parse regex from array

Question

I am trying to parse out IP addresses from Outlook email headers. I've started writing some stuff in C# (because that is the example I was leveraging) and have come up with something close.

I can split the headers with the string lines[] = Regex.Split(headers, @"\r\n"); command okay, but when I try to iterate through the lines[] array, my regex for IP address fails and does not store the value in a second array:

Code:

private void button1_Click(object sender, EventArgs e)
 {
 // use a string constant to define the mapi property
 string PidTagTransportMessageHeaders = @"http://schemas.microsoft.com/mapi/proptag/0x007D001E";
 string mypattern = @"(#{1,3}\.)(#{1,3}\.)([0-9]{1,3}\.)([0-9]{1,3})";
 // string[] ip = Regex.Split(lines[i], (@"(\(|\[)(#{1,3}\.)(#{1,3}\.)([0-9]{1,3}\.)([0-9]{1,3})(\)|\])"));
 // get a handle on the current message
 Outlook.MailItem message = (Outlook.MailItem)this.OutlookItem;
 // use the property accessor to retreive the header
 string headers = string.Empty;
 try
 {
 headers = (string)message.PropertyAccessor.GetProperty(PidTagTransportMessageHeaders);
 }
 catch { 
 }
 // if getting the internet headers is successful, put into textbox
 string[] lines = Regex.Split(headers, "\r\n");
 Regex regexObj = new Regex(mypattern);
 for (int i = 0; i < lines.Length; i++)
 {
 MatchCollection matches = regexObj.Matches(lines[i]); 
 } 
 //eventually write the found IP array into textBox1.Text
 textBox1.Text = headers;
 }
 }
}

Any help or suggestions?

• Marc B: What about IPv6 addresses? Can't ignore those anymore these days.
• Grinn: Can you give us an example or two of what the headers variable contains?
• spender: is ###.###.999.999 an ipaddress?

Answer 1

Change your #'s to \d's:

string mypattern = @"(\d{1,3}\.)(\d{1,3}\.)(\d{1,3}\.)(\d{1,3})";

Note that a more accurate IPv4 address capture regular expression would be something like:

\b(?:(?:25[0-5]|2[0-4][0-9]|1[0-9][0-9]|[1-9]?[0-9])\.){3}(?:25[0-5]|2[0-4][0-9]|1[0-9][0-9]|[1-9]?[0-9])\b

...or at least add word boundaries...

\b(\d{1,3}\.)(\d{1,3}\.)(\d{1,3}\.)(\d{1,3})\b

For a simple IPv6 (standard) I like:

(?<![:.\w])(?:[A-F0-9]{1,4}:){7}[A-F0-9]{1,4}(?![:.\w])

• spender: is there a difference between \d and [0-9]? If not, you really should be consistent.
• MRAB: \d matches any digit; [0-9] is just a subset.
• spender: ...consistency question still applies here. Would you want to include the decimal chars of other character sets on the first two octets of an IP address but exclude them from the remainder? [0-9] throughout seems like a better idea.
• Grinn: @spender: Actually, I'd go the other direction. \d is just short-hand for [0-9], but you have a good point. Updated.
• spender: [0-9] is any digit 0,1,2,3,4,5,6,7,8,9. Apparently there are other digits in unicode that are included in /d

Answer 2

IPAddress.Parse Method do not reinvent the wheel.

• spender: You have to find them first, or do you propose chucking all the headers at this method? That won't work.
• user854301: Yes all rows could be parsed, why not.
• user1585685: How would you parse the header all at once?

Answer 3

If you're trying to match off IPv4, then try this beast, should be fairly close to what an actual IPv4 can be, the enclosing \b mean beginning and end of a word, so you should be able to remove those and tweak to your heart's content to get the ip based on your header format

\b(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\b