带有长非ascii名称的电子邮件附件

Question

我尝试发送System.Net.Mail.MailMessage带System.Net.Mail.Attachment.

附件名称是"Счёт-договор№4321от4июля.pdf"

附件创建代码:

var nameEncoding = Encoding.UTF8;
return new System.Net.Mail.Attachment(new MemoryStream(bytes), 
                                      MessageBodiesHelpers.EncodeAttachmentName(fileName, nameEncoding),
                                      attachment.MediaType)
       {
            TransferEncoding = TransferEncoding.Base64,
            NameEncoding = nameEncoding
       };

代码内容MessageBodiesHelpers.EncodeAttachmentName取自https://social.msdn.microsoft.com/Forums/en-US/b6c764f7-4697-4394-b45f-128a24306d55/40-smtpclientsend-attachments-mit-umlauten-im-dateinamen?forum=dotnetframeworkde

如果我将该附件发送到gmail或ms exchange,则附件名称已成功解码.但!

如果我将附件发送到icloud,那么我会得到"???? - ???????№4321от4????.pdf"

邮件附件标题:

来自ms exchange:

Content-Type: application/pdf;
    name="=?utf-8?B?0KHRh9GR0YIt0LTQ?==?utf-8?B?vtCz0L7QstC+0YAg?==?utf-8?B?4oSWNDMyMSDQvtGC?==?utf-8?B?IDQg0LjRjtC70Y8u?==?utf-8?B?cGRm?="
Content-Transfer-Encoding: base64
Content-Disposition: attachment

来自icloud:

Content-Transfer-Encoding: BASE64
Content-Type: APPLICATION/PDF;
    name="????-??????? =?utf-8?B?4oSWNDMyMSDQvtGC?= 4 ????.pdf"

如何格式化icloud的名称？

UPD

如果我将消息从outlook(ms exchange)转发到icloud,则附件名称已成功解码.头:

Content-Transfer-Encoding: BASE64
Content-Disposition: ATTACHMENT;
    size=200702;
    modification-date="Mon, 04 Jul 2016 13:40:22 GMT";
    filename*=utf-8''%D0%A1%D1%87%D1%91%D1%82%2D%D0%B4%D0%BE%D0%B3%D0%BE%D0%B2%D0%BE%D1%80%20%E2%84%964321%20%D0%BE%D1%82%204%20%D0%B8%D1%8E%D0%BB%D1%8F.pdf;
    creation-date="Mon, 04 Jul 2016 13:40:22 GMT"
Content-Type: APPLICATION/PDF;
    name*=utf-8''%D0%A1%D1%87%D1%91%D1%82%2D%D0%B4%D0%BE%D0%B3%D0%BE%D0%B2%D0%BE%D1%80%20%E2%84%964321%20%D0%BE%D1%82%204%20%D0%B8%D1%8E%D0%BB%D1%8F.pdf

UPD2

如果我使用icloud(icloud.com)的web界面阅读消息,则附件名称已成功解码.

Answer 1

在 .NET 4.0 SmtpClient 现在实现了 RFC 行长度限制（76 个字符）。这需要对编码过程进行大量更改，并且没有涵盖您描述的一些问题。

在您的情况下，非 ascii 附件名称存在问题，该名称将被编码为超过 41 个 utf-8 字节 (Encoding.UTF8.GetByteCount(fileName);)。在这种情况下，名称被编码两次并且其中可能有额外的换行符。唯一已知的解决方法是限制非 ASCII 文件名的长度。

您可以阅读这篇文章以了解有关您的问题的更多信息