正则验证函数异常：粘贴字符串验证失败，需移除非法字符

你的需求很明确：当前方法只是判断输入是否完全符合正则规则，不符合就返回false，但实际需要的是自动移除不符合规则的字符，保留合法内容，尤其是处理粘贴过来的包含特殊字符的文本时。

原方法的核心问题 #

你的skipMetadataSpecChar函数用val.match(...)做全量匹配校验——只有当整个字符串的每一个字符都在允许的范围内时才返回true，一旦出现非法字符（比如测试字符串里的ﻁﻁﻉﻁﻉﻁﻉ这类非ASCII字符），就直接返回false，完全没有处理非法字符的逻辑。

修改后的实现方案 #

我们把逻辑从「校验」改成「过滤」：用正则匹配所有不允许的字符，然后替换为空字符串，直接返回过滤后的合法内容。

function filterInvalidChars(val) {
  // 匹配所有不在允许列表中的字符，全局替换为空
  return val.replace(/[^`_,:\\/="';%@#!~-&+^[\]{}$()<>|*?_\.,\s0-9a-zA-Z]/g, '');
}

代码细节说明：

1. 正则规则调整：把原正则的^[...]（要求整个字符串由合法字符组成）改成[^...]（匹配所有不在合法列表里的字符）
2. 添加g修饰符：确保替换所有匹配到的非法字符，而不是只替换第一个
3. 简化转义写法：原正则里的\057（斜杠/）、\134（反斜杠\\）、\075（等号=）都换成了更易读的直接转义写法

测试你的示例文本 #

把你提供的测试字符串传入这个函数后，那些ﻁﻁﻉﻁﻉﻁﻉ这类非法字符会被自动移除，剩下的合法内容会完整保留，比如处理后会得到：

This is test comment for the length of text as there was a issue in prod due to comment length in the box.This is test comment for the length of text as there was a issue in prod due to comment length in the box.This is test comment for the length of text as there was a issue in prod due to comment length in the box.This is test comment for the length of text as there was a issue in prod due to comment length in the box.This is test comment for the length of text as there was a issue in prod due to comment length in the box.This is test comment for the length of text as there was a issue in prod due to comment length in the box.This is test comment for the length of text as there was a issue in prod due to comment length in the box.This is test comment for the length of text as there was a issue in prod due to comment length in the box.;%@#This is test comment for the length of text as there was a issue in prod due to comment length in the box. This is test data to vhe issue an$

额外扩展：保留校验功能 #

如果你还需要判断输入是否包含非法字符，可以在过滤后对比原字符串和结果：

function hasInvalidChars(val) {
  return val !== filterInvalidChars(val);
}

这样就能同时满足「自动过滤非法字符」和「校验是否存在非法字符」的双重需求了。

内容的提问来源于stack exchange，提问作者Prashant Chaudhary