.NET 的 IMAP 文件夹路径编码 (IMAP UTF-7)？

Question

IMAP 规范（RFC 2060，5.1.3. 邮箱国际命名约定）描述了如何处理文件夹名称中的非 ASCII 字符。它定义了一个修改的 UTF-7 编码：

按照惯例，国际邮箱 名称是使用指定的 UTF-7 编码的修改版本 在 [UTF-7] 中描述。目的 这些修改是为了纠正 UTF-7 存在以下问题：

1. UTF-7 使用“+”字符进行移位；这与 邮箱名称中“+”的常见用法，尤其是 USENET 新闻组名称。

2. UTF-7 的编码是使用“/”字符的 BASE64；这 与使用“/”作为流行的层次分隔符冲突。

3. UTF-7 禁止“\”的未编码使用；这与 使用“\”作为流行的层次分隔符。

4. UTF-7 禁止“~”的未编码使用；这与 在某些服务器中使用“~”作为主目录指示符。

5. UTF-7 允许多种替代形式来表示相同的 细绳;特别是，可打印的 US-ASCII 字符可以 以编码形式表示。

在修改后的 UTF-7 中，可打印的 US-ASCII 字符（“&”除外）代表它们自己； 也就是说，八位字节值为 0x20-0x25 的字符 和 0x27-0x7e。性格 ”＆” (0x26) 由两个八位字节序列“&-”表示。

所有其他字符（八位字节值 0x00-0x1f、0x7f-0xff 和所有 Unicode 16 位八位字节）表示 在修改后的 BASE64 中，进一步 从 [UTF-7] 修改“，”是 用于代替“/”。
修改后的 BASE64 不得用于表示 任何打印的 US-ASCII 字符 可以代表自己。

"&" 用于转换为已修改 BASE64 和“-”转换回 US-ASCII。所有名称均以 US-ASCII 开头， 并且必须以 US-ASCII 结尾（即， 以 Unicode 16 位结尾的名称 八位字节必须以“-”结尾）。

在我开始实现它之前，我的问题是：是否有一些 .NET 代码/库（甚至在框架中）可以完成这项工作？我找不到 .NET 资源（仅限 implementations for other languages/frameworks）。

谢谢！

Answer 1

//
// ImapEncoding.cs
//
// Author: Jeffrey Stedfast <jestedfa@microsoft.com>
//
// Copyright (c) 2013-2019 Microsoft Corp. (www.microsoft.com)
//
// Permission is hereby granted, free of charge, to any person obtaining a copy
// of this software and associated documentation files (the "Software"), to deal
// in the Software without restriction, including without limitation the rights
// to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
// copies of the Software, and to permit persons to whom the Software is
// furnished to do so, subject to the following conditions:
//
// The above copyright notice and this permission notice shall be included in
// all copies or substantial portions of the Software.
//
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
// FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
// AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
// LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
// OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
// THE SOFTWARE.
//

using System.Text;

namespace MailKit.Net.Imap {
    static class ImapEncoding
    {
        const string utf7_alphabet = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+,";

        static readonly byte[] utf7_rank = {
            255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
            255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,
            255,255,255,255,255,255,255,255,255,255,255, 62, 63,255,255,255,
             52, 53, 54, 55, 56, 57, 58, 59, 60, 61,255,255,255,255,255,255,
            255,  0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14,
             15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25,255,255,255,255,255,
            255, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
             41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51,255,255,255,255,255,
        };

        public static string Decode (string text)
        {
            var decoded = new StringBuilder ();
            bool shifted = false;
            int bits = 0, v = 0;
            int index = 0;
            char c;

            while (index < text.Length) {
                c = text[index++];

                if (shifted) {
                    if (c == '-') {
                        // shifted back out of modified UTF-7
                        shifted = false;
                        bits = v = 0;
                    } else if (c > 127) {
                        // invalid UTF-7
                        return text;
                    } else {
                        byte rank = utf7_rank[(byte) c];

                        if (rank == 0xff) {
                            // invalid UTF-7
                            return text;
                        }

                        v = (v << 6) | rank;
                        bits += 6;

                        if (bits >= 16) {
                            char u = (char) ((v >> (bits - 16)) & 0xffff);
                            decoded.Append (u);
                            bits -= 16;
                        }
                    }
                } else if (c == '&' && index < text.Length) {
                    if (text[index] == '-') {
                        decoded.Append ('&');
                        index++;
                    } else {
                        // shifted into modified UTF-7
                        shifted = true;
                    }
                } else {
                    decoded.Append (c);
                }
            }

            return decoded.ToString ();
        }

        static void Utf7ShiftOut (StringBuilder output, int u, int bits)
        {
            if (bits > 0) {
                int x = (u << (6 - bits)) & 0x3f;
                output.Append (utf7_alphabet[x]);
            }

            output.Append ('-');
        }

        public static string Encode (string text)
        {
            var encoded = new StringBuilder ();
            bool shifted = false;
            int bits = 0, u = 0;

            for (int index = 0; index < text.Length; index++) {
                char c = text[index];

                if (c >= 0x20 && c < 0x7f) {
                    // characters with octet values 0x20-0x25 and 0x27-0x7e
                    // represent themselves while 0x26 ("&") is represented
                    // by the two-octet sequence "&-"

                    if (shifted) {
                        Utf7ShiftOut (encoded, u, bits);
                        shifted = false;
                        bits = 0;
                    }

                    if (c == 0x26)
                        encoded.Append ("&-");
                    else
                        encoded.Append (c);
                } else {
                    // base64 encode
                    if (!shifted) {
                        encoded.Append ('&');
                        shifted = true;
                    }

                    u = (u << 16) | (c & 0xffff);
                    bits += 16;

                    while (bits >= 6) {
                        int x = (u >> (bits - 6)) & 0x3f;
                        encoded.Append (utf7_alphabet[x]);
                        bits -= 6;
                    }
                }
            }

            if (shifted)
                Utf7ShiftOut (encoded, u, bits);

            return encoded.ToString ();
        }
    }
}

Answer 2

这太专业了，无法出现在框架中。尽管我见过的许多不完整的“实现”根本不关心转换，但 codeplex 上可能有一些东西，并且很乐意将所有非 us-ascii 字符传递到 IMAP 服务器。

不过，我过去已经实现了这一点，它实际上只有 30 行代码。您遍历字符串中的所有字符，如果它们在 0x20 和 0x7e 之间的范围内输出它们（不要忘记在“&”之后附加“-”），否则收集所有非 us-ascii 并使用 UTF7 转换它们（或UTF8 + base64，我不太确定）用“，”替换“/”。此外，您需要保持“转移状态”，例如无论您当前是在编码非 us-ascii 还是输出 us-ascii 并在状态更改时附加转换标记“&”和“-”。

Answer 3

未经测试，但this MIT 许可的代码看起来不错，如果应用了Alekseys 错误修复：

    /// <summary>
    /// Takes a UTF-16 encoded string and encodes it as modified UTF-7.
    /// </summary>
    /// <param name="s">The string to encode.</param>
    /// <returns>A UTF-7 encoded string</returns>
    /// <remarks>IMAP uses a modified version of UTF-7 for encoding international mailbox names. For
    /// details, refer to RFC 3501 section 5.1.3 (Mailbox International Naming Convention).</remarks>
    internal static string UTF7Encode(string s) {
        StringReader reader = new StringReader(s);
        StringBuilder builder = new StringBuilder();
        while (reader.Peek() != -1) {
            char c = (char)reader.Read();
            int codepoint = Convert.ToInt32(c);
            // It's a printable ASCII character.
            if (codepoint > 0x1F && codepoint < 0x7F) {
                builder.Append(c == '&' ? "&-" : c.ToString());
            } else {
                // The character sequence needs to be encoded.
                StringBuilder sequence = new StringBuilder(c.ToString());
                while (reader.Peek() != -1) {
                    codepoint = Convert.ToInt32((char)reader.Peek());
                    if (codepoint > 0x1F && codepoint < 0x7F)
                        break;
                    sequence.Append((char)reader.Read());
                }
                byte[] buffer = Encoding.BigEndianUnicode.GetBytes(
                    sequence.ToString());
                string encoded = Convert.ToBase64String(buffer).Replace('/', ',').
                    TrimEnd('=');
                builder.Append("&" + encoded + "-");
            }
        }
        return builder.ToString();
    }

    /// <summary>
    /// Takes a modified UTF-7 encoded string and decodes it.
    /// </summary>
    /// <param name="s">The UTF-7 encoded string to decode.</param>
    /// <returns>A UTF-16 encoded "standard" C# string</returns>
    /// <exception cref="FormatException">The input string is not a properly UTF-7 encoded
    /// string.</exception>
    /// <remarks>IMAP uses a modified version of UTF-7 for encoding international mailbox names. For
    /// details, refer to RFC 3501 section 5.1.3 (Mailbox International Naming Convention).</remarks>
    internal static string UTF7Decode(string s) {
        StringReader reader = new StringReader(s);
        StringBuilder builder = new StringBuilder();
        while (reader.Peek() != -1) {
            char c = (char)reader.Read();
            if (c == '&' && reader.Peek() != '-') {
                // The character sequence needs to be decoded.
                StringBuilder sequence = new StringBuilder();
                while (reader.Peek() != -1) {
                    if ((c = (char)reader.Read()) == '-')
                        break;
                    sequence.Append(c);
                }
                string encoded = sequence.ToString().Replace(',', '/');
                int pad = encoded.Length % 4;
                if (pad > 0)
                    encoded = encoded.PadRight(encoded.Length + (4 - pad), '=');
                try {
                    byte[] buffer = Convert.FromBase64String(encoded);
                    builder.Append(Encoding.BigEndianUnicode.GetString(buffer));
                } catch (Exception e) {
                    throw new FormatException(
                        "The input string is not in the correct Format.", e);
                }
            } else {
                if (c == '&' && reader.Peek() == '-')
                    reader.Read();
                builder.Append(c);
            }
        }
        return builder.ToString();
    }

---

不要在当前状态下使用this 代码，它包含[...] UTF7.GetBytes([...]) [...] .Replace('+', '&') - 它使用现有的.Net UTF-7 编码例程并且（除其他外）在结果中将+ 替换为& .这是错误，因为它不仅将“移位字符”从+ 更改为&（这是有意且正确的），而且还更改了base64 编码区域内的所有+ 字符（即不得更改为&)。