电子邮件的文本挖掘

Question

I have a set of emails in a text file. I want to extract the body out of it. The sample document is shown below.

Email: 1
 ===============


  MIME-Version: 1.0
  Received: by 10.68.8.6 with HTTP; Sat, 7 Apr 2012 01:04:45 -0700 (PDT)
  Date: Sat, 7 Apr 2012 13:34:45 +0530
  Delivered-To: twistyprincess22@gmail.com
  Message-ID: <CAGibXq7_Gjqmp=jOCu2X8+Xngb5QuoqqMQ_ZKbu9jHCoJnFYgA@mail.gmail.com>
  Subject: hello
  From: twisty princess <twistyprincess22@gmail.com>
  To: twisty princess <twistyprincess22@gmail.com>
   Content-Type: multipart/alternative; boundary=047d7b33d826e6762004bd1239b5
  --047d7b33d826e6762004bd1239b5            
  Content-Type: text/plain; charset=ISO-8859-1

   hey How are you doing?

   --047d7b33d826e6762004bd1239b5       
    Content-Type: text/html; charset=ISO-8859-1

     <br><br>hey How are you doing?<br>

     --047d7b33d826e6762004bd1239b5--

So from this text, I just want "hey How are you doing?". I want this done using Regular Expressions and C#. Thanks

Answer 1

Use regex boundary=([^\s]+) to find boundary name

var bname = _boundaryRegex.Match(text).Groups[1].Value;

Then format text capturing regex using bname

var textCapturer = new Regex(string.Format("--{0}(?<text>.*?)(?=--)",bname);
foreach(var match in textCapturer.Matches(text))
{
    Console.WriteLine(match.Groups["text"]);
}

It finds value of boundary parameter and then tries to match text beetween --BOUNDARY lines.

Though I don't recomend you to do this kind of parsing using regex.