Ok, consider this image.

I develop an IE extension in c# and I would :
  - the distance in red, between top of screen and top of `visible webpage`
  - the distance in red between left of screen and left of `visible webpage`
  - the width/heigth of the visible webpage

Of course considering that i have the whole screen size. If i have red and black I can calculate green.

What the point ?

I have thousand screen coordinates (X,Y), i have to calcul the coordinate relative to the webpage.

Example : 

Considering
  Screen size : 1200 * 800
  Webpage size : 400*300
  Red distance between left screen border and left webpage border  : 200
  Red distance between top screen border and top webpage border  : 300

So my coordinates screen => relative webpage becomes :
  ( 100, 100 ) => OUTSIDE WEBPAGE( ignored )
  ( 1100, 650 ) => OUTSIDE WEBPAGE ( ignored )
  ( 200, 300 ) => ( 0,0 )
  ( 250, 400 ) => ( 50, 100 )

Actually i have this code, this is inherited from AddinExpress.IE.ADXIEModule, thetoolbarObj is the toolbar that I added to InternetExplorer. So i can use pointToScreen on it and i'm not far of what I need, but the left corner of the toolbar is not what I need, I need the leftcorner of the webpage.

public void getUtilsDimension()
{
    Rectangle resolution = Screen.PrimaryScreen.Bounds;
    Int32 screenWidth = resolution.Width;
    Int32 screenHeight = resolution.Height;

    AddinExpress.IE.ADXIEToolBarItem toolbarItem = this.ToolBars[0];
    AddinExpress.IE.ADXIEToolbar toolbarObj = toolbarItem.ToolBarObj;
    Point leftCornerWebPage = toolbarObj.PointToScreen(new Point(0, 0));
    Int32 toolbarHeight = toolbarObj.Height;
    Int32 toolbarWidth = toolbarObj.Width;

    Debug.WriteLine("Largeur écran : " + screenWidth);
    Debug.WriteLine("Hauteur écran : " + screenHeight);
    Debug.WriteLine("LeftCornerX : " + leftCornerWebPage.X);
    Debug.WriteLine("LeftCornerY : " + leftCornerWebPage.Y);
    Debug.WriteLine("toolbarHeight : " + toolbarHeight);
    Debug.WriteLine("toolbarWidth : " + toolbarWidth);

}

This is what I get actually, the screen is 1600*900, pointToScreen return the coordinates of the red cross ( 484,158 ). But I need the coordinates of the blue cross, as the width and heigh of visible webpage. I know I can get that with $(window) in Jquery, but i don't know how with c#.

I can access at the HTLMDocument (typeof mshtml.HTMLDocument) with this.HTMLDocument, unfortunately pointToScreen is not available on HTMLDocument object.

Edit : It s chrome on the first screenshot but of course that should be IE

Update 08/12

OK I have the width and height of the visible webpage ( black line on my screen shot ) The only missing thing is the coordinates of blue cross on my screenshot 2

var heightVisibleWebPage = HTMLDocument.documentElement.offsetHeight;
var widthVisibleWebPage = HTMLDocument.documentElement.offsetWidth;

For the bounty, I need the exact coordinates of the blue cross. No matter how. It should work no matter the Internet explorer version, favorites/tool/command/state bar displayed or not.

Update 08/12 HTMLDocument

HTMLDocument is from AddinExpress, it's not a System.Windows.Forms.HtmlDocument

public mshtml.HTMLDocument HTMLDocument
{
    get
    {
        return (this.HTMLDocumentObj as mshtml.HTMLDocument);
    }

}

His parent HTMLDocument.parentWindows is a IHTMLWindow2 object

HTMLDocumentObj is a member of

public class ADXIEModule : Component, IRemoteModule2, IRemoteModule, IObjectWithSite, IWin32Window
{
     ...
    //
    // Résumé :
    //     Gets the automation object (a COM object) of the active document, if any.
    //
    // Notes :
    //     When the active document is an HTML page, this property provides access to
    //     the contents of the HTML Document Object Model (DOM). Specifically, it returns
    //     an HTMLDocument object reference. The HTMLDocument object is functionally
    //     equivalent to the HTML document object used in HTML page script. It supports
    //     all the properties and methods necessary to access the entire contents of
    //     the active HTML document.
    //     The HTMLDocument object can be used through the IHTMLDocument interface,
    //     the IHTMLDocument2 interface, and the IHTMLDocument3 interface.
    //     When other document types are active, such as a Microsoft Word document,
    //     this property returns the document automation object of that document. For
    //     Word documents, this is the Document object.
    [Browsable(false)]
    public object HTMLDocumentObj { get; }

    ...

}

Explain when -1 for the community please ;)