Ideally these would go in a class called "FileNaming" or maybe just rename Encode to "FileNameEncode". Note: these are not designed to handle Full Paths, just the folder and/or file names. Ideally you would Split("/") your full path first and then check the pieces. And obviously instead of a union, you could just add the "%" character to the list of chars not allowed in Windows, but I think it's more helpful/readable/factual this way. Decode() is exactly the same but switches the Replace(Uri.HexEscape(s[0]), s) "escaped" with the character.

public static List<string> urlEncodedCharacters = new List<string>
{
  "/", "\\", "<", ">", ":", "\"", "|", "?", "%" //and others, but not *
};
//Since this is a superset of urlEncodedCharacters, we won't be able to only use UrlEncode() - instead we'll use HexEncode
public static List<string> specialCharactersNotAllowedInWindows = new List<string>
{
  "/", "\\", "<", ">", ":", "\"", "|", "?", "*" //windows dissallowed character set
};

    public static string Encode(string fileName)
    {
        //CheckForFullPath(fileName); // optional: make sure it's not a path?
        List<string> charactersToChange = new List<string>(specialCharactersNotAllowedInWindows);
        charactersToChange.AddRange(urlEncodedCharacters.
            Where(x => !urlEncodedCharacters.Union(specialCharactersNotAllowedInWindows).Contains(x)));   // add any non duplicates (%)

        charactersToChange.ForEach(s => fileName = fileName.Replace(s, Uri.HexEscape(s[0])));   // "?" => "%3f"

        return fileName;
    }

感谢@simon-tewsi提供的非常有用的表格!

您应该只对用户名或URL中可能无效的其他部分进行编码。URL编码URL可能会导致这样的问题:

string url = HttpUtility.UrlEncode("http://www.google.com/search?q=Example");

将产生

http 3a % 2f 2fwww。google。com % 2fsearch 3fq % 3dExample

这显然不会有很好的效果。相反，你应该在查询字符串中只编码键/值对的值，就像这样:

string url = "http://www.google.com/search?q=" + HttpUtility.UrlEncode("Example");

希望这有帮助。另外，正如teedyay提到的，您仍然需要确保非法文件名字符被删除，否则文件系统将不喜欢该路径。

编辑:注意，这个答案现在已经过时了。请参阅下面Siarhei Kuchuk的答案以获得更好的解决方案

UrlEncoding会做你在这里建议的事情。对于c#，您只需使用HttpUtility，如前所述。

您还可以Regex非法字符，然后进行替换，但这变得更加复杂，因为您必须有某种形式的状态机(switch…用正确的字符替换。由于UrlEncode预先完成了这个操作，所以相当简单。

至于Linux和windows，有一些字符在Linux中是可以接受的，但在windows中是不可以接受的，但我并不担心这一点，因为文件夹名可以通过解码Url字符串返回，使用UrlDecode，因此您可以往返更改。

对于。net核心用户，使用这个

Microsoft.AspNetCore.Http.Extensions.UriHelper.Encode(Uri uri)

更好的方法是使用

Uri。EscapeUriString

不引用。net 4的完整配置文件。

(更新)

根据OP的要求，推荐的API应该是

Uri。EscapeDataString

(谢谢@ykadaru)

Ideally these would go in a class called "FileNaming" or maybe just rename Encode to "FileNameEncode". Note: these are not designed to handle Full Paths, just the folder and/or file names. Ideally you would Split("/") your full path first and then check the pieces. And obviously instead of a union, you could just add the "%" character to the list of chars not allowed in Windows, but I think it's more helpful/readable/factual this way. Decode() is exactly the same but switches the Replace(Uri.HexEscape(s[0]), s) "escaped" with the character.

public static List<string> urlEncodedCharacters = new List<string>
{
  "/", "\\", "<", ">", ":", "\"", "|", "?", "%" //and others, but not *
};
//Since this is a superset of urlEncodedCharacters, we won't be able to only use UrlEncode() - instead we'll use HexEncode
public static List<string> specialCharactersNotAllowedInWindows = new List<string>
{
  "/", "\\", "<", ">", ":", "\"", "|", "?", "*" //windows dissallowed character set
};

    public static string Encode(string fileName)
    {
        //CheckForFullPath(fileName); // optional: make sure it's not a path?
        List<string> charactersToChange = new List<string>(specialCharactersNotAllowedInWindows);
        charactersToChange.AddRange(urlEncodedCharacters.
            Where(x => !urlEncodedCharacters.Union(specialCharactersNotAllowedInWindows).Contains(x)));   // add any non duplicates (%)

        charactersToChange.ForEach(s => fileName = fileName.Replace(s, Uri.HexEscape(s[0])));   // "?" => "%3f"

        return fileName;
    }

感谢@simon-tewsi提供的非常有用的表格!