sicos1977 / msgreader Goto Github PK

Nice tool!! I can now use it on my server to read msg files, as I do not have Outlook on the server.

Request - could you make the exe to remember the last window size and position? I find that each time I open the app, I have to resize the window. It would be nice if the tool could store this info, and resize each time it is opened.

Thanks again for this great tool!!

Hello! When i try load message https://drive.google.com/file/d/0B1gFe2F9vE6FTDhRRXVFbmsyTzQ/view?usp=sharing it returns message with Date delivery = 01.01.0001, but if you check UnknownHeaders you will find Delivery-date.
Do you know why?
Thank you

Hi Kees,

The conversion of the e-mail is not always going well. Sometimes it converts Dutch written text to (what I think is) Greece.

Any clue what that might be?

Thanks!

Jouke van Rossum

In commit 77b81b5 there is some refactoring done that removes a large block of code from Reader.cs. In this block of code, there is a check for attachments, and then attachments are added into the header. Now instead this block was removed and replaced with a very simplified version, ExtractMsgEmailHeader(message, hyperlinks), which does nothing with attachmentList. This change needs to be reverted or fixed.

Hello,i've just installed your library via Nuget. First, thank you for your work.
I've successfully extracted the content of several .msg files and put them in separate folders.
I would like to access to the metadata of each message (i.e. recipient and sender addresses, mail subject, timestamps etc.) but I can't understand how to do it.
Thanks in advantage,
Sebastiano

Hi Kees,

My bulk import went very well thanks to your great library. I encountered only two issues.

[1] In some cases the return statement in the following line

https://github.com/Sicos1977/MSGReader/blob/master/MsgReader/Outlook/Message.cs#L1279

Prevents the code to reach the lines at 1295 and 1296:

GetHeaders();
SetEmailSenderAndRepresentingSender();

That caused me to end up with empty Sender fields. I worked around this issue by moving the two method calls to lines 1225 and 1226, so before the if statement.
I can't send you an example msg because of customer confidentiality. I know that sucks.

[2] In some other rare case the e-mail addresses contained the Exchange Legacy Distinguished Name, so something like /o=SomeDomain/ou=SomeGroup/cn=Recipients/cn=Morgan Cheng instead of an e-mail address.

In those cases I could work around most of them by running the returned e-mail addresses through the following quick hack AD lookup method (found somewhere on Stack Overflow ...I think)...

using System.DirectoryServices;

public string ResolveEmailAddress(string email)
{
    // Check if the address is set and if it's an Exchange Distinguished Name (i.e. Canonical CN=... etc)
    if (string.IsNullOrWhiteSpace(email) || !email.Contains("CN=")) return email;

    using (var ds = new DirectorySearcher())
    {
        ds.Filter = $"(&(ObjectClass=User)(legacyExchangeDN={email}))";
        var src = ds.FindAll();

        // Return the Exchange Distinguished Name in case we can't get the e-mail address for some reason
        if (src.Count != 1 || src[0] == null || src[0].Properties == null || src[0].Properties.Count == 0 || 
            !src[0].Properties.Contains("Mail") || src[0].Properties["Mail"].Count == 0) return email;

        return string.IsNullOrWhiteSpace(src[0].Properties["Mail"][0].ToString())
            ? email
            : src[0].Properties["Mail"][0].ToString();
    }
}

Hello Kees,

many thanks for your work.
We are using your project for displaying e-mails exported from outlook in an outlook addin.
Unfortunately we got an e-mail which is not displayed correctly, can you please look at it,
my knowledge goes not far enough to fix it by myself.

Best regards
Dennis

MSGReader issue.zip

Hi Kees,

Noticed a small issue on the HeaderParser for Mime Types.

When parsing the content-type in the public static ContentType ParseContentType(string headerValue) method i've had to add in the following line // Check for illegal content-type var v = value.ToUpperInvariant(); if (v.Equals("TEXT") || v.Equals("TEXT/")) value = "text/plain"; if (v.EndsWith("\0")) value = v.Replace("\0", String.Empty); << As some Content types are coming through with a "\0" suffix eg "IMAGE/GIF\0" which is causing an assignment error, removing the "\0" resolves the issue.

Hi, I just tried to use this project to open some msg-files I was handed.
But, somehow this does not work as it should.

I am using the same code as in the readme:

using (var msg = new MsgReader.Outlook.Storage.Message("c:\\test.msg"))
{
        var from = msg.Sender;
        var sentOn = msg.SentOn;
        var recipientsTo = msg.GetEmailRecipients(Storage.Recipient.RecipientType.To, false, false);
        var recipientsCc = msg.GetEmailRecipients(Storage.Recipient.RecipientType.Cc, false, false);
        var subject = msg.Subject
        var htmlBody = msg.BodyHtml;
        // etc...
}

It fails when trying to read with the following exception:
<System.ArgumentException: Object must be of type Int32. at System.Int32.CompareTo(Object value)

I am attaching the zipped msg-file for you here. It was created with Outlook 2016: Test.zip

Is there something I am doing wrong? Is Outlook 2016 supported? Are there anything special with the file I uploaded (encoding, language etc) that makes it hard/impossible to read? As far as I know it was created by saving/dragging it from Outlook onto the desktop, where Outlook suggested the msg extension.

Hi and thanks for this # great library.
I'm currently having problems with any msg-files.
There seems to be an issue in https://github.com/Sicos1977/MSGReader/blob/master/MsgReader/Outlook/MapiTagMapper.cs#L95

I do not understand exactly which properties of the email throws an exception,
but it seems that the array entryStreamBytes[] is not allocated correclty and is not big enough.

The exception returned in "Index was outside the bounds of the array."
could you help me?

Tnx Jljch
best reguards

Hello Kees,
some lines in the 'German Translation' should be corrected:

public const string EmailSignedBy = "Ondertekend Tür";
To
public const string EmailSignedBy = "Signiert von";

public const string EmailSignedByOn = "op";
To
public const string EmailSignedByOn = "Signiert am";

Also the filter in 'ViewerForm.cs' should be enlarged to allow EML-Files:

Filter = "MSG Files (.msg)|.msg|EML Files (.eml)|.eml",

Best regards,
Ron
BTW: Thanks for the great work you have done.

You guys might want to checkout MimeKit for message parsing - it'll make your lives a whole lot easier (it's very robust as well).

It can also handle S/MIME and PGP/MIME messages.

https://github.com/jstedfast/MimeKit

when to upload other than English like Japanese language mail then message shown as question mark ?? means not read correctly

Please provide appropriate solution

Hi Kees

I have an e-mail which is throwing the above exception. The CleanUpCharacters method is not helping. As a .eml file the Convert method succeeds, but as a .msg it throws. I have attached both files:

eml-and-msg.zip

Any help would be appreciated.

Kind regards
Michael

delete this!!!

using (var outlookMsg = new MsgReader.Outlook.Storage.Message(pathPSgFile))
{
Console.WriteLine("Subject: {0}", outlookMsg.Subject);
Console.WriteLine("Body: {0}", outlookMsg.BodyText);

            Console.WriteLine("{0} Recipients", outlookMsg.Recipients.Count);
            foreach (MsgReader.Outlook.Storage.Recipient recip in outlookMsg.Recipients)
            {
                Console.WriteLine(" {0}:{1}", recip.Type, recip.Email);
            }

            Console.WriteLine("{0} Attachments", outlookMsg.Attachments.Count);
            foreach (MsgReader.Outlook.Storage.Attachment attach in outlookMsg.Attachments)
            {
                Console.WriteLine(" {0}, {1}b", attach.FileName, attach.Data.Length);
            }

Attached is msg template

        }

IDENTIFIER HTML Test Mail w Att - Smoke-Test.zip

Is it possible to expose the underlying Stream of Storage.Attachment.Data?
For checking the dimensions of an image for example this would be useful as you don't need to load the whole attachment into memory...

I am experiencing an issue when using Nuget restore with MSGReader package 1.9.3. This issue is documented https://nuget.codeplex.com/workitem/4489.

WARNING: Unexpected XML declaration. The XML declaration must be the first node in the document, and no white space characters are allowed to appear before it.

Hi Kees,

A very good day to you.

Could I ask how does one open a .MSG file in order to delete the attachments? My attempts to do so have resulted in errors claiming that 'MsgReader.Exceptions.MRCannotRemoveAttachment: Cannot remove attachments when the file is not opened in Write or ReadWrite mode'

I have tried using the following code:

private static void TestMsgReader()
{
System.IO.FileStream filetoread = new System.IO.FileStream(@"D:\wph\sources\ReadAnOutlookMsg\bin\Debug\testsomethings.msg", FileMode.Open, FileAccess.ReadWrite);

        Storage.Message mymessage = new Storage.Message(filetoread);            

        List<object> mymessageattachments = mymessage.Attachments;

        object currattachment = null;

        for (int x = 0; x < mymessageattachments.Count; x++)
        {
            currattachment = mymessageattachments.IndexOf(x);
            mymessage.DeleteAttachment(currattachment);                
        }

       mymessage.Dispose();
        filetoread.Close();
        filetoread.Dispose();
    }

Many thanks and regards,
Gerald

I noticed, that the MsgReader.resources.dlls are missing in the nuget-package. Could you please provide the de/MsgReader.resources.dll, fr/MsgReader.resources.dll and so on in the nuget-package (or provide a way to localize the strings)?
Thanks a lot!

Hi,

I currently work on a french mail extractor, from more sources. I treat .msg and .eml files.

In France we often use accent on the sender name, and lot of software save this display name in the Mime Headers with UTF-8 encoding but the subject is word encoded in a .eml file. Example :

From: "Xx santé xxxxxx" <[email protected]>
...
Subject: =?UTF-8?B?RG...

So my question : Can I add some methods with encoding different from ASCII although it does not correspond to the RFC? Or should I use another approach to retrieve those emails in UTF-8 ?

Thanks,

Yanos

Hi and thanks for this great library. I'm currently having problems with german msg-files. There seems to be an issue in https://github.com/Sicos1977/MSGReader/blob/master/MsgReader/Rtf/DomDocument.cs#L3345

(Rtf/DomDocuments.cs, line 3345). The hexBuffer is empty at this point, and therefore byte.Parse throws an exception. reader.CurrentToken.Hex however contains 0xFC (a german Umlaut in 1252 encoding: 'ü'). If I add
if (hexBuffer == string.Empty) hexBuffer = reader.CurrentToken.Hex;
The code works, but the font-size changes at the Umlauts position.

Please see this testmail: link removed

Hi Kees,

I ran into the following issue. When creating a new Storage.Message like this:

var msgFilePath = @"c:\temp\email.msg";
return new Storage.Message(msgFilePath);

I got a 'Storage can not be null' exception from Storage.cs.
When using the overloaded Storage.Message constructor which expects a Stream, no exception occurs:

var msgFilePath = @"c:\temp\email.msg";
return new Storage.Message(File.OpenRead(msgFilePath));

Hallo,

I have used the library for an test with EML Mails, generated by Lotus Notes..
In the Header is a time as send date with an EST zone Tag. The regex in the file "Rfc2822DateTime.cs" interpret this tag as +0500-0600-0700 Hours.
The Regex interprets E = +0500 S = -0600 T= -0700.
I have changed the regex - in a private branch- from the file "Rfc2822DateTime.cs" from
private const string RegexOldTimezoneFormats = @"UT|GMT|EST|EDT|CST|CDT|MST|MDT|PST|MSK|PDT|[A-I]|[K-Y]|Z";
into
private const string RegexOldTimezoneFormats = @"\b((UT|GMT|EST|EDT|CST|CDT|MST|MDT|PST|MSK|PDT)|([A-IK-Za-ik-z]))\b";
Now the Time interpreter works fine :-)

Best Regards
MB

Note to myself:

Opvolginsmarkering should be Opvolgingsmarkering

I was able to get this to run and successfully convert .msg to .eml but cannot compile with xamarin.ios I really would love to use this feature but am stopped by the xamarin.ios issue. It would be great if this could work on xamarin as well thanks in advance.

An unhandled exception of type 'System.ArgumentException' occurred in MsgReader.dll

Additional information: Unknown parameter in Content-Disposition. Ask developer to fix! Parameter: VOICE

We have a request to read MSG files and store the first x characters from the body into a SQL Server table. The MSGReader.dll works great except when a MSG file has a voicemail attachment. Any workarounds? we were using 1.9.6 but I tried 1.9.10 and received the same error.

When forwarding an email which has been forwarded I see this error message returned:

ExceptionMessage":"The provided stream is not a valid IStorage\r\nParameter name: storageStream","ExceptionType":"System.ArgumentException","StackTrace":" at MsgReader.Outlook.Storage..ctor(Stream storageStream, FileAccess fileAccess) in C:\Users\Kees\Documents\GitHub\MsgReader\MsgReader\Outlook\Storage.cs:line 196\r\n

and the thread from the previous email is not attached.

Thanks!

James

In previous versions of the MSGReader the HTML output showed the attachments of the e-mail. This no longer occurs.

I'm using version 2.0.0.0 of the MSGReader.

Hi,

the file name is derived from the mails subject.
If the length of file name and path equals or exceeds MaxPath (248 characters) the first ExtractToFolder runs as expected.
The second call to ExtractToFolder with the same mail file loops infinitely in FileManager.FileExistsMakeNew because the shortened

I'd like to suggest to replace

        while (File.Exists(tempFileName))
        {
            tempFileName = path + fileNameWithoutExtension + "_" + i + extension;
            tempFileName = validateLongFileName ? ValidateLongFileName(tempFileName, extraTruncateSize) : tempFileName;
            i += 1;
        }

by

        while (File.Exists(tempFileName))
        {
            tempFileName = validateLongFileName ? ValidateLongFileName(tempFileName, extraTruncateSize) : tempFileName;
            fileNameWithoutExtension = GetFileNameWithoutExtension(tempFileName);
            tempFileName = path + fileNameWithoutExtension + "_" + i + extension;
            i += 1;
        }

Thanks!
Sincerely
Matthias

I have someone with an RTF msg with inline pdf. We hit a stack overflow with it,

There are a couple hundred levels of stack all at the same spot of code, around line 368 of Rtf.DomDocument.cs:

if (reader.TokenType == RtfTokenType.GroupStart)
  Load(reader,format); // this is where recursion is at

Unfortunately I am not able to reproduce this in a msg file that I can send to you (it is a sensitive email). Any idea as to what could cause the issue? Is the iterator not changing positions?

Hello Kees,

I'm having a problem when accessing the BodyHtml property in Outlook.Storage.Message class.

If the email format is RTF, like this one written and sent with Outlook 2010: broken_rtf.zip I get the following exception:

--NullReferenceException at MsgReader.Rtf.DomDocument.Load(Reader reader, DocumentFormatInfo parentFormat) in C:\Users\Kees\Documents\GitHub\MsgReader\MsgReader\Rtf\DomDocument.cs:line 1466 at MsgReader.Rtf.DomDocument.Load(Reader reader, DocumentFormatInfo parentFormat) in C:\Users\Kees\Documents\GitHub\MsgReader\MsgReader\Rtf\DomDocument.cs:line 367 at MsgReader.Rtf.DomDocument.LoadRtfText(String rtfText) in C:\Users\Kees\Documents\GitHub\MsgReader\MsgReader\Rtf\DomDocument.cs:line 293 at MsgReader.Outlook.Storage.Message.get_BodyHtml() in C:\Users\Kees\Documents\GitHub\MsgReader\MsgReader\Outlook\Message.cs:line 1066

I tried to debug it by myself, but I don't have enough knowledge about RTF to know what is happening here.

Hello,
I found msgreader to be able to cope with a lot of problematic msg format mails where other libraries fail. The only problem I have is that I need to use this assembly so that I can extract attachments, mail body text and meta data (from, to, cc, subject, etc.) into memory. I found that msgreader only supports saving into files first which is very inefficient in my scenario.
Could you add support for extracting everything (files and metadata) into streams/byte arrays and properties? I guess you do that internally, so you just need to expose these as public properties.
Thanks.

Hi,
I am working with the Version 1.6.1.
If I send a message via Outlook 2013 with quoted text as html-mail the MSGReader deletes the first quote and all text within the quotes and returns just the final quote.
Example to reproduce the issue
Sent message
This is a "quoted test" to reproduce the issue.
Returns
This is a " to reproduce the issue
Regards
Ron

Hi,

The first time the DocDocument.ReadHtmlContent method encounters an escaped character (e.g., \'c6) it generates a span tag with font styling which is not required (and also wrong, style-wise).

This RTF snippet (generated by Outlook 2013):

{\*\htmltag64 <p class=MsoNormal>}\htmlrtf {\htmlrtf0 X
{\*\htmltag244 <o:p>}
{\*\htmltag252 </o:p>}\htmlrtf\par}\htmlrtf0

{\*\htmltag72 </p>}
{\*\htmltag64 <p class=MsoNormal>}\htmlrtf {\htmlrtf0 \'c6
{\*\htmltag244 <o:p>}
{\*\htmltag252 </o:p>}\htmlrtf\par}\htmlrtf0

{\*\htmltag72 </p>}
{\*\htmltag64 <p class=MsoNormal>}\htmlrtf {\htmlrtf0 \'d8
{\*\htmltag244 <o:p>}
{\*\htmltag252 </o:p>}\htmlrtf\par}\htmlrtf0

becomes:

<p class=MsoNormal>
X<o:p>
</o:p>
</p>
<p class=MsoNormal>
<span style = 'font-family:"Arial";font-size:12pt'>Æ<o:p>
</o:p>
</p>
<p class=MsoNormal>
Ø<o:p>
</o:p>
</p>

Complete RTF from msg file: body.rtf.txt and HTML generated by MSGReader: body.html.txt.

The problem can be solved by removing the following code from DocDocument.ReadHtmlContent:

if (FontTable != null && fontIndex.HasValue && fontIndex <= FontTable.Count)
{
	// <span style = 'font-size:12.0pt;font-family:"Arial",sans-serif' >
	var font = FontTable[fontIndex.Value];
	if (!spanTagWritten)
	{
		stringBuilder.Append("<span style = 'font-family:\"" + font.Name + "\";");
		if (fontSize.HasValue)
			stringBuilder.Append("font-size:" + fontSize + "pt");
		stringBuilder.Append("'>");
		spanTagWritten = true;
		encoding = font.Encoding ?? _defaultEncoding;
	}
}

This have worked for my use cases so far and there does not seem to be any reason to treat escaped characters differently, at least for files generated by Outlook 2013.

Thanks!
Henning

Hi,

Is the write privilege on the following line required?

https://github.com/Sicos1977/MSGReader/blob/master/MsgReader/Outlook/Storage.cs#L109

Unsure how structured storage works, but it'd be good for a use case I've got (where I can't write to the particular location where the .MSG to parse lives) to be able to parse the message without the filesystem complaining

Hello Kees,

We are having issues opening a number of MSG files in our system using your MSGViewer we are getting the "compressed-RTF CRC32 failed" error. The body of the MSG for these files is RTF. The files open normally in Outlook. I have attached 2 sample files and would appreciate any assistance you could provide.

Example MSGs.zip

Hello,

Regarding your code, this roughly what i need for a command line tool that lets me extract recipents addresses from all msg files in a directory. Such an utility would basically help me extract addresses from abuse RFC822 complaint mails in order to blacklist them.
I'm not good at reading C#, would it be hard to add such a command line tool ?

Regards,
Ozy.

Could we get a nuget package for this?

Hey Kees,

I'm currently having an issue where an msg appointment has optional attendees but are not being extracted by the program

Hi Kees, thanks for this great and useful library! It came just on time!

I'm importing a large quantity of Outlook *.msg e-mail files into our CRM system. In CRM I can set a property on e-mails to differ between received mails and sent mails. Is there a way to determine this for the *.msg files? I couldn't find a property on the MsgReader.Outlook.Storage.Message Type. My test file also returned filled SentOn and ReceivedOn properties, so those can't be used either.
Any help appreciated.

I created a console app that loops through a directory with msg files. Every msg is analyzed and some data is extracted. It all works fine except when a msg file is opened. Of course it throws an exception and used a try/catch to catch the exception.

This all works fine until the program is done executing. When I place breakpoints at the end of the main() function, it executes everything as it should, but after that it throws a null reference exception:

An unhandled exception of type 'System.NullReferenceException' occurred in mscorlib.dll
Additional information: Object reference not set to an instance of an object.

mscorlib.dll!System.Runtime.InteropServices.Marshal.ReleaseComObject(object o) MsgReader.dll!MsgReader.Outlook.Storage.ReferenceManager.Finalize()

I've parsed non-English msg files and then message shown as marks or ??.

for example:
msg files in MsgReadzip\MsgRead\bin\Debug\Inbox shows：

Chinese part ¡
ÁŤ¤¦Ã

Japanese part ¡


fullwidth ¡
¢¢¢¡¢¢¢¢¢¡¢¢¡¢¢¢

I've tried MsgReader.Reader.ExtractMsgEmailBody(sr, false, "text/plain"); but still.

Please help.
thanks

MsgRead.zip

I’m working on an internal project for storing e-mails and documents and we are trying to extract the attachments of e-mails and store the attachments separately.

But we would like to remove the extracted attachments documents from the e-mail before we save the e-mail to our storage.

This solution works great for extracting attachments, but I have not been able to remove attachments from the e-mail.

Is that something you can add to this project?

I noticed that if I open a msg file that has web links, and then click on one of the links, it will actually navigate to the target contents, much like a web browser does. This is a nice feature indeed, but I found that I had no way to navigate back to the original msg/email content.

So, my request is to add a "Reload message" button to the form so we could easily get back to the original message.

Can you please specify what license your project is using - public domain, apache, LGPL, etc. This project looks great and I would love to be able to use it, but I want to make sure that I legally can. Thanks. Keep up the good work.

I have a Msg file (trying to see if I am allowed to upload) that was created by outlook (reproducible). This file causes PreProcessMsgFile to die at the following code:

fileInfo = new FileInfo(FileManager.FileExistsMakeNew(outputFolder + attachmentFileName));
File.WriteAllBytes(fileInfo.FullName, attach.Data);

The issue is attach.Data is null.
The file does have an attachment, and actually message.Attachments.length==2. The first attachment is good (pdf), but second attachment has filename of Nameless.

I've added the following if() continue;:

if (attachment is Storage.Attachment)
{
   var attach = (Storage.Attachment)attachment;
   if ( attach.Data == null )
      continue;
   attachmentFileName = attach.FileName;

Hello,
If message contains incorrect base64 of embedded attachments it could be loaded without body, but i think it is much better to add option to load message with ignore load bad embedded attachment.
Thank you

Sorry I forgot something, after sending the regular expression. In the RFC 2822 specification part "4.3. Obsolete Date and Time" is specified that the military Zone in Upper and lower case are allowed.
So in the regular expression this digits are used as lower and upper case sign, but in the method "MatchEvaluator" the switch/case accepted only Uppercase.
Please add the case for the lowercase Military sign like my sample:

    /// <summary>
    /// Convert timezones in older formats to [+-]dddd format.
    /// </summary>
    /// <param name="match">The match that was found</param>
    /// <returns>The string to replace the matched string with</returns>
    /// 
    /// <remarks>
    /// 
    ///
    /// .. Look here to check the RFC 2822: http://www.rfc-base.org/rfc-2822.html
    ///     
    /// ... here the snippet
    ///     
    ///     4.3. Obsolete Date and Time  
    /// 
    ///         The syntax for the obsolete date format allows a 2 digit year in the
    ///         date field and allows for a list of alphabetic time zone
    ///         specifications that were used in earlier versions of this standard.
    ///         It also permits comments and folding white space between many of the
    ///         tokens.
    ///     
    ///         obs-day-of-week =       [CFWS] day-name [CFWS]
    ///     
    ///         obs-year        =       [CFWS] 2*DIGIT [CFWS]
    ///     
    ///         obs-month       =       CFWS month-name CFWS
    ///     
    ///         obs-day         =       [CFWS] 1*2DIGIT [CFWS]
    ///     
    ///         obs-hour        =       [CFWS] 2DIGIT [CFWS]
    ///     
    ///         obs-minute      =       [CFWS] 2DIGIT [CFWS]
    ///     
    ///         obs-second      =       [CFWS] 2DIGIT [CFWS]
    ///     
    ///         obs-zone        =       "UT" / "GMT" /          ; Universal Time
    ///     
    ///     
    ///     
    ///         Resnick                     Standards Track                    [Page 31]
    ///         
    ///         RFC 2822                Internet Message Format               April 2001
    ///     
    ///     
    ///                                                         ; North American UT
    ///                                                         ; offsets
    ///                                 "EST" / "EDT" /         ; Eastern:  - 5/ - 4
    ///                                 "CST" / "CDT" /         ; Central:  - 6/ - 5
    ///                                 "MST" / "MDT" /         ; Mountain: - 7/ - 6
    ///                                 "PST" / "PDT" /         ; Pacific:  - 8/ - 7
    ///     
    ///                                 %d65-73 /               ; Military zones - "A"
    ///                                 %d75-90 /               ; through "I" and "K"
    ///                                 %d97-105 /              ; through "Z", both
    ///                                 %d107-122               ; upper and lower case -- imported lower and upper
    ///
    /// .. end of snippet
    /// 
    /// 
    /// </remarks>
    private static string MatchEvaluator(Match match)
    {
        if (!match.Success)
        {
            throw new ArgumentException("Match success are always true");
        }

        switch (match.Value)
        {
                // "A" through "I" and "a" through "i"
                // are equivalent to "+0100" through "+0900" respectively
            case "A": case "a":
                return "+0100";
            case "B": case "b":
                return "+0200";
            case "C": case "c":
                return "+0300";
            case "D": case "d":
                return "+0400";
            case "E": case "e":
                return "+0500";
            case "F": case "f":
                return "+0600";
            case "G": case "g":
                return "+0700";
            case "H": case "h":
                return "+0800";
            case "I": case "i":
                return "+0900";

                // "K", "L", and "M" and "k", "l" and "m"
                // are equivalent to "+1000", "+1100", and "+1200" respectively
            case "K": case "k":
                return "+1000";
            case "L": case "l":
                return "+1100";
            case "M": case "m":
                return "+1200";

                // "N" through "Y" and "n" through "y"
                // are equivalent to "-0100" through "-1200" respectively
            case "N": case "n":
                return "-0100";
            case "O": case "o":
                return "-0200";
            case "P": case "p":
                return "-0300";
            case "Q": case "q":
                return "-0400";
            case "R": case "r":
                return "-0500";
            case "S": case "s":
                return "-0600";
            case "T": case "t":
                return "-0700";
            case "U": case "u":
                return "-0800";
            case "V": case "v":
                return "-0900";
            case "W": case "w":
                return "-1000";
            case "X": case "x":
                return "-1100";
            case "Y": case "y":
                return "-1200";

                // "Z", "z", "UT" and "GMT"
                // is equivalent to "+0000"
            case "Z": case "z":
            case "UT":
            case "GMT":
                return "+0000";

                // US time zones
            case "EDT":
                return "-0400"; // EDT is semantically equivalent to -0400
            case "EST":
                return "-0500"; // EST is semantically equivalent to -0500
            case "CDT":
                return "-0500"; // CDT is semantically equivalent to -0500
            case "CST":
                return "-0600"; // CST is semantically equivalent to -0600
            case "MDT":
                return "-0600"; // MDT is semantically equivalent to -0600
            case "MST":
                return "-0700"; // MST is semantically equivalent to -0700
            case "PDT":
                return "-0700"; // PDT is semantically equivalent to -0700
            case "PST":
                return "-0800"; // PST is semantically equivalent to -0800

                // EU time zones
            case "MSK":
                return "+0400"; // MSK is semantically equivalent to +0400

            default:
                throw new ArgumentException("Unexpected input");
        }
    }   

Sorry I have forget this, but in my branch I have changed it.

Best regards
MB