Convert HTML Email Body to Plain Text

C# example code to load an email and convert the HTML body to plain text. This uses both the Chilkat Mail and Chilkat Html-to-Xml components.

Download Chilkat .NET for 2.0 Framework

Download Chilkat .NET for 1.0 / 1.1 Framework

//  
string XmlToText(string xmlStr)
        {
            StringBuilder sb = new StringBuilder();

            Chilkat.Xml xml = new Chilkat.Xml();
            xml.LoadXml(xmlStr);

            Stack<int> curIndex = new Stack<int>(100);
            Stack<int> numChildren = new Stack<int>(100);

            // We are on the 1st child for the current node...
            curIndex.Push(0);
            numChildren.Push(xml.NumChildren);

            while (true)
            {
                int idx = curIndex.Peek();
                int nChildren = numChildren.Peek();

                // Have we finished with the children at this node?
                // If so, move back up to the parent.
                if (idx == nChildren)
                {
                    curIndex.Pop();
                    numChildren.Pop();
                    if (curIndex.Count == 0)
                    {
                        // We're done...
                        break;
                    }
                    xml.GetParent2();
                }
                else
                {
                    // Traverse to the Nth child.
                    xml.GetChild2(idx);

                    // Tags that should represent a CRLF in plain-text are
                    // caught here...
                    if (xml.Tag.Equals("br") ||
                        xml.Tag.Equals("div") ||
                        xml.Tag.Equals("tr") ||
                        xml.Tag.Equals("p"))
                    {
                        sb.Append("\r\n");
                    }
                    sb.Append(xml.Content);

                    // Increment the current index at the top of the stack.
                    int tempIdx = curIndex.Pop();
                    curIndex.Push(tempIdx + 1);

                    // Does this child have any children?
                    int n = xml.NumChildren;
                    if (n > 0)
                    {
                        curIndex.Push(0);
                        numChildren.Push(n);
                    }
                    else
                    {
                        // Just move back up.
                        xml.GetParent2();
                    }

                }

            }

            return sb.ToString();

        }

        private void button1_Click(object sender, EventArgs e)
        {
        //  Create a MailMan first for the purpose
        //  of unlocking the component.
        Chilkat.MailMan mailman = new Chilkat.MailMan();
        mailman.UnlockComponent("anything for 30-day trial");

        Chilkat.Email email = new Chilkat.Email();

        //  Load an email from a .eml
        bool success;
        success = email.LoadEml("myEmail.eml");
        if (success != true) {
            MessageBox.Show(email.LastErrorText);
            return;
        }

        // Get the body of the email, which may or may not be HTML.
        string bodyStr = email.Body;

        // Assume the body is HTML...
        // Convert it to XML:
        Chilkat.HtmlToXml htmlConv = new Chilkat.HtmlToXml();
        htmlConv.UnlockComponent("Anything for 30-day trial");

        // br tags are dropped by default, but we want them...
        htmlConv.UndropTagType("br");

        // Drop some tag types to simplify the resultant XML...
        htmlConv.DropTagType("font");
        htmlConv.DropTagType("b");
        htmlConv.DropTagType("i");
        // Remove scripts.
        htmlConv.DropTagType("script");
        // Exclude style sheets.
        htmlConv.DropTagType("style");

        htmlConv.Html = bodyStr;
        string xml = htmlConv.ToXml();
        string plainText = XmlToText(xml);

        // We don't want more than 2 CRLF's in a row:
        while (plainText.Contains("\r\n\r\n\r\n"))
        {
            plainText = plainText.Replace("\r\n\r\n\r\n", "\r\n\r\n");
        }

        textBox1.Text = plainText;

        }