PHP正则表达式不适用于数据库中的字符串

问题描述

preg_replace 当我在从数据库获取的字符串上使用它时不会返回所需的结果.

preg_replace does not return desired result when I use it on string fetched from database.

$result = DB::connection("connection")->select("my query");
foreach($result as $row){

    //prints run-d.m.c.
    print($row->artist . "\n");

    //should print run.d.m.c
    //prints run-d.m.c
    print(preg_replace("/-/", ".", $row->artist) . "\n");
}

这仅在我尝试替换 -(破折号)时发生.我可以替换任何其他字符.但是，如果我在简单的字符串上尝试这个正则表达式，它会按预期工作:

This occurs only when i try to replace - (dash). I can replace any other character.However if I try this regex on simple string it works as expected:

$str = "run-d.m.c";

//prints run.d.m.c
print(preg_replace("/-/", ".", $str) . "\n");

我在这里遗漏了什么?

推荐答案

事实证明您的字符串中有 Unicode 破折号.要匹配所有 Unicode 破折号，请使用

It turns out you have Unicode dashes in your strings. To match all Unicode dashes, use

/[\p{Pd}\xAD]/u

查看正则表达式演示

\p{Pd} 匹配 Unicode Character Category 'Punctuation, Dash' 但是软连字符 \xAD，因此它应该与 \p 结合{Pd} 在字符类中.

The \p{Pd} matches any hyphen in the Unicode Character Category 'Punctuation, Dash' but a soft hyphen, \xAD, hence it should be combined with \p{Pd} in a character class.

/u 修饰符使模式识别 Unicode，并使正则表达式引擎将输入字符串视为 Unicode 代码点序列，而不是字节序列.

The /u modifier makes the pattern Unicode aware and makes the regex engine treat the input string as Unicode code point sequence, not a byte sequence.

这篇关于PHP正则表达式不适用于数据库中的字符串的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！