用PHP將Unicode 轉換為UTF-8

function unescape($str) { $str = rawurldecode($str);

<code> preg_match_all(<span>"/(?:<span>%u</span>.{4})|&#x.{4};|&#\d+;|.+/U"</span>,<span>$str</span>,<span>$r</span>); <span>$ar</span> = <span>$r</span>[<span>0</span>]; <span>//print</span>_r(<span>$ar</span>); <span>foreach</span>(<span>$ar</span> as <span>$k</span>=><span>$v</span>) { <span>if</span>(<span>substr</span>(<span>$v</span>,<span>0</span>,<span>2</span>) == <span>"<span>%u</span>"</span>){ <span>$ar</span>[<span>$k</span>] = iconv(<span>"UCS-2BE"</span>,<span>"UTF-8"</span>,<span>pack</span>(<span>"H4"</span>,<span>substr</span>(<span>$v</span>,-<span>4</span>)));
  }
        elseif(<span>substr</span>(<span>$v</span>,<span>0</span>,<span>3</span>) == <span>"&#x"</span>){ <span>$ar</span>[<span>$k</span>] = iconv(<span>"UCS-2BE"</span>,<span>"UTF-8"</span>,<span>pack</span>(<span>"H4"</span>,<span>substr</span>(<span>$v</span>,<span>3</span>,-<span>1</span>)));
  }
        elseif(<span>substr</span>(<span>$v</span>,<span>0</span>,<span>2</span>) == <span>"&#"</span>) { <span>$ar</span>[<span>$k</span>] = iconv(<span>"UCS-2BE"</span>,<span>"UTF-8"</span>,<span>pack</span>(<span>"n"</span>,<span>substr</span>(<span>$v</span>,<span>2</span>,-<span>1</span>)));
        }
    } <span>return</span><span>join</span>(<span>""</span>,<span>$ar</span>);
}
echo unescape(<span>"紫星蓝"</span>);</code>

今天有用户反馈，表单系统用户提交的数据中文会乱码。测试发现问题出在 iconv 转换上。
iconv(‘UCS-2’, ‘GBK’, ‘中文’)
Google 搜索发现，原因是 Linux 服务器上 UCS-2 编码方式与 Winodws 不一致。
于是，我改成 iconv(‘UCS-2BE’, ‘GBK’, ‘中文’) 试试，中文正常了

以下是有关两个平台 UCS-2 编码的潜规则：

1, UCS-2 不等于 UTF-16。 UTF-16 每个字节使用 ASCII 字符范围编码，而 UCS-2 对每个字节的编码可以超出 ASCII 字符范围。UCS-2 和 UTF-16 对每个字符至多占两个字节，但是他们的编码是不一样的。

2, 对于 UCS-2, windows 下默认是 UCS-2LE。用 MultibyteToWidechar（或者A2W）生成的是 UCS-2LE 的 unicode。windows记事本可以将文本保存为 UCS-2BE，相当于多了层转换。

3, 对于 UCS-2, linux 下默认是 UCS-2BE。用iconv(指定UCS-2)来转换生成的是 UCS-2BE 的 unicode。如果转换windows平台过来的 UCS-2, 需要指定 UCS-2LE。

4, 鉴于windows和linux等多个平台对 UCS-2 的理解不同（UCS-2LE,UCS-2BE）。MS 主张 unicode 有个引导标志(UCS-2LE FFFE, UCS-2BE FEFF)，以表明下面的字符是 unicode 并且判别 big-endian 或 little-endian。 所以从 windows 平台过来的数据发现有这个前缀，不用慌张。

5, linux 的编码输出，比如从文件输出，从 printf 输出，需要控制台做适当的编码匹配（如果编码不匹配，一般和该程序编译时的编码有若干关系），而控制台的转换输入需要查看当前的系统编码。比如控制台当前的编码是 UTF-8, 那么 UTF-8 编码的东西能正确显示，GBK 就不能；同样，当前编码是 GBK, 就能显示 GBK 编码，后来的系统应该更智能的处理好更多的转换了。不过通过 putty 等终端还是需要设置好终端的编码转换以解除乱码的烦恼。

PHP中对汉字进行UNICODE编码和解码的实现

<code><span>//将内容进行UNICODE编码</span><span><span>function</span><span>unicode_encode</span><span>(<span>$name</span>)</span> {</span><span>$name</span> = iconv(<span>'UTF-8'</span>, <span>'UCS-2'</span>, <span>$name</span>); <span>$len</span> = strlen(<span>$name</span>); <span>$str</span> = <span>''</span>; <span>for</span> (<span>$i</span> = <span>0</span>; <span>$i</span> < <span>$len</span> - <span>1</span>; <span>$i</span> = <span>$i</span> + <span>2</span>)
    { <span>$c</span> = <span>$name</span>[<span>$i</span>]; <span>$c2</span> = <span>$name</span>[<span>$i</span> + <span>1</span>]; <span>if</span> (ord(<span>$c</span>) > <span>0</span>)
        { <span>// 两个字节的文字</span><span>$str</span> .= <span>'\u'</span>.base_convert(ord(<span>$c</span>), <span>10</span>, <span>16</span>).base_convert(ord(<span>$c2</span>), <span>10</span>, <span>16</span>);
        } <span>else</span> { <span>$str</span> .= <span>$c2</span>;
        }
    } <span>return</span><span>$str</span>;
} <span>$name</span> = <span>'MY,你大爷的'</span>; <span>$unicode_name</span>=unicode_encode(<span>$name</span>); <span>echo</span><span>'<h3>'</span>.<span>$unicode_name</span>.<span>'</h3>'</span>; <span>// 将UNICODE编码后的内容进行解码</span><span><span>function</span><span>unicode_decode</span><span>(<span>$name</span>)</span> {</span><span>// 转换编码，将Unicode编码转换成可以浏览的utf-8编码</span><span>$pattern</span> = <span>'/([\w]+)|(\\\u([\w]{4}))/i'</span>;
    preg_match_all(<span>$pattern</span>, <span>$name</span>, <span>$matches</span>); <span>if</span> (!<span>empty</span>(<span>$matches</span>))
    { <span>$name</span> = <span>''</span>; <span>for</span> (<span>$j</span> = <span>0</span>; <span>$j</span> < count(<span>$matches</span>[<span>0</span>]); <span>$j</span>++)
        { <span>$str</span> = <span>$matches</span>[<span>0</span>][<span>$j</span>]; <span>if</span> (strpos(<span>$str</span>, <span>'\\u'</span>) === <span>0</span>)
            { <span>$code</span> = base_convert(substr(<span>$str</span>, <span>2</span>, <span>2</span>), <span>16</span>, <span>10</span>); <span>$code2</span> = base_convert(substr(<span>$str</span>, <span>4</span>), <span>16</span>, <span>10</span>); <span>$c</span> = chr(<span>$code</span>).chr(<span>$code2</span>); <span>$c</span> = iconv(<span>'UCS-2'</span>, <span>'UTF-8'</span>, <span>$c</span>); <span>$name</span> .= <span>$c</span>;
            } <span>else</span> { <span>$name</span> .= <span>$str</span>;
            }
        }
    } <span>return</span><span>$name</span>;
} <span>echo</span><span>'MY,\u4f60\u5927\u7237\u7684 -> '.unicode_decode(<span>$unicode_name</span>);</code>

以上就介绍了用PHP将Unicode 转化为UTF-8，包括了方面的内容，希望对PHP教程有兴趣的朋友有所帮助。