// Sample class for Copying and Pasting HTML fragments to and from the clipboard.
// Mike Stall. http://blogs.msdn.com/jmstall
using System;
using System.Diagnostics;

using System.Windows.Forms;
using System.Text.RegularExpressions;

using System.Collections.Generic;
using System.Text;
using System.IO;

/// <summary>
/// Helper class to decode HTML from the clipboard.
/// See http://blogs.msdn.com/jmstall/archive/2007/01/21/html-clipboard.aspx for details.
/// </summary>
class HtmlFragment
    #region Read and decode from clipboard
    /// <summary>
    /// Get a HTML fragment from the clipboard.
    /// </summary>    
    /// <example>
    ///    string html = "<b>Hello!</b>";
    ///    HtmlFragment.CopyToClipboard(html);
    ///    HtmlFragment html2 = HtmlFragment.FromClipboard();
    ///    Debug.Assert(html2.Fragment == html);
    /// </example>
    static public HtmlFragment FromClipboard()
        string rawClipboardText = Clipboard.GetText(TextDataFormat.Html);
        HtmlFragment h = new HtmlFragment(rawClipboardText);
        return h;

    /// <summary>
    /// Create an HTML fragment decoder around raw HTML text from the clipboard. 
    /// This text should have the header.
    /// </summary>
    /// <param name="rawClipboardText">raw html text, with header.</param>
    public HtmlFragment(string rawClipboardText)
        // This decodes CF_HTML, which is an entirely text format using UTF-8.
        // Format of this header is described at:
        // http://msdn.microsoft.com/library/default.asp?url=/workshop/networking/clipboard/htmlclipboard.asp
        // Note the counters are byte counts in the original string, which may be Ansi. So byte counts
        // may be the same as character counts (since sizeof(char) == 1).
        // But System.String is unicode, and so byte couns are no longer the same as character counts,
        // (since sizeof(wchar) == 2). 
        int startHMTL = 0;
        int endHTML = 0;

        int startFragment = 0;
        int endFragment = 0;

        Regex r;
        Match m;

        r = new Regex("([a-zA-Z]+):(.+?)[\r\n]",
            RegexOptions.IgnoreCase | RegexOptions.Compiled);

        for (m = r.Match(rawClipboardText); m.Success; m = m.NextMatch())
            string key = m.Groups[1].Value.ToLower();
            string val = m.Groups[2].Value;
                // Version number of the clipboard. Starting version is 0.9. 
                case "version":
                    m_version = val;

                // Byte count from the beginning of the clipboard to the start of the context, or -1 if no context
                case "starthtml":
                    if (startHMTL != 0) throw new FormatException("StartHtml is already declared");
                    startHMTL = int.Parse(val);

                // Byte count from the beginning of the clipboard to the end of the context, or -1 if no context.
                case "endhtml":
                    if (startHMTL == 0) throw new FormatException("StartHTML must be declared before endHTML");
                    endHTML = int.Parse(val);

                    m_fullText = rawClipboardText.Substring(startHMTL, endHTML - startHMTL);

                //  Byte count from the beginning of the clipboard to the start of the fragment.
                case "startfragment":
                    if (startFragment != 0) throw new FormatException("StartFragment is already declared");
                    startFragment = int.Parse(val);

                // Byte count from the beginning of the clipboard to the end of the fragment.
                case "endfragment":
                    if (startFragment == 0) throw new FormatException("StartFragment must be declared before EndFragment");
                    endFragment = int.Parse(val);
                    m_fragment = rawClipboardText.Substring(startFragment, endFragment - startFragment);

                // Optional Source URL, used for resolving relative links.
                case "sourceurl":
                    m_source = new System.Uri(val);
        } // end for

        if (m_fullText == null && m_fragment == null)
            throw new FormatException("No data specified");

    // Data. See properties for descriptions.
    string m_version;
    string m_fullText;
    string m_fragment;
    System.Uri m_source;

    /// <summary>
    /// Get the Version of the html. Usually something like "1.0".
    /// </summary>
    public string Version
        get { return m_version; }

    /// <summary>
    /// Get the full text (context) of the HTML fragment. This includes tags that the HTML is enclosed in.
    /// May be null if context is not specified.
    /// </summary>
    public string Context
        get { return m_fullText; }

    /// <summary>
    /// Get just the fragment of HTML text.
    /// </summary>
    public string Fragment
        get {  return m_fragment; }

    /// <summary>
    /// Get the Source URL of the HTML. May be null if no SourceUrl is specified. This is useful for resolving relative urls.
    /// </summary>
    public System.Uri SourceUrl
        get { return m_source; }

    #endregion // Read and decode from clipboard

    #region Write to Clipboard
    // Helper to convert an integer into an 8 digit string.
    // String must be 8 characters, because it will be used to replace an 8 character string within a larger string.    
    static string To8DigitString(int x)
        return String.Format("{0,8}", x);

    /// <summary>
    /// Clears clipboard and copy a HTML fragment to the clipboard. This generates the header.
    /// </summary>
    /// <param name="htmlFragment">A html fragment.</param>
    /// <example>
    ///    HtmlFragment.CopyToClipboard("<b>Hello!</b>");
    /// </example>
    public static void CopyToClipboard(string htmlFragment)
        CopyToClipboard(htmlFragment, null, null);

    /// <summary>
    /// Clears clipboard and copy a HTML fragment to the clipboard, providing additional meta-information.
    /// </summary>
    /// <param name="htmlFragment">a html fragment</param>
    /// <param name="title">optional title of the HTML document (can be null)</param>
    /// <param name="sourceUrl">optional Source URL of the HTML document, for resolving relative links (can be null)</param>
    public static void CopyToClipboard(string htmlFragment, string title, Uri sourceUrl)
        if (title == null) title = "From Clipboard"; 

        System.Text.StringBuilder sb = new System.Text.StringBuilder();

        // Builds the CF_HTML header. See format specification here:
        // http://msdn.microsoft.com/library/default.asp?url=/workshop/networking/clipboard/htmlclipboard.asp

        // The string contains index references to other spots in the string, so we need placeholders so we can compute the offsets. 
        // The <<<<<<<_ strings are just placeholders. We'll backpatch them actual values afterwards.
        // The string layout (<<<) also ensures that it can't appear in the body of the html because the <
        // character must be escaped.
        string header =
@"Format:HTML Format

        string pre =
@"<!DOCTYPE HTML PUBLIC ""-//W3C//DTD HTML 4.0 Transitional//EN"">
<HTML><HEAD><TITLE>" + title + @"</TITLE></HEAD><BODY><!--StartFragment-->";

        string post = @"<!--EndFragment--></BODY></HTML>";

        if (sourceUrl != null)
            sb.AppendFormat("SourceURL:{0}", sourceUrl);
        int startHTML = sb.Length;

        int fragmentStart = sb.Length;

        int fragmentEnd = sb.Length;

        int endHTML = sb.Length;

        // Backpatch offsets
        sb.Replace("<<<<<<<1", To8DigitString(startHTML));
        sb.Replace("<<<<<<<2", To8DigitString(endHTML));
        sb.Replace("<<<<<<<3", To8DigitString(fragmentStart));
        sb.Replace("<<<<<<<4", To8DigitString(fragmentEnd));

        // Finally copy to clipboard.
        string data = sb.ToString();
        Clipboard.SetText(data, TextDataFormat.Html);

    #endregion // Write to Clipboard
} // end of class