我喜欢这种不使用正则表达式的方式，所以我将其移植到PHP。我刚刚添加了一个名为is_between的函数来检查字符:

function is_between($val, $min, $max)
{
    $val = (int) $val; $min = (int) $min; $max = (int) $max;

    return ($val >= $min && $val <= $max);
}

function international_char_to_ascii($char)
{
    if (mb_strpos('àåáâäãåa', $char) !== false)
    {
        return 'a';
    }

    if (mb_strpos('èéêëe', $char) !== false)
    {
        return 'e';
    }

    if (mb_strpos('ìíîïi', $char) !== false)
    {
        return 'i';
    }

    if (mb_strpos('òóôõö', $char) !== false)
    {
        return 'o';
    }

    if (mb_strpos('ùúûüuu', $char) !== false)
    {
        return 'u';
    }

    if (mb_strpos('çccc', $char) !== false)
    {
        return 'c';
    }

    if (mb_strpos('zzž', $char) !== false)
    {
        return 'z';
    }

    if (mb_strpos('ssšs', $char) !== false)
    {
        return 's';
    }

    if (mb_strpos('ñn', $char) !== false)
    {
        return 'n';
    }

    if (mb_strpos('ýÿ', $char) !== false)
    {
        return 'y';
    }

    if (mb_strpos('gg', $char) !== false)
    {
        return 'g';
    }

    if (mb_strpos('r', $char) !== false)
    {
        return 'r';
    }

    if (mb_strpos('l', $char) !== false)
    {
        return 'l';
    }

    if (mb_strpos('d', $char) !== false)
    {
        return 'd';
    }

    if (mb_strpos('ß', $char) !== false)
    {
        return 'ss';
    }

    if (mb_strpos('Þ', $char) !== false)
    {
        return 'th';
    }

    if (mb_strpos('h', $char) !== false)
    {
        return 'h';
    }

    if (mb_strpos('j', $char) !== false)
    {
        return 'j';
    }
    return '';
}

function url_friendly_title($url_title)
{
    if (empty($url_title))
    {
        return '';
    }

    $url_title = mb_strtolower($url_title);

    $url_title_max_length   = 80;
    $url_title_length       = mb_strlen($url_title);
    $url_title_friendly     = '';
    $url_title_dash_added   = false;
    $url_title_char = '';

    for ($i = 0; $i < $url_title_length; $i++)
    {
        $url_title_char     = mb_substr($url_title, $i, 1);

        if (strlen($url_title_char) == 2)
        {
            $url_title_ascii    = ord($url_title_char[0]) * 256 + ord($url_title_char[1]) . "\r\n";
        }
        else
        {
            $url_title_ascii    = ord($url_title_char);
        }

        if (is_between($url_title_ascii, 97, 122) || is_between($url_title_ascii, 48, 57))
        {
            $url_title_friendly .= $url_title_char;

            $url_title_dash_added = false;
        }
        elseif(is_between($url_title_ascii, 65, 90))
        {
            $url_title_friendly .= chr(($url_title_ascii | 32));

            $url_title_dash_added = false;
        }
        elseif($url_title_ascii == 32 || $url_title_ascii == 44 || $url_title_ascii == 46 || $url_title_ascii == 47 || $url_title_ascii == 92 || $url_title_ascii == 45 || $url_title_ascii == 47 || $url_title_ascii == 95 || $url_title_ascii == 61)
        {
            if (!$url_title_dash_added && mb_strlen($url_title_friendly) > 0)
            {
                $url_title_friendly .= chr(45);

                $url_title_dash_added = true;
            }
        }
        else if ($url_title_ascii >= 128)
        {
            $url_title_previous_length = mb_strlen($url_title_friendly);

            $url_title_friendly .= international_char_to_ascii($url_title_char);

            if ($url_title_previous_length != mb_strlen($url_title_friendly))
            {
                $url_title_dash_added = false;
            }
        }

        if ($i == $url_title_max_length)
        {
            break;
        }
    }

    if ($url_title_dash_added)
    {
        return mb_substr($url_title_friendly, 0, -1);
    }
    else
    {
        return $url_title_friendly;
    }
}

重写Jeff的代码，使其更简洁

    public static string RemapInternationalCharToAscii(char c)
    {
        var s = c.ToString().ToLowerInvariant();

        var mappings = new Dictionary<string, string>
        {
            { "a", "àåáâäãåą" },
            { "c", "çćčĉ" },
            { "d", "đ" },
            { "e", "èéêëę" },
            { "g", "ğĝ" },
            { "h", "ĥ" },
            { "i", "ìíîïı" },
            { "j", "ĵ" },
            { "l", "ł" },
            { "n", "ñń" },
            { "o", "òóôõöøőð" },
            { "r", "ř" },
            { "s", "śşšŝ" },
            { "ss", "ß" },
            { "th", "Þ" },
            { "u", "ùúûüŭů" },
            { "y", "ýÿ" },
            { "z", "żźž" }
        };

        foreach(var mapping in mappings)
        {
            if (mapping.Value.Contains(s))
                return mapping.Key;
        }

        return string.Empty;
    }

有一个叫PermalinkFu的小Ruby on Rails插件可以做到这一点。escape方法将转换为适合URL的字符串。看一下代码;那个方法很简单。

为了删除非ascii字符，它使用iconv lib将'utf-8'转换为'ascii//ignore//translit'。然后空格变成破折号，所有东西都被缩小，等等。

Brian的Ruby代码:

title.downcase.strip.gsub(/\ /, '-').gsub(/[^\w\-]/, '')

Downcase将字符串转换为小写，strip删除开头和结尾的空格，第一个gsub调用全局地用破折号替换空格，第二个调用删除所有不是字母或破折号的内容。

我知道这是一个非常老的问题，但由于大多数浏览器现在都支持unicode url，我在XRegex中找到了一个很好的解决方案，它可以转换除字母以外的所有内容(在所有语言中都是'-')。

这可以在几种编程语言中实现。

模式是\\p{^L}+，然后你只需要用它来替换所有非字母到'-'。

node.js中xregex模块的工作示例。

var text = 'This ! can @ have # several $ letters % from different languages such as עברית or Español';

var slugRegEx = XRegExp('((?!\\d)\\p{^L})+', 'g');

var slug = XRegExp.replace(text, slugRegEx, '-').toLowerCase();

console.log(slug) ==> "this-can-have-several-letters-from-different-languages-such-as-עברית-or-español"

下面是我的(稍慢，但写起来很有趣)Jeff的代码版本:

public static string URLFriendly(string title)
{
    char? prevRead = null,
        prevWritten = null;

    var seq = 
        from c in title
        let norm = RemapInternationalCharToAscii(char.ToLowerInvariant(c).ToString())[0]
        let keep = char.IsLetterOrDigit(norm)
        where prevRead.HasValue || keep
        let replaced = keep ? norm
            :  prevWritten != '-' ? '-'
            :  (char?)null
        where replaced != null
        let s = replaced + (prevRead == null ? ""
            : norm == '#' && "cf".Contains(prevRead.Value) ? "sharp"
            : norm == '+' ? "plus"
            : "")
        let _ = prevRead = norm
        from written in s
        let __ = prevWritten = written
        select written;

    const int maxlen = 80;  
    return string.Concat(seq.Take(maxlen)).TrimEnd('-');
}

public static string RemapInternationalCharToAscii(string text)
{
    var seq = text.Normalize(NormalizationForm.FormD)
        .Where(c => CharUnicodeInfo.GetUnicodeCategory(c) != UnicodeCategory.NonSpacingMark);

    return string.Concat(seq).Normalize(NormalizationForm.FormC);
}

我的测试字符串:

“我喜欢c#， f#， c++，还有……焦糖布丁! !他们看到我在编程……他们hatin”……想抓住我的下流行为……”