public static Encoding GetEncoding(int codepage)
{
return(Encoding.GetEncoding(codepage));
}
internal static bool TryParseMailbox(byte[] text, int startIndex, ref int index, int endIndex, string name, int codepage, bool throwOnError, out InternetAddress address)
{
Encoding encoding = Encoding.GetEncoding(codepage);
DomainList route = null;
address = null;
// skip over the '<'
index++;
if (index >= endIndex)
{
if (throwOnError)
{
throw new ParseException(string.Format("Incomplete mailbox at offset {0}", startIndex), startIndex, index);
}
return(false);
}
if (text[index] == (byte)'@')
{
if (!DomainList.TryParse(text, ref index, endIndex, throwOnError, out route))
{
if (throwOnError)
{
throw new ParseException(string.Format("Invalid route in mailbox at offset {0}", startIndex), startIndex, index);
}
return(false);
}
if (index + 1 >= endIndex || text[index] != (byte)':')
{
if (throwOnError)
{
throw new ParseException(string.Format("Incomplete route in mailbox at offset {0}", startIndex), startIndex, index);
}
return(false);
}
index++;
}
string addrspec;
if (!TryParseAddrspec(text, ref index, endIndex, (byte)'>', throwOnError, out addrspec))
{
return(false);
}
if (!ParseUtils.SkipCommentsAndWhiteSpace(text, ref index, endIndex, throwOnError))
{
return(false);
}
if (index >= endIndex || text[index] != (byte)'>')
{
if (throwOnError)
{
throw new ParseException(string.Format("Unexpected end of mailbox at offset {0}", startIndex), startIndex, index);
}
return(false);
}
if (route != null)
{
address = new MailboxAddress(encoding, name, route, addrspec);
}
else
{
address = new MailboxAddress(encoding, name, addrspec);
}
index++;
return(true);
}
/// <summary>
/// Initializes a new instance of the <see cref="MimeKit.IO.Filters.CharsetFilter"/> class.
/// </summary>
/// <remarks>
/// Creates a new <see cref="CharsetFilter"/> to convert text from the specified
/// source encoding into the target charset encoding.
/// </remarks>
/// <param name="sourceCodePage">Source code page.</param>
/// <param name="targetCodePage">Target code page.</param>
/// <exception cref="System.ArgumentOutOfRangeException">
/// <para><paramref name="sourceCodePage"/> is less than zero or greater than 65535.</para>
/// <para>-or-</para>
/// <para><paramref name="targetCodePage"/> is less than zero or greater than 65535.</para>
/// </exception>
/// <exception cref="System.NotSupportedException">
/// <para>The <paramref name="sourceCodePage"/> is not supported by the system.</para>
/// <para>-or-</para>
/// <para>The <paramref name="targetCodePage"/> is not supported by the system.</para>
/// </exception>
public CharsetFilter(int sourceCodePage, int targetCodePage)
: this(Encoding.GetEncoding(sourceCodePage), Encoding.GetEncoding(targetCodePage))
{
}
static CharsetUtils()
{
int gb2312;
#if NETSTANDARD
Encoding.RegisterProvider(CodePagesEncodingProvider.Instance);
#endif
try {
Latin1 = Encoding.GetEncoding(28591, new EncoderExceptionFallback(), new DecoderExceptionFallback());
} catch (NotSupportedException) {
// Note: Some ASP.NET web hosts such as GoDaddy's Windows environment do not have
// iso-8859-1 support, they only have the built-in text encodings, so we need to
// hack around it by using an alternative encoding.
// Try to use Windows-1252 if it is available...
Latin1 = Encoding.GetEncoding(1252, new EncoderExceptionFallback(), new DecoderExceptionFallback());
}
// Note: Encoding.UTF8.GetString() replaces invalid bytes with a unicode '?' character,
// so we use our own UTF8 instance when using GetString() if we do not want it to do that.
UTF8 = Encoding.GetEncoding(65001, new EncoderExceptionFallback(), new DecoderExceptionFallback());
aliases = new Dictionary <string, int> (MimeUtils.OrdinalIgnoreCase);
AddAliases(aliases, 65001, -1, "utf-8", "utf8");
// ANSI_X3.4-1968 is used on some systems and should be
// treated the same as US-ASCII.
AddAliases(aliases, 20127, -1, "ansi_x3.4-1968");
// ANSI_X3.110-1983 is another odd-ball charset that appears
// every once in a while and seems closest to ISO-8859-1.
AddAliases(aliases, 28591, -1, "ansi_x3.110-1983", "latin1");
// Macintosh aliases
AddAliases(aliases, 10000, -1, "macintosh");
AddAliases(aliases, 10079, -1, "x-mac-icelandic");
// Korean charsets (aliases for euc-kr)
// 'upgrade' ks_c_5601-1987 to euc-kr since it is a superset
AddAliases(aliases, 51949, -1,
"ks_c_5601-1987",
"ksc-5601-1987",
"ksc-5601_1987",
"ksc-5601",
"5601",
"ks_c_5861-1992",
"ksc-5861-1992",
"ksc-5861_1992",
"euckr-0",
"euc-kr");
// Chinese charsets (aliases for big5)
AddAliases(aliases, 950, -1, "big5", "big5-0", "big5-hkscs", "big5.eten-0", "big5hkscs-0");
// Chinese charsets (aliases for gb2312)
gb2312 = AddAliases(aliases, 936, -1, "gb2312", "gb-2312", "gb2312-0", "gb2312-80", "gb2312.1980-0");
// Chinese charsets (euc-cn and gbk not supported on Mono)
// https://bugzilla.mozilla.org/show_bug.cgi?id=844082 seems to suggest falling back to gb2312.
AddAliases(aliases, 51936, gb2312, "euc-cn", "gbk-0", "x-gbk", "gbk");
// Chinese charsets (hz-gb-2312 not suported on Mono)
AddAliases(aliases, 52936, gb2312, "hz-gb-2312", "hz-gb2312");
// Chinese charsets (aliases for gb18030)
AddAliases(aliases, 54936, -1, "gb18030-0", "gb18030");
// Japanese charsets (aliases for euc-jp)
AddAliases(aliases, 51932, -1, "eucjp-0", "euc-jp", "ujis-0", "ujis");
// Japanese charsets (aliases for Shift_JIS)
AddAliases(aliases, 932, -1, "shift_jis", "jisx0208.1983-0", "jisx0212.1990-0", "pck");
// Note from http://msdn.microsoft.com/en-us/library/system.text.encoding.getencodings.aspx
// Encodings 50220 and 50222 are both associated with the name "iso-2022-jp", but they
// are not identical. Encoding 50220 converts half-width Katakana characters to
// full-width Katakana characters, whereas encoding 50222 uses a shift-in/shift-out
// sequence to encode half-width Katakana characters. The display name for encoding
// 50222 is "Japanese (JIS-Allow 1 byte Kana - SO/SI)" to distinguish it from encoding
// 50220, which has the display name "Japanese (JIS)".
//
// If your application requests the encoding name "iso-2022-jp", the .NET Framework
// returns encoding 50220. However, the encoding that is appropriate for your application
// will depend on the preferred treatment of the half-width Katakana characters.
AddAliases(aliases, 50220, -1, "iso-2022-jp");
}
static CharsetUtils()
{
aliases = new Dictionary <string, int> (StringComparer.OrdinalIgnoreCase);
aliases.Add("utf-8", 65001);
aliases.Add("utf8", 65001);
// ANSI_X3.4-1968 is used on some systems and should be
// treated the same as US-ASCII.
aliases.Add("ansi_x3.4-1968", 20127);
// ANSI_X3.110-1983 is another odd-ball charset that appears
// every once in a while and seems closest to ISO-8859-1.
aliases.Add("ansi_x3.110-1983", 28591);
// Macintosh aliases
aliases.Add("macintosh", 10000);
aliases.Add("x-mac-icelandic", 10079);
// Korean charsets (aliases for euc-kr)
// 'upgrade' ks_c_5601-1987 to euc-kr since it is a superset
aliases.Add("ks_c_5601-1987", 51949);
aliases.Add("5601", 51949);
aliases.Add("ksc-5601", 51949);
aliases.Add("ksc-5601-1987", 51949);
aliases.Add("ksc-5601_1987", 51949);
aliases.Add("ks_c_5861-1992", 51949);
aliases.Add("euckr-0", 51949);
aliases.Add("euc-kr", 51949);
// Chinese charsets (aliases for big5)
aliases.Add("big5", 950);
aliases.Add("big5-0", 950);
aliases.Add("big5-hkscs", 950);
aliases.Add("big5.eten-0", 950);
aliases.Add("big5hkscs-0", 950);
// Chinese charsets (aliases for gb2312)
aliases.Add("gb2312", 936);
aliases.Add("gb-2312", 936);
aliases.Add("gb2312-0", 936);
aliases.Add("gb2312-80", 936);
aliases.Add("gb2312.1980-0", 936);
// Chinese charsets (euc-cn and gbk not supported on Mono)
try {
Encoding.GetEncoding(51936);
aliases.Add("euc-cn", 51936);
aliases.Add("gbk-0", 51936);
aliases.Add("x-gbk", 51936);
aliases.Add("gbk", 51936);
} catch {
// https://bugzilla.mozilla.org/show_bug.cgi?id=844082 seems to suggest falling back to gb2312.
// fall back to gb2312
aliases.Add("euc-cn", 936);
aliases.Add("gbk-0", 936);
aliases.Add("x-gbk", 936);
aliases.Add("gbk", 936);
}
// Chinese charsets (hz-gb-2312 not suported on Mono)
try {
Encoding.GetEncoding(52936);
aliases.Add("hz-gb-2312", 52936);
aliases.Add("hz-gb2312", 52936);
} catch {
// fall back to gb2312
aliases.Add("hz-gb-2312", 936);
aliases.Add("hz-gb2312", 936);
}
// Chinese charsets (aliases for gb18030)
aliases.Add("gb18030-0", 54936);
aliases.Add("gb18030", 54936);
// Japanese charsets (aliases for euc-jp)
aliases.Add("eucjp-0", 51932);
aliases.Add("euc-jp", 51932);
aliases.Add("ujis-0", 51932);
aliases.Add("ujis", 51932);
// Japanese charsets (aliases for Shift_JIS)
aliases.Add("jisx0208.1983-0", 932);
aliases.Add("jisx0212.1990-0", 932);
aliases.Add("pck", 932);
// Note from http://msdn.microsoft.com/en-us/library/system.text.encoding.getencodings.aspx
// Encodings 50220 and 50222 are both associated with the name "iso-2022-jp", but they
// are not identical. Encoding 50220 converts half-width Katakana characters to
// full-width Katakana characters, whereas encoding 50222 uses a shift-in/shift-out
// sequence to encode half-width Katakana characters. The display name for encoding
// 50222 is "Japanese (JIS-Allow 1 byte Kana - SO/SI)" to distinguish it from encoding
// 50220, which has the display name "Japanese (JIS)".
//
// If your application requests the encoding name "iso-2022-jp", the .NET Framework
// returns encoding 50220. However, the encoding that is appropriate for your application
// will depend on the preferred treatment of the half-width Katakana characters.
}
static void ExtractMapiProperties(TnefReader reader, MimeMessage message, BodyBuilder builder)
{
var prop = reader.TnefPropertyReader;
var recipient = new EmailAddress();
var sender = new EmailAddress();
string normalizedSubject = null;
string subjectPrefix = null;
MailboxAddress mailbox;
var msgid = false;
while (prop.ReadNextProperty())
{
switch (prop.PropertyTag.Id)
{
case TnefPropertyId.InternetMessageId:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
message.MessageId = prop.ReadValueAsString();
msgid = true;
}
break;
case TnefPropertyId.TnefCorrelationKey:
// According to MSDN, PidTagTnefCorrelationKey is a unique key that is
// meant to be used to tie the TNEF attachment to the encapsulating
// message. It can be a string or a binary blob. It seems that most
// implementations use the Message-Id string, so if this property
// value looks like a Message-Id, then us it as one (unless we get a
// InternetMessageId property, in which case we use that instead.
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Binary)
{
if (!msgid)
{
var value = prop.ReadValueAsString();
if (value.Length > 5 && value[0] == '<' && value[value.Length - 1] == '>' && value.IndexOf('@') != -1)
{
message.MessageId = value;
}
}
}
break;
case TnefPropertyId.Subject:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
message.Subject = prop.ReadValueAsString();
}
break;
case TnefPropertyId.SubjectPrefix:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
subjectPrefix = prop.ReadValueAsString();
}
break;
case TnefPropertyId.NormalizedSubject:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
normalizedSubject = prop.ReadValueAsString();
}
break;
case TnefPropertyId.SenderName:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
sender.Name = prop.ReadValueAsString();
}
break;
case TnefPropertyId.SenderEmailAddress:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
sender.Addr = prop.ReadValueAsString();
}
break;
case TnefPropertyId.SenderSearchKey:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Binary)
{
sender.SearchKey = prop.ReadValueAsString();
}
break;
case TnefPropertyId.SenderAddrtype:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
sender.AddrType = prop.ReadValueAsString();
}
break;
case TnefPropertyId.ReceivedByName:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
recipient.Name = prop.ReadValueAsString();
}
break;
case TnefPropertyId.ReceivedByEmailAddress:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
recipient.Addr = prop.ReadValueAsString();
}
break;
case TnefPropertyId.ReceivedBySearchKey:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Binary)
{
recipient.SearchKey = prop.ReadValueAsString();
}
break;
case TnefPropertyId.ReceivedByAddrtype:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
recipient.AddrType = prop.ReadValueAsString();
}
break;
case TnefPropertyId.RtfCompressed:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Binary)
{
var rtf = new TextPart("rtf");
var converter = new RtfCompressedToRtf();
var content = new MemoryBlockStream();
using (var filtered = new FilteredStream(content)) {
filtered.Add(converter);
using (var compressed = prop.GetRawValueReadStream()) {
compressed.CopyTo(filtered, 4096);
filtered.Flush();
}
}
rtf.Content = new MimeContent(content);
content.Position = 0;
builder.Attachments.Add(rtf);
}
break;
case TnefPropertyId.BodyHtml:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Binary)
{
var html = new TextPart("html");
Encoding encoding;
if (prop.PropertyTag.ValueTnefType != TnefPropertyType.Unicode)
{
encoding = Encoding.GetEncoding(reader.MessageCodepage);
}
else
{
encoding = CharsetUtils.UTF8;
}
html.SetText(encoding, prop.ReadValueAsString());
builder.Attachments.Add(html);
}
break;
case TnefPropertyId.Body:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Binary)
{
var plain = new TextPart("plain");
Encoding encoding;
if (prop.PropertyTag.ValueTnefType != TnefPropertyType.Unicode)
{
encoding = Encoding.GetEncoding(reader.MessageCodepage);
}
else
{
encoding = CharsetUtils.UTF8;
}
plain.SetText(encoding, prop.ReadValueAsString());
builder.Attachments.Add(plain);
}
break;
case TnefPropertyId.Importance:
// https://msdn.microsoft.com/en-us/library/ee237166(v=exchg.80).aspx
switch (prop.ReadValueAsInt32())
{
case 2: message.Importance = MessageImportance.High; break;
case 1: message.Importance = MessageImportance.Normal; break;
case 0: message.Importance = MessageImportance.Low; break;
}
break;
case TnefPropertyId.Priority:
// https://msdn.microsoft.com/en-us/library/ee159473(v=exchg.80).aspx
switch (prop.ReadValueAsInt32())
{
case 1: message.Priority = MessagePriority.Urgent; break;
case 0: message.Priority = MessagePriority.Normal; break;
case -1: message.Priority = MessagePriority.NonUrgent; break;
}
break;
case TnefPropertyId.Sensitivity:
// https://msdn.microsoft.com/en-us/library/ee217353(v=exchg.80).aspx
// https://tools.ietf.org/html/rfc2156#section-5.3.4
switch (prop.ReadValueAsInt32())
{
case 1: message.Headers[HeaderId.Sensitivity] = "Personal"; break;
case 2: message.Headers[HeaderId.Sensitivity] = "Private"; break;
case 3: message.Headers[HeaderId.Sensitivity] = "Company-Confidential"; break;
case 0: message.Headers.Remove(HeaderId.Sensitivity); break;
}
break;
}
}
if (string.IsNullOrEmpty(message.Subject) && !string.IsNullOrEmpty(normalizedSubject))
{
if (!string.IsNullOrEmpty(subjectPrefix))
{
message.Subject = subjectPrefix + normalizedSubject;
}
else
{
message.Subject = normalizedSubject;
}
}
if (sender.TryGetMailboxAddress(out mailbox))
{
message.From.Add(mailbox);
}
if (recipient.TryGetMailboxAddress(out mailbox))
{
message.To.Add(mailbox);
}
}
static void ExtractMapiProperties(TnefReader reader, MimeMessage message, BodyBuilder builder)
{
var prop = reader.TnefPropertyReader;
while (prop.ReadNextProperty())
{
switch (prop.PropertyTag.Id)
{
case TnefPropertyId.InternetMessageId:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
message.MessageId = prop.ReadValueAsString();
}
break;
case TnefPropertyId.Subject:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode)
{
message.Subject = prop.ReadValueAsString();
}
break;
case TnefPropertyId.RtfCompressed:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Binary)
{
var rtf = new TextPart("rtf");
rtf.ContentType.Name = "body.rtf";
var converter = new RtfCompressedToRtf();
var content = new MemoryBlockStream();
using (var filtered = new FilteredStream(content)) {
filtered.Add(converter);
using (var compressed = prop.GetRawValueReadStream()) {
compressed.CopyTo(filtered, 4096);
filtered.Flush();
}
}
rtf.Content = new MimeContent(content);
content.Position = 0;
builder.Attachments.Add(rtf);
}
break;
case TnefPropertyId.BodyHtml:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Binary)
{
var html = new TextPart("html");
html.ContentType.Name = "body.html";
Encoding encoding;
if (prop.PropertyTag.ValueTnefType != TnefPropertyType.Unicode)
{
encoding = Encoding.GetEncoding(reader.MessageCodepage);
}
else
{
encoding = CharsetUtils.UTF8;
}
html.SetText(encoding, prop.ReadValueAsString());
builder.Attachments.Add(html);
}
break;
case TnefPropertyId.Body:
if (prop.PropertyTag.ValueTnefType == TnefPropertyType.String8 ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Unicode ||
prop.PropertyTag.ValueTnefType == TnefPropertyType.Binary)
{
var plain = new TextPart("plain");
plain.ContentType.Name = "body.txt";
Encoding encoding;
if (prop.PropertyTag.ValueTnefType != TnefPropertyType.Unicode)
{
encoding = Encoding.GetEncoding(reader.MessageCodepage);
}
else
{
encoding = CharsetUtils.UTF8;
}
plain.SetText(encoding, prop.ReadValueAsString());
builder.Attachments.Add(plain);
}
break;
case TnefPropertyId.Importance:
switch (prop.ReadValueAsInt32())
{
case 2: message.Importance = MessageImportance.High; break;
case 0: message.Importance = MessageImportance.Low; break;
}
break;
case TnefPropertyId.Priority:
switch (prop.ReadValueAsInt32())
{
case 2: message.Priority = MessagePriority.Urgent; break;
case 0: message.Priority = MessagePriority.NonUrgent; break;
}
break;
}
}
}
/// <summary>
/// Initializes a new instance of the <see cref="MimeKit.IO.Filters.CharsetFilter"/> class.
/// </summary>
/// <remarks>
/// Creates a new <see cref="CharsetFilter"/> to convert text from the specified
/// source encoding into the target charset encoding.
/// </remarks>
/// <param name="sourceEncodingName">Source encoding name.</param>
/// <param name="targetEncodingName">Target encoding name.</param>
/// <exception cref="System.ArgumentNullException">
/// <para><paramref name="sourceEncodingName"/> is <c>null</c>.</para>
/// <para>-or-</para>
/// <para><paramref name="targetEncodingName"/> is <c>null</c>.</para>
/// </exception>
/// <exception cref="System.NotSupportedException">
/// <para>The <paramref name="sourceEncodingName"/> is not supported by the system.</para>
/// <para>-or-</para>
/// <para>The <paramref name="targetEncodingName"/> is not supported by the system.</para>
/// </exception>
public CharsetFilter(string sourceEncodingName, string targetEncodingName)
: this(Encoding.GetEncoding(sourceEncodingName), Encoding.GetEncoding(targetEncodingName))
{
}
