0

我会尽量简单地呈现它：我使用 json_encode() 对来自不同语言的多个 utf-8 字符串进行编码，我注意到当它们属于 ASCII 表时字符保持不变，但其他所有内容都返回为 ' \unnnn'，其中 'nnnn' 是一个十六进制数。

见代码：

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> 
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="content-type" content="application/xhtml+xml; charset=UTF-8" />
 <title>Multibyte string functions</title>
</head>
<body>
<h3>Multibyte string functions</h3>
<p>
<?php
//present json encode errors nicely:
//assign integer values to keys and error names to values
echo '<br /><b>Define JSON errors</b><br />';
$constants = get_defined_constants(true);
$json_errors = array();
foreach ($constants["json"] as $name => $value) {
   if (!strncmp($name, "JSON_ERROR_", 11)) {
      $json_errors[$value] = $name;
   }
}
echo nl2br(print_r($json_errors, true), true);

//Display current detection order
echo "<br /><b>Current detection order 'mb_detect_order()':</b> ", implode(", ", mb_detect_order());
//Display internal encoding
echo "<br /><b>Internal encoding 'mb_internal_encoding()':</b> ",  mb_internal_encoding();
//Get current language
echo "<br /><b>Current detection language 'mb_language()' ('neutral' for utf8):</b> ", mb_language();

//our test data
//a nowdoc that can break a <input> field;
$str = <<<'STR'
O'Reilly(\n) "& 'Big\Two @ <span>bo\tld</span>"
STR;
$strings = array(
   $str,
   "Latin: tell me the answer and I might find the question!",
   "Greek: πες μου την ερώτηση και ίσως βρω την απάντηση!",
   "Chinese simplified: 告诉我答复，并且我也许发现问题!",
   "Arabic: أخبرني الاجابة, انا قد تجد مسالة!",
   "Portuguese: mais coisas a pensar sobre diário ou dois!",
   "French: plus de choses à penser à journalier ou à deux!",
   "Spanish: ¡más cosas a pensar en diario o dos!",
   "Italian: più cose da pensare circa giornaliere o due!",
   "Danish: flere ting å tenke på hver dag eller to!",
   "Chech: Další věcí, přemýšlet o každý den nebo dva!",
   "German: mehr über Spaß spät schönen",
   "Albanian: më vonë gjatë fun bukur",
   "Hungarian: több mint szórakozás késő csodálatos kenyér"
);

//show encoding and then encode
foreach( $strings as $string ){
   echo "<br /><br />$string :", mb_detect_encoding($string);
   $json = json_encode($string);
   echo "<br />Error? ", $json_errors[json_last_error()];
   echo '<br />json=', $json;
}

上面的代码将输出：

Define JSON errors
Array
(
[0] => JSON_ERROR_NONE
[1] => JSON_ERROR_DEPTH
[2] => JSON_ERROR_STATE_MISMATCH
[3] => JSON_ERROR_CTRL_CHAR
[4] => JSON_ERROR_SYNTAX
[5] => JSON_ERROR_UTF8
)

Current detection order 'mb_detect_order()': ASCII, UTF-8
Internal encoding 'mb_internal_encoding()': ISO-8859-1
Current detection language 'mb_language()' ('neutral' for utf8): neutral

O'Reilly(\n) "& 'Big\Two @ bo\tld" :ASCII
Error? JSON_ERROR_NONE
json="O'Reilly(\\n) \"& 'Big\\Two @ bo\\tld<\/span>\""

Latin: tell me the answer and I might find the question! :ASCII
Error? JSON_ERROR_NONE
json="Latin: tell me the answer and I might find the question!"

Greek: πες μου την ερώτηση και ίσως βρω την απάντηση! :UTF-8
Error? JSON_ERROR_NONE
json="Greek: \u03c0\u03b5\u03c2 \u03bc\u03bf\u03c5 \u03c4\u03b7\u03bd \u03b5\u03c1\u03ce\u03c4\u03b7\u03c3\u03b7 \u03ba\u03b1\u03b9 \u03af\u03c3\u03c9\u03c2 \u03b2\u03c1\u03c9 \u03c4\u03b7\u03bd \u03b1\u03c0\u03ac\u03bd\u03c4\u03b7\u03c3\u03b7!"

Chinese simplified: 告诉我答复，并且我也许发现问题! :UTF-8
Error? JSON_ERROR_NONE
json="Chinese simplified: \u544a\u8bc9\u6211\u7b54\u590d\uff0c\u5e76\u4e14\u6211\u4e5f\u8bb8\u53d1\u73b0\u95ee\u9898!"

Arabic: أخبرني الاجابة, انا قد تجد مسالة! :UTF-8
Error? JSON_ERROR_NONE
json="Arabic: \u0623\u062e\u0628\u0631\u0646\u064a \u0627\u0644\u0627\u062c\u0627\u0628\u0629, \u0627\u0646\u0627 \u0642\u062f \u062a\u062c\u062f \u0645\u0633\u0627\u0644\u0629!"

Portuguese: mais coisas a pensar sobre diário ou dois! :UTF-8
Error? JSON_ERROR_NONE
json="Portuguese: mais coisas a pensar sobre di\u00e1rio ou dois!"

French: plus de choses à penser à journalier ou à deux! :UTF-8
Error? JSON_ERROR_NONE
json="French: plus de choses \u00e0 penser \u00e0 journalier ou \u00e0 deux!"

Spanish: ¡más cosas a pensar en diario o dos! :UTF-8
Error? JSON_ERROR_NONE
json="Spanish: \u00a1m\u00e1s cosas a pensar en diario o dos!"

Italian: più cose da pensare circa giornaliere o due! :UTF-8
Error? JSON_ERROR_NONE
json="Italian: pi\u00f9 cose da pensare circa giornaliere o due!"

Danish: flere ting å tenke på hver dag eller to! :UTF-8
Error? JSON_ERROR_NONE
json="Danish: flere ting \u00e5 tenke p\u00e5 hver dag eller to!"

Chech: Další věcí, přemýšlet o každý den nebo dva! :UTF-8
Error? JSON_ERROR_NONE
json="Chech: Dal\u0161\u00ed v\u011bc\u00ed, p\u0159em\u00fd\u0161let o ka\u017ed\u00fd den nebo dva!"

German: mehr über Spaß spät schönen :UTF-8
Error? JSON_ERROR_NONE
json="German: mehr \u00fcber Spa\u00df sp\u00e4t sch\u00f6nen"

Albanian: më vonë gjatë fun bukur :UTF-8
Error? JSON_ERROR_NONE
json="Albanian: m\u00eb von\u00eb gjat\u00eb fun bukur"

Hungarian: több mint szórakozás késő csodálatos kenyér :UTF-8
Error? JSON_ERROR_NONE
json="Hungarian: t\u00f6bb mint sz\u00f3rakoz\u00e1s k\u00e9s\u0151 csod\u00e1latos keny\u00e9r"

正如您在大多数语言中所看到的（英语除外），存在 utf-8 字符的十六进制转换。是否可以通过不替换我的 unicode 字符来进行编码？安全吗？其他人做什么？

您应该考虑来自页面中的用户输入并存储到 mysql 的此类编码。

谢谢。

4

2 回答 2

3

也许您应该尝试json_encode($string, JSON_UNESCAPED_UNICODE)，或http://php.net/manual/fr/function.json-encode.php中的任何方法，这可能对您的各种情况有用。

于 2013-03-02T22:33:45.663 回答
1

Ok, really thanks for the answer!

The problem is that I'm on version PHP Version 5.3.10 and json_encode($string, JSON_UNESCAPED_UNICODE) isn't an option.

Fortunately, a guy called "Mr Swordsteel" posted a comment at php's manual http://www.php.net/manual/en/function.json-encode.php which actually does the trick (thank you Mr Swordsteel!) The real paradox is that it emulates completely json_encode function and gives a hint if we want to port it to another language like javascript and keep our libraries communicative.

function my_json_encode($in){
   $_escape = function ($str) {
      return addcslashes($str, "\v\t\n\r\f\"\\/");
   };
   $out = "";
   if (is_object($in)){
      $class_vars = get_object_vars(($in));
      $arr = array();
      foreach ($class_vars as $key => $val){
         $arr[$key] = "\"{$_escape($key)}\":\"{$val}\"";
       }
      $val = implode(',', $arr);
      $out .= "{{$val}}";
   }elseif (is_array($in)){
      $obj = false;
      $arr = array();
      foreach($in as $key => $val){
         if(!is_numeric($key)){
            $obj = true;
         }
         $arr[$key] = my_json_encode($val);
      }
      if($obj){
         foreach($arr AS $key => $val){
            $arr[$key] = "\"{$_escape($key)}\":{$val}";
         }
         $val = implode(',', $arr);
         $out .= "{{$val}}";
      }else {
         $val = implode(',', $arr);
         $out .= "[{$val}]";
      }
   }elseif (is_bool($in)){
      $out .= $in ? 'true' : 'false';
   }elseif (is_null($in)){
      $out .= 'null';
   }elseif (is_string($in)){
      $out .= "\"{$_escape($in)}\"";debug('in='.$in.', $_escape($in)='.$_escape($in).', out='.$out);
      }else{
         $out .= $in;
      }
      return "{$out}";
   }

I gave it a lot of tests and couldn't break it! It would be very interesting now to re-implement json_decode!

Thanks.

于 2013-03-05T17:58:21.520 回答