汉字转换成拼音

隐藏行号 复制代码 ？ 这是一段程序代码。 
  1. <?php
  2. /**
  3.  * 汉字转换成拼音
  4.  * @author sunny5156
  5.  * @version 1.0.0
  6.  * */
  7. class Pinyin {
  8.     private $_outCoding = 'gb2312';
  9.     private $_inCoding = '';
  10.     private $str = '';
  11.     private $pix = ' '; //获取全拼时的分隔符
  12.     private $_dataKey = "a|ai|an|ang|ao|ba|bai|ban|bang|bao|bei|ben|beng|bi|bian|biao|bie|bin|bing|bo|bu|ca|cai|can|cang|cao|ce|ceng|cha|chai|chan|chang|chao|che|chen|cheng|chi|chong|chou|chu|chuai|chuan|chuang|chui|chun|chuo|ci|cong|cou|cu|cuan|cui|cun|cuo|da|dai|dan|dang|dao|de|deng|di|dian|diao|die|ding|diu|dong|dou|du|duan|dui|dun|duo|e|en|er|fa|fan|fang|fei|fen|feng|fo|fou|fu|ga|gai|gan|gang|gao|ge|gei|gen|geng|gong|gou|gu|gua|guai|guan|guang|gui|gun|guo|ha|hai|han|hang|hao|he|hei|hen|heng|hong|hou|hu|hua|huai|huan|huang|hui|hun|huo|ji|jia|jian|jiang|jiao|jie|jin|jing|jiong|jiu|ju|juan|jue|jun|ka|kai|kan|kang|kao|ke|ken|keng|kong|kou|ku|kua|kuai|kuan|kuang|kui|kun|kuo|la|lai|lan|lang|lao|le|lei|leng|li|lia|lian|liang|liao|lie|lin|ling|liu|long|lou|lu|lv|luan|lue|lun|luo|ma|mai|man|mang|mao|me|mei|men|meng|mi|mian|miao|mie|min|ming|miu|mo|mou|mu|na|nai|nan|nang|nao|ne|nei|nen|neng|ni|nian|niang|niao|nie|nin|ning|niu|nong|nu|nv|nuan|nue|nuo|o|ou|pa|pai|pan|pang|pao|pei|pen|peng|pi|pian|piao|pie|pin|ping|po|pu|qi|qia|qian|qiang|qiao|qie|qin|qing|qiong|qiu|qu|quan|que|qun|ran|rang|rao|re|ren|reng|ri|rong|rou|ru|ruan|rui|run|ruo|sa|sai|san|sang|sao|se|sen|seng|sha|shai|shan|shang|shao|she|shen|sheng|shi|shou|shu|shua|shuai|shuan|shuang|shui|shun|shuo|si|song|sou|su|suan|sui|sun|suo|ta|tai|tan|tang|tao|te|teng|ti|tian|tiao|tie|ting|tong|tou|tu|tuan|tui|tun|tuo|wa|wai|wan|wang|wei|wen|weng|wo|wu|xi|xia|xian|xiang|xiao|xie|xin|xing|xiong|xiu|xu|xuan|xue|xun|ya|yan|yang|yao|ye|yi|yin|ying|yo|yong|you|yu|yuan|yue|yun|za|zai|zan|zang|zao|ze|zei|zen|zeng|zha|zhai|zhan|zhang|zhao|zhe|zhen|zheng|zhi|zhong|zhou|zhu|zhua|zhuai|zhuan|zhuang|zhui|zhun|zhuo|zi|zong|zou|zu|zuan|zui|zun|zuo";
  13.     private $_dataValue = "-20319|-20317|-20304|-20295|-20292|-20283|-20265|-20257|-20242|-20230|-20051|-20036|-20032|-20026|-20002|-19990|-19986|-19982|-19976|-19805|-19784|-19775|-19774|-19763|-19756|-19751|-19746|-19741|-19739|-19728|-19725|-19715|-19540|-19531|-19525|-19515|-19500|-19484|-19479|-19467|-19289|-19288|-19281|-19275|-19270|-19263|-19261|-19249|-19243|-19242|-19238|-19235|-19227|-19224|-19218|-19212|-19038|-19023|-19018|-19006|-19003|-18996|-18977|-18961|-18952|-18783|-18774|-18773|-18763|-18756|-18741|-18735|-18731|-18722|-18710|-18697|-18696|-18526|-18518|-18501|-18490|-18478|-18463|-18448|-18447|-18446|-18239|-18237|-18231|-18220|-18211|-18201|-18184|-18183|-18181|-18012|-17997|-17988|-17970|-17964|-17961|-17950|-17947|-17931|-17928|-17922|-17759|-17752|-17733|-17730|-17721|-17703|-17701|-17697|-17692|-17683|-17676|-17496|-17487|-17482|-17468|-17454|-17433|-17427|-17417|-17202|-17185|-16983|-16970|-16942|-16915|-16733|-16708|-16706|-16689|-16664|-16657|-16647|-16474|-16470|-16465|-16459|-16452|-16448|-16433|-16429|-16427|-16423|-16419|-16412|-16407|-16403|-16401|-16393|-16220|-16216|-16212|-16205|-16202|-16187|-16180|-16171|-16169|-16158|-16155|-15959|-15958|-15944|-15933|-15920|-15915|-15903|-15889|-15878|-15707|-15701|-15681|-15667|-15661|-15659|-15652|-15640|-15631|-15625|-15454|-15448|-15436|-15435|-15419|-15416|-15408|-15394|-15385|-15377|-15375|-15369|-15363|-15362|-15183|-15180|-15165|-15158|-15153|-15150|-15149|-15144|-15143|-15141|-15140|-15139|-15128|-15121|-15119|-15117|-15110|-15109|-14941|-14937|-14933|-14930|-14929|-14928|-14926|-14922|-14921|-14914|-14908|-14902|-14894|-14889|-14882|-14873|-14871|-14857|-14678|-14674|-14670|-14668|-14663|-14654|-14645|-14630|-14594|-14429|-14407|-14399|-14384|-14379|-14368|-14355|-14353|-14345|-14170|-14159|-14151|-14149|-14145|-14140|-14137|-14135|-14125|-14123|-14122|-14112|-14109|-14099|-14097|-14094|-14092|-14090|-14087|-14083|-13917|-13914|-13910|-13907|-13906|-13905|-13896|-13894|-13878|-13870|-13859|-13847|-13831|-13658|-13611|-13601|-13406|-13404|-13400|-13398|-13395|-13391|-13387|-13383|-13367|-13359|-13356|-13343|-13340|-13329|-13326|-13318|-13147|-13138|-13120|-13107|-13096|-13095|-13091|-13076|-13068|-13063|-13060|-12888|-12875|-12871|-12860|-12858|-12852|-12849|-12838|-12831|-12829|-12812|-12802|-12607|-12597|-12594|-12585|-12556|-12359|-12346|-12320|-12300|-12120|-12099|-12089|-12074|-12067|-12058|-12039|-11867|-11861|-11847|-11831|-11798|-11781|-11604|-11589|-11536|-11358|-11340|-11339|-11324|-11303|-11097|-11077|-11067|-11055|-11052|-11045|-11041|-11038|-11024|-11020|-11019|-11018|-11014|-10838|-10832|-10815|-10800|-10790|-10780|-10764|-10587|-10544|-10533|-10519|-10331|-10329|-10328|-10322|-10315|-10309|-10307|-10296|-10281|-10274|-10270|-10262|-10260|-10256|-10254";
  14.     
  15.     // 设置字符串
  16.     public function setStr($str) {
  17.         $this->str = $str;
  18.     }
  19.     // 获取字符串
  20.     public function getStr() {
  21.         return $this->str;
  22.     }
  23.     // 获取全拼
  24.     public function getPinyin() {
  25.         $_TDataKey = explode ( '|', $this->_dataKey );
  26.         $_TDataValue = explode ( '|', $this->_dataValue );
  27.         
  28.         $data = (PHP_VERSION >= '5.0') ? array_combine ( $_TDataKey, $_TDataValue ) : $this->_Array_Combine ( $_TDataKey, $_TDataValue );
  29.         
  30.         arsort ( $data );
  31.         reset ( $data );
  32.         echo self::getStr () . '<br>';
  33.         $str = $this->safe_encoding ( self::getStr () );
  34.         $_Res = '';
  35.         $arr = array();
  36.         for($i = 0; $i < strlen ( $str ); $i ++) {
  37.             $_P = ord ( substr ( $str, $i, 1 ) );
  38.             if ($_P > 160) {
  39.                 $_Q = ord ( substr ( $str, ++ $i, 1 ) );
  40.                 $_P = $_P * 256 + $_Q - 65536;
  41.             }
  42. //             $_Res .= $this->_Pinyin ( $_P, $data ) . $this->pix;
  43.             $arr[] = $this->_Pinyin ( $_P, $data );
  44.         }
  45.         
  46.         return implode($arr, $this->pix);
  47.     }
  48.     
  49.     // 获取一个字符的拼音首字母
  50.     private function getFristChar($str) {
  51.         if (! $str)
  52.             return null;
  53.             // echo $str{0};
  54.         $fchar = ord ( $str {0} );
  55.         if ($fchar >= ord ( "A" ) and $fchar <= ord ( "z" ))
  56.             return strtoupper ( $str {0} );
  57.         $s = $this->safe_encoding ( $str );
  58.         $asc = ord ( $s {0} ) * 256 + ord ( $s {1} ) - 65536;
  59.         if ($asc >= -20319 and $asc <= - 20284)
  60.             return "A";
  61.         if ($asc >= -20283 and $asc <= -19776)
  62.             return "B";
  63.         if ($asc >= -19775 and $asc <= -19219)
  64.             return "C";
  65.         if ($asc >= -19218 and $asc <= -18711)
  66.             return "D";
  67.         if ($asc >= -18710 and $asc <= -18527)
  68.             return "E";
  69.         if ($asc >= -18526 and $asc <= -18240)
  70.             return "F";
  71.         if ($asc >= -18239 and $asc <= -17923)
  72.             return "G";
  73.         if ($asc >= -17922 and $asc <= -17418)
  74.             return "H";
  75.         if ($asc >= -17417 and $asc <= -16475)
  76.             return "J";
  77.         if ($asc >= -16474 and $asc <= -16213)
  78.             return "K";
  79.         if ($asc >= -16212 and $asc <= -15641)
  80.             return "L";
  81.         if ($asc >= -15640 and $asc <= -15166)
  82.             return "M";
  83.         if ($asc >= -15165 and $asc <= -14923)
  84.             return "N";
  85.         if ($asc >= -14922 and $asc <= -14915)
  86.             return "O";
  87.         if ($asc >= -14914 and $asc <= -14631)
  88.             return "P";
  89.         if ($asc >= -14630 and $asc <= -14150)
  90.             return "Q";
  91.         if ($asc >= -14149 and $asc <= -14091)
  92.             return "R";
  93.         if ($asc >= -14090 and $asc <= -13319)
  94.             return "S";
  95.         if ($asc >= -13318 and $asc <= -12839)
  96.             return "T";
  97.         if ($asc >= -12838 and $asc <= -12557)
  98.             return "W";
  99.         if ($asc >= -12556 and $asc <= -11848)
  100.             return "X";
  101.         if ($asc >= -11847 and $asc <= -11056)
  102.             return "Y";
  103.         if ($asc >= -11055 and $asc <= -10247)
  104.             return "Z";
  105.         return null;
  106.     }
  107.     
  108.     // 分离字符串
  109.     private function getSplitChar() {
  110.         $str = self::getStr ();
  111.         $arr = array ();
  112.         // echo mb_strlen($str,'utf-8').'<br>';
  113.         for($i = 0; $i < mb_strlen( $str, 'utf-8' ); $i ++) {
  114.             $arr [] = mb_substr ( $str, $i, 1, 'utf-8' );
  115.         }
  116.         
  117.         return json_encode ( $arr );
  118.     }
  119.     
  120.     // 获取字符串每个字符的首字母
  121.     public function getStrFirstChar() {
  122.         $pix = '';
  123.         $str = array ();
  124.         $data = json_decode ( self::getSplitChar () );
  125.         foreach ( $data as $item ) {
  126.             $str [] = self::getFristChar ( $item );
  127.         }
  128.         
  129.         return implode ( $str, $pix );
  130.     }
  131.     private function _Pinyin($_Num, $_Data) {
  132.         if ($_Num > 0 && $_Num < 160)
  133.             return chr ( $_Num );
  134.         elseif ($_Num < - 20319 || $_Num > - 10247)
  135.             return '';
  136.         else {
  137.             foreach ( $_Data as $k => $v ) {
  138.                 if ($v <= $_Num)
  139.                     break;
  140.             }
  141.             return $k;
  142.         }
  143.     }
  144.     
  145.     
  146.     private function safe_encoding($string) {
  147.         $encoding = "utf-8";
  148.         for($i = 0; $i < strlen ( $string ); $i ++) {
  149.             if (ord ( $string {$i} ) < 128)
  150.                 continue;
  151.             if ((ord ( $string {$i} ) & 224) == 224) { // 第一个字节判断通过
  152.                 $char = $string {++ $i};
  153.                 if ((ord ( $char ) & 128) == 128) { // 第二个字节判断通过
  154.                     $char = $string {++ $i};
  155.                     if ((ord ( $char ) & 128) == 128) {
  156.                         $encoding = "UTF-8";
  157.                         break;
  158.                     }
  159.                 }
  160.             }
  161.             if ((ord ( $string {$i} ) & 192) == 192) { // 第一个字节判断通过
  162.                 $char = $string {++ $i};
  163.                 if ((ord ( $char ) & 128) == 128) { // 第二个字节判断通过
  164.                     $encoding = "GB2312";
  165.                     break;
  166.                 }
  167.             }
  168.         }
  169.         if (strtoupper ( $encoding ) == strtoupper ( $this->_outCoding ))
  170.             return $string;
  171.         else
  172.             return iconv ( $encoding, $this->_outCoding, $string );
  173.     }
  174.     
  175.     
  176.     private function _Array_Combine($_Arr1, $_Arr2) {
  177.         for($i = 0; $i < count ( $_Arr1 ); $i ++)
  178.             $_Res [$_Arr1 [$i]] = $_Arr2 [$i];
  179.         return $_Res;
  180.     }
  181. }
posted @ 2012-11-01 10:51  sunny5156  阅读(1294)  评论(0)  编辑  收藏  举报