';
echo intval(($t2-$t1)*1000),'毫秒';
?>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment