Question

我正在尝试使用preg_replace替换除chars nums和unicode chars之外的所有内容。这是我尝试过并遇到的错误：

preg_replace（）：编译失败：字符类在偏移量22处的排列范围混乱

这是我的正则表达式：

[^A-Za-z0-9 \x{0080}-\x{FFFF}]

我想将文本转换为以下示例：

CAFÉ? CREATORS WERE HERE!#1 => CAFÉ CREATORS WERE HERE1

-编辑-

我尝试了以下解决方案，并收到此错误：

$str = 'CAFÉ? CREATORS WERE HERE!#1';


$alphaNumStr = preg_replace('/[^A-Za-z0-9 x{0080}-x{FFFF}]/u', '', $str);

echo 'TEXT: ' . $alphaNumStr;

TEXT：preg_replace（）：编译失败：字符类的范围乱序，第4行的偏移量20

Answer 1

您需要将u（unicode）标志添加到您的正则表达式中：

$text = 'CAFÉ? CREATORS WERE HERE!#1';
echo preg_replace('/[^A-Za-z0-9 \x{0080}-\x{FFFF}]/u', '', $text);

输出

CAFÉ CREATORS WERE HERE1

Answer 2

如果要保留所有语言的所有字母，请使用：

$str = 'CAFÉ? CREATORS WERE HERE!#1';
echo preg_replace('/[^\p{L}\d\s]+/u', '', $str);

输出：

CAFÉ CREATORS WERE HERE1

\p{L}代表任何字母。