Clarion Accessories
CapeSoft Logo

CapeSoft File Explorer feBrowser

Download Latest Version
Installed Version Latest Version
How To

Control a webpage using the FeBrowser control

Typically you'd want to load a webpage into the feBrowser control, and then programmatically manipulate the data entry to that webpage to move through the site automatically. In order to do this, you need to determine the names of the controls that you would like to manipulate, and then code the webpage control into your application to populate the data into the entry fields and "click buttons" to control the webpage.

As an example, we'll use, enter a street address to display in the control.

Step 1 is to determine the control names that you need to populate with data (in this case the street address).

File Explorer Classes - FeBrowser

Formerly known as the Html3 (or Html1 using COM) file type / object, the FeBrowser Object is used to browse html pages. This object acts in the same manner as Internet Explorer, so most things that work in IE work well with the feBrowser object. For example you can load a PDF into the feBrowser the same as you would in Internet Explorer.

FeBrowser - Methods

Functional Reference - Methods Listed by Usage

This list organizes the above methods by their functional use and describes them briefly. See the list below for an alphabetized quick reference.
Navigation and Loading
BackGo to the previously loaded page - navigates to the previous page on the navigation stack.
ForwardGo forward through the navigation stack (only applies if Back() has been used at least once.
LoadLoads an URL or file into the control
RefreshRefreshes the currently loaded page and provides control over how the refresh is done
ReloadLoad the currently loaded page or file again
StopStops the current navigation
LoadFromStreamLoads the Browser from memory
HomeNavigates to the Home page set in the Internet Explorer settings
SaveAs Saves the page as an HTML file on disk
Event Handling
EventCallbackThis gets called with each event that the Browser control fires. There are also individual callback methods for handling specific events. See Summary of callback events for more information.
TranslateEventIDTranslates an Event ID passed to the EventCallback method into a string that describes the event. This is done automatically in the specific event callback methods. See Summary of callback events.
Control Management
HideControlHides the Browser control. Hiding the Clarion Region control populated by the template will have no affect on the browser control.
UnhideControlUnhides the Browser control.
ResizeResizes the browser control. If you are populating the control using the template (the normal method of using the Browser control) then this is called for you when the window resizes if you have enabled resizing in the template settings.
SetFocusSets the focus to the window and the Browser.
Browser Management
GetInfoRetrieves a property of the Browser, such as the contents of the loaded document, the title of the page etc.
UpdateUpdates a property of the Browser
InitIntializes the Browser control.
KillDisposes the Browser and cleans up the interface and all handles etc.
NameOfFrameReturns the name of a frame that is part of the frameset in an HTML page.
NumberOfFramesReturns the number of frames in an HTML page.
HTML DOM (Document Object Model)Methods to allow the objects in an HTML page to be manipulated. This allows each tag, input, button etc to be treated as an "object" on the page and hence controlled.
ClickByNameClick onany object on the page based on the object name. Not all ta gs are named, this uses the Name attribute and will only work for named tags.
ClickByRecordNumber"Clicks" on any object on the page based on its record number in the queue of elements (populated by calling GetElementQ). This allows anything on the page to be clicked on, regardless of whether it has a name or ID.
GetElementQPopulates a queue of all elements (objects) on a page. This queue contains all the HTML tags and their values. For elements such as form inputs that may contain a large amount of data, this method is limited in the amount of data that it retrieves. The entire value, regardless of size, can be retrieved using GetValueByID and GetValueByName.
GetValueByIDRetrieve the value of an element using the element ID to locate it.
GetValueByNameRetrieve the value of an element using the element name to locate it.
SetValueByIDSets the value of an element on the page. The ID is used to locate the element.
SetValueByNameSets the value of an element on the page. The name is used to locate the element.
SubmitFormSubmits a particular form on a page. The SetValue methods allow forms to be filled and this submits the form without the ClickByName or ClickByID methods having to be used and the form submission method determined.
Tool functions
SearchPageSearches the page for a string.
EditProvides access to the Edit command such as Cut, Copy, Paste etc.
PrintMePrints the currently loaded page, either with out without user interaction. See How to load and print a page in the background and then close the FE window for an example.
GetHtmlGraphicsInfoRetrieves a list of all images used in a loaded page, including their paths.

Quick Reference

BackNavigate to the previous page
ClickByNameTriggers a click on an item on the page
ClickByRecordNumberTriggers a click on an item on the page
EditProvides access to a wide variety of "Edit" functionality such as Cut, Copy, Paste, Save, Find, Zoom, Close, Print Preview, Page Setup etc.
EmbedImagesReturns a list of images used in the document and optionally replaces references to the images with the format used for embedding images when sending HTML via email.
EventCallbackHandles events (called when a page is loaded, an item is clicked on, the mouse moves over an item etc.)
ForwardNagivates to the next page in the history
GetElementQRetrieves a queue of HTML DOM elements
GetHtmlGraphicsInfoDeprecated, use EmbedImages.
GetInfoRetrieves a variety of properties for the control including the HTML or Text for the loaded page, the current URL etc.
GetValueByIDRetrieves the value of an HTML DOM element
GetValueByNameRetrieves the value of an HTML DOM element
HideControlHides the Browse control
HomeNavigates to the Browser Home page (if one is set)
InitInitializes the control
KillDestroys the control
Load Loads a page into the control
LoadFromStreamLoad from memory (limited and not recommend, see the notes for this method for alternative approaches)
NameOfFrameReturns the name of the frame
NumberOfFramesReturns the number of frames
PrintMePrints the contents of the control
RefreshReloads the page
ReloadLoads the last document again
ResizeHandles control resizing
SaveAsSaves the currently loaded page as an HTML file on disk
SearchPageSearches the page for the specified terms
SetFocusSets focus to the Browser control
SetValueByIDSets the value on an HTML element
SetValueByNameSets the value on an HTML element
StopCancels navigation
SubmitFormSubmits the form on the page
TranslateEventIDConverts an event ID to a descriptive string
UnhideControlUnhides the control if hidden
UpdateUpdates a wide variety of properties that relate to the control


Back ( ) ,byte,proc

Description Example


ClickByName ( string ElementName, long FrameNumber=0, long NameInstance=0 ) ,byte,proc

Description Example
ThisViewer.ClickByName ('Button1')
ThisViewer.ClickByName ('Button1', 2)


ClickByRecordNumber ( long pRecordNumber, long pFrameNumber=0 ) ,byte,proc

Description Example
ThisViewer.ClickByRecordNumber (5)


Edit (long EditType, long pAction = 0, <*cstring pValue>) , byte, proc


The Edit method is used to call various "Editing" commands. These range from commands that are in the Edit menu of the browser such as "Copy", "Paste", "Cut", "Undo" and so on, to those that are in the File menu, such as "Save", "Print", "Close" and so on.

Parameter Description
long EditType An equate that specifies what action the Edit method needs to perform. The values for this equate are as follows:
fe:Browser_SaveThe same as the File->Save command, saves the contents of the Browser control to a file. This has no effect on the Browse control unless the cotents of the page have been modified
fe:Browser_SaveAsSaves the document to disk and displays a dialog box to the user to request the file name.

Use the Save As method to save the document to disk as an HTML file without user interaction.
fe:Browser_PrintFile->Print. Prints the document. Again the behaviour of the print command and whether the print options dialog is displayed is controlled by the pAction parameter.
fe:Browser_PrintPreview File->Print Preview. Displays the Print Preview dialog box. This has been greatly improved with IE7.
fe:Browser_PageSetup File->Page Setup. Displays the Page Setup dialog box.
fe:Browser_PropertiesFile->Properties. Displays information about the currently loaded page.
fe:Browser_CutEdit->Cut (not exposed in IE itself). Cuts the current selection (removes it from the browser and places it on the clipboard). Unlike a standard browser, the FE browser actually allows you to remove content from a document.
fe:Browser_CopyEdit->Copy. Copies the current selection to the clipboard.
fe:Browser_UndoEdit->Undo (not exposed in IE itself). Undoes the last action, typically not used for the Browser control, unless you are using the Cut or Delete actions (which are not typically used with the Html Browser).
fe:Browser_RedoEdit->Redo (not exposed in IE itself). Re-does the last action that was undone using the Undo command. If no Undo has been performed this has no effect whatsoever. You can cycle through the "undo" list using the Undo and Redo commands to move back and forth through the performed actions.
fe:Browser_SelectAllEdit->Select All. Selects everything within the browser control.
fe:Browser_ClearSelectionEdit->Clear Selection. Deselects everything, so that nothing in the control is selected.
fe:Browser_FindEdit->Find. Displays the Find dialog and allows the user to type a string to locate within the loaded document. Each time the Find button is pressed the next instance of the string is located and selected.
fe:Browser_DeleteEdit->Delete (not exposed in IE itself). Deletes the current selection (removes it from the document).
fe:Browser_StopDownloading Equivalent to pressing the Stop button in IE. This stops the current document from loading.
fe:Browser_CloseDocumentFile->Exit (no exact equivalent command in IE). Closes the current document. In the case of FileExplorer this does not close the Window that the FE control is on, it simply closes the loaded document and displays a blank page.
fe:Browser_ZoomThe text zoom value, takes an integer value between 0 (smallest) and 4 (largest) with 2 being the standard size (normal). (Deprecated in later browsers)
long pAction = 0 (optional) The optional second parameter allows you to specify how the command is executed for commands that normally require user feedback. For example the fe:Browser_SaveAs action normally displays a Save As dialog box, you can circumvent the display of the dialog by passing pAction as fe:OLECMDEXECOPT_DONTPROMPTUSER as the second paremeter.

There are four possible values for this parameter:

Prompt the user for input or not, whichever is the default behavior.

Execute the command after obtaining user input.

Execute the command without prompting the user. For example, clicking the Print toolbar button causes a document to be immediately printed without user input.

Show help for the corresponding command, but do not execute.
string pValue (optional) The third parameter allows a value to be passed for those actions that require one, or can optionally use one. For example, the SaveAs action allows the file name to save the document to, and in combination with, the pAction set to fe:OLECMDEXECOPT_DONTPROMPTUSER this would save the document to a specific file name, without prompting the user.

Currently supported actions:

fe:Browser_SaveAs - is documented as supporting this property, however the implementation of the COM object does not use the passed path unfortunately, and does not respect the value of the pAction parameter (a dialog is always displayed).
Return Values

Returns a byte indicating success (1) or failure (0).

    !--- Copy the entire loaded document onto the clipboard

    !--- Note that the below example could be done simply by calling the feBrowser.PrintMe() method:

    !--- Print the document without displaying the "print" dialog box
    ThisViewer.Edit(fe:Browser_Print, fe:OLECMDEXECOPT_DONTPROMPTUSER)


EventCallback ( long EventID, *cstring EventString, <*cstring Parm1>, <*cstring Parm2>, <*cstring Parm3>, <*cstring Parm4>, <*cstring Parm5>, <*cstring Parm6>, <*cstring Parm7> ) ,long


Generic event handling method. This method would not typically be used, as FileExplorer provides specific event handling methods for each event - See Summary of callback events for more information. For an example of using a callback method see the "How to Print a web page in the Background" example at the end of this document.
  1. Internet Explorer fires an event
  2. File Explorer "catches" that event, and the File Explorer dll calls the FeBrowser::EventCallback method
  3. Any code in your app, before the call to parent.EventCallback
  4. File Explorer processes the code inside the FeBrowser::EventCallback method (you have no control over this)
  5. Any code in your app, after the call to parent.EventCallback
  6. Your app returns from the EventCallback (virtual) method, and depending on the value of "ReturnVar" which your app returns, File Explorer will act accordingly


Forward ( ) ,byte,proc



GetElementQ ( <string TagName, fe:QParam_Type QParm, long FrameNumber=0 )

ThisViewer.GetElementQ ('input', locMyQ )  ! scan for all "input" elements
ThisViewer.GetElementQ ( , locMyQ )  ! scan for all elements


GetHtmlGraphicsInfo ( *string HtmlSource ), string

ThisViewer.GetInfo()  ! updates ThisViewer.Source
TempString = ThisViewer.Source  ! needed only for the next line
ThisViewer.GetHtmlGraphicsInfo (TempString)  ! updates the Source and Graphics list properties
ThisEmailSend.MessageHtml = ThisViewer.Source
ThisEmailSend.EmbedList = ThisViewer.GraphicsList


GetInfo ( ) GetInfo (string Parm1, long FrameNumber=0 ) ,*cstring,proc



GetValueByID ( string ElementID ) ,*cstring

Description <input id="id5" type="text" name="LastName" size="20">
MyCStringPtr &= ThisViewer.GetValueByID ('id5')


GetValueByName ( string ElementName, long FrameNumber=0 ) ,*cstring


<input id="id5" type="text" name="LastName" size="20">
MyCStringPtr &= ThisViewer.GetValueByName ('LastName')


HideControl ()

ThisViewer.HideControl ()


Home ( ) ,byte,proc



Init (long ParentControl, <string LoadPath>, long Handle1=0, long Handle2=0, byte SkipCallbacks=false) ,byte,proc

ThisViewer.Init (?feControl)
ThisViewer.Init (?feControl, '',,,true)


Kill ( byte level=0 ) ,byte,proc



Load (string CurrentPath, <long CurrentControl>, <string pFlags>, <string pTargetFrameName>, <string pPostData>, <string pHeaders> ) ,byte,proc


Loads a document (passed as the CurrentPath parameter) into the html viewer control.

Parameter Description
string CurrentPathThe URL to navigate to. This can be the address of a web site, the path to a document on disk etc.
long currentControl(Optional) The control to load the document into. If the CurrentControl parameter is omitted, the control which you passed in the Init() method is used.
long pFlags Reserved for future use
string pTargetFrameName:
  • (Optional) Used to set which frame the document will load into. Have a look in the example, in the procedure called "Advanced_feBrowser_Frames1" to see how this is implemented. Possible values are:
_BLANK:Load the link into a new unnamed window.
_PARENT: Load the link into the immediate parent of the document the link is in.
_SELF: Load the link into the same window the link was clicked in.
_TOP: Load the link into the full body of the current window.
<WINDOW_NAME >: A named HTML frame. If no frame or window exists that matches the specified target name, a new window is opened for the specified link.
string pPostData Reserved for future use
string pHeaders(Optional) Used to send HTTP headers to the server, which will be added to the default Microsoft Internet Explorer headers.
Return Value

Returns a byte indicating success (1) or failure (0).

ThisViewer.Load('',,, '_top')
ThisViewer.Load('',,, 'BannerFrame')


LoadFromStream (*cstring szHtml)


Loads the Browser control from a string in memory. This can be used to load web pages stored in BLOB, HTML contents directly back into the browser and using .GetInfo() to retrieve the content, modifying it, and then .LoadFromStream() to update the contents of the browser etc.

Parameter Description
*cstring szHtml A cstring (null terminated string) that contains the HTML to load. This parameter is passed by address, but is not modifying by the function.
Return Value

This method has no return value.

Note: If an error occurs then the error information will be output to the system debug log (this and all other debug information can be viewed using SysInternals DebugView, available from

htmlData       cstring(4096)  
! Note: Typically the string should be dynamically allocated unless the exact size is known.
    htmlData = '<html><head><title>Loaded From Memory</title></head>' |
               & '<body bgcolor="white"><h1>Hello World</h1>' |
               & '<p>Loaded from memory!</p></html>'



NameOfFrame ( long FrameNumber ) ,string

frameName  STRING(20)
frameName  = ThisViewer.NameOfFrame(2) ! returns the name of frame 2


NumberOfFrames ( ) ,long

totalFrames      long
    totalFrames = ThisViewer.NumberOfFrames()


PrintMe ( byte ShowDialog=true ) ,byte,proc

Description Example


Refresh (byte Level = fe:Browser_RefreshLevel_Default), byte, proc


Provides the same effect as the "Refresh" button in Internet Explorer. This forces the control to reload the page from the server, the parameter will determine how the refresh is done (for example whether a cached version of the page can be used etc.)

Parameter Description
fe:Browser_RefreshLevel_NormalOptionally, you can pass this method a byte (Level), which will determine the "type" / "level" or refreshing you want done. This can be one of the following values
fe:Browser_RefreshLevel_DefaultRefresh that does not include sending the HTTP " pragma:nocache" header to the server. The server may return a cached version of the page.
fe:Browser_RefreshLevel_IfExpiredBehaves in the same manner as fe:Browser_RefreshLevel_Normal.
fe:Browser_RefreshLevel_CompletelyRefresh that occurs if the page has expired. This refreshes the page, and if the cached version has expired then a new version will be fetched.

Refresh that includes sending a "pragma:nocache " header to the server (HTTP URLs only). This refreshes the actual page, and asks the server not to returned a cached version. This is potentially the slowest (the most data needs to be transferred) however it ensures the current version of the actual page is retrieved.
Return Values

Returns a byte, indicating success (1), or failure (0)


Reload (<long CurrentControl>)



Resize ( <long ParentControl>, long XMargin=0, long YMargin=0, long WidthAdjust=0, long HeightAdjust=0)



SaveAs (string fileName, long pAppend = false)


Saves the document to disk as an HTML file.

Parameter Description
string fileNameThe name and path of the file to save to. The HTML of the page is saved, so this should typically have the HTM or HTML file extension.
long pAppend = falseIf this is set to True (1) then if the file already exists then the page HTML is appended to it. Otherwise the file is overwritten if it exists.
Return Values

Returns True if successful and False if an error occurs.


  1. Only supports HTML pages, does not support saving PDFs or other content loaded into the control. Use the SaveAs option provided by the Edit() method to perform user interactive saving of alternative content.
  2. This functionality is implemented using the feFile class, which provides generic file reading and writing using the Windows API. This can be used to save any content to disk. For example the loaded page can be saves as Plain Text as follows:
    SavePageText Procedure(*feBrowser Browser, string fileName, long pAppend=0)
    fFile           feFile
        return fFile.Save(browser.GetInfo_OuterText, fileName, pAppend)


SearchPage ( ) ,byte,proc



SelectElementByID ( string ElementID )



SelectElementByName ( string ElementName, long FrameNumber=0, long NameInstance=0 )

Note: Certain elements (eg. FileInputElements) are marked read-only, and cannot be set using methods such as SetValueByName. The following code does work, but is a bit of a hack.. Provided here as an additional example:

ThisViewer.SelectElementByName ('MyField1')

ThisViewer.SelectElementByName ('MyField1')
ThisViewer.SelectElementByName ('MyField1', 3)


SetFocus ( )



SetValueByID ( string ElementID, string ElementValue )

<input id="id5" type="text" name="LastName" size="20">
ThisViewer.SetValueByID ('id5', 'James')


SetValueByName ( string ElementName, string ElementValue, long FrameNumber=0,long NameInstance=0 )

ThisViewer.SetValueByName ('LastName', 'James')


Stop ( ) ,byte,proc



SubmitForm ( <string FormName>, long FrameNumber=0 )

ThisViewer.SubmitForm ()
ThisViewer.SubmitForm ('form2')


TranslateEventID ( long EventID ) ,string

TempString  STRING(40)
TempString = ThisViewer1.TranslateEventID (SomeEventID)


UnhideControl ( )

ThisViewer.UnhideControl ()


Update ( string Property, string PropValue, byte FrameNumber=0)

ThisViewer1.Update ('PrinterHeader', 'This is my header')
ThisViewer1.Update ('PrinterFooter', 'This is my footer')
ThisViewer1.Update ('PrinterOrientation', 'Landscape')
ThisViewer1.Update ('PrinterOrientation', 'Portrait')
ThisViewer1.Update ('SilentMode', true)
FeBrowser - Properties (alphabetical)
Busy (Byte)Holds true (1) if a page is still being loaded, or false (0) if not. See the GetInfo method.
DocumentType (String)Holds a description of the type of document which you have loaded. See the GetInfo method.
GraphicsList (String)See the notes on the GetHtmlGraphicsInfo method. This property contains the list of graphic files found in the loaded html document.
NewWindow (Byte)By default this property is set to false (0). Setting it to true (1) prior to calling the Load method results in the url you are loading being loaded into a new window ( your default browser will load and the file will be loaded into that window ). Note that this property will not affect browsing: if you click on a hyperlink the new page will still be loaded into the parent window, unless the hyperlink properties are set to open the page in a new window.
OuterText (CString)Holds the currently loaded page's text content. See the GetInfo method.
PageTitle (String)The title of the page being displayed, e.g. 'CapeSoft Accessories Home Page'. See the GetInfo method. Note that this property will only hold a title if the page you are viewing has one! If not this property will hold the same thing that the Url property does.
RootFolder (String)This property is used in the GetHtmlGraphicsInfo method, for the following purpose. When you are trying to establish the list of graphics in an html document ( primarily useful for when you are using NetTalk to send an email written using File Explorer - see the GraphicsList property ) there may be graphics with paths such as " /MySubfolder/MyImage.gif ".  The problem arises that when you pass NetTalk a graphic path like that ( to embed the graphic into the email at send-time ), it fails as the path is not "valid". File Explorer handles this problem as follows: If you have first set the HtmlRootFolder property, before calling the GetHtmlGraphicsInfo method, then File Explorer automatically constructs valid paths for these graphics, using the HtmlRootFolder property to establish valid paths for these files. Example...

ThisViewer1.RootFolder = "c:\My Documents\Website1"
! Here we set this property with the folder that contains the actual html page we're working with. Any graphics in this html file that have no path ( Image1.gif ) or an "incomplete" path ( /Subfoler/Image.gif ) will be assumed to be in this folder or have this folder as their root.

If you do not set the RootFolder property before calling the GetHtmlGraphicsInfo method, File Explorer will assume that such graphics stem from the exe directory.
Source (CString)Holds the html component (source / text) of the loaded page.  Updates when you call the GetInfo or the GetHtmlGraphicsInfomethods.
Url (String)e.g. Holds the URL of the page currently displayed. See the GetInfo method.

FeBrowser - Events

See the FeBrowser - Summary of callback Events below, which now lists every single callback method, along with the parameters which are passed to it and a full description of what the method does and when it is called.

FeBrowser - Templates

At this time there is quite an overlap in terms of multiple File Explorer classes being implemented by the same templates. In the future we may revisit this, but for now all templates are covered in the main (FileExplorer.htm) help document, please click here.

Useful References

FeBrowser - Summary of callback events

See the FeBrowser_Callbacks procedure in the example application for more info. This is still a work in progress, we will be adding support for additional callback events as well as improving the existing callbacks. The FE Class has specific callback methods for the events, you can override these methods to implement specific behaviour for callbacks. This is a far easier and cleaner way of responding to callbacks than embedding code directly in the EventCallback() method and checking which equate was passed (although this is still supported if you prefer to take this approach).

Note: For methods that take a pDisp parameter that points to the object that the event relates to it is not advised that this variable is used within your code, it is an internal FE pointer.
Event NameEquate Description
BeforeNavigatefe:WBEvt_BeforeNavigateThis is sent before navigation to give a chance to abort.


This event is obsolete. Use the BeforeNavigate2 event instead.
BeforeNavigate2fe:WBEvt_BeforeNavigate2 Fired before navigate occurs, as in when a hyperlink is clicked on, a url is entered, or the back or forward buttons are pressed.


The FeBrowser.EventBeforeNavigate2 method is called:

FeBrowser.EventBeforeNavigate2 (long pDisp, string pURL, long pFlags, string pTargetFrameName, string pPostData, string pHeaders, byte pCancel)

  • pDisp
    Object that evaluates to the top level or frame WebBrowser object corresponding to the navigation. Used internally by FE, do not modify this property (applies to all methods that take a pDisp parameter).
  • url
    String expression that evaluates to the URL to which the browser is navigating.
  • Flags
  • TargetFrameName
    String expression that evaluates to the name of the frame in which the resource will be displayed, or Null if no named frame is targeted for the resource.
  • PostData
    Data to send to the server if the HTTP POST transaction is being used.
  • Headers
    Value that specifies the additional HTTP headers to send to the server (HTTP URLs only). The headers can specify such things as the action required of the server, the type of data being passed to the server, or a status code.
  • Cancel
    Boolean value that the container can set to True to cancel the navigation operation, or to False to allow it to proceed.

The BeforeNavigate2 event replaces the BeforeNavigate and FrameBeforeNavigate events, which should no longer be used. The pDisp parameter that specifies the corresponding WebBrowser object clearly identifies where the navigation is to occur. Microsoft Internet Explorer 3.0 required the FrameBeforeNavigate and a target frame name to identify the target of navigation. However, this solution was potentially ambiguous as frame names are not unique.

Note Internet Explorer 4.0 and later continue to fire the BeforeNavigate and FrameBeforeNavigate events for compatibility with Internet Explorer 3.0.

The pDisp parameter specifies the WebBrowser object of the top-level frame corresponding to the navigation. Navigating to a different URL could happen as a result of external automation, internal automation from a script, or the user clicking a link or typing in the address bar. The processing of this navigation can be modified by setting the Cancel parameter to True and either ignoring or reissuing a modified navigation method to the WebBrowser object.

When reissuing a navigation for the WebBrowser object, the Stop method must first be executed for pDisp. This prevents the display of a web page that declares a cancelled navigation from appearing while the new navigation is being processed.
ClientToHostWindowfe:WBEvt_ClientToHostWindowFired to request client sizes be converted to host window sizes.
CommandStateChangefe:WBEvt_CommandStateChange Fires when the enabled state of a command changes. The FeBrowser.EventCommandStateChange method is called:


FeBrowser.EventCommandStateChange(long pCommand, bool pEnable)

  • pCommand
    CommandStateChangeConstants enumeration value that specifies the command that changed. This can have one of three values:
    Enabled state of a toolbar button might have changed.

    Enabled state of the Forward button has changed.

    Enabled state of the Back button has changed.
  • pEnable
        Boolean value that specifies the enabled state:
            Command is disabled.
            Command is enabled.
DocumentCompletefe:WBEvt_DocumentCompleteFired when the document being navigated to reaches ReadyState_Complete.


FeBrowser.EventDocumentComplete (long pDisp, string pURL)

  • pDisp
    Object that specifies the top-level or frame WebBrowser object corresponding to the event.
  • URL
    String that specifies the URL, Universal Naming Convention (UNC) file name, or pointer to an item identifier list (PIDL) of the loaded document.
DownloadBeginfe:WBEvt_DownloadBeginDownload of a page started.


FeBrowser.EventDownloadBegin ()


This event is fired shortly after the BeforeNavigate or BeforeNavigate2 event, unless the navigation is canceled.

Each DownloadBegin event will have a corresponding DownloadComplete event.
DownloadCompletefe:WBEvt_DownloadCompleteDownload of page complete.


FeBrowser.EventDownloadBegin ()


Unlike NavigateComplete or NavigateComplete2, which are fired only when a URL is successfully navigated to, this event is always fired after a navigation starts

Each DownloadBegin event will have a corresponding DownloadComplete event.
FileDownloadfe:WBEvt_FileDownloadFires to indicate that a file download is about to occur. If a file download dialog is to be displayed, this event is fired prior to the display of the dialog.
FrameBeforeNavigatefe:WBEvt_FrameBeforeNavigateSee the BeforeNavigate event for a full description.
FrameNavigateCompletefe:WBEvt_FrameNavigateCompleteSee the NavigateComplete event for a full description.
NavigateCompletefe:WBEvt_NavigateCompleteIn async, this is sent when we have enough to show. Fires after a navigation to a link is completed on either a window or frameSet element.


FeBrowser.EventNavigateComplete2 (long pDisp, string pURL)

  • pDisp
    Object that evaluates to the top-level or frame WebBrowser object corresponding to the event.
  • URL
    String expression that evaluates to the URL, UNC file name, or PIDL that was navigated to.

This event is obsolete. Use the NavigateComplete2 event instead.
NavigateComplete2fe:WBEvt_NavigateComplete2 Fired when the document being navigated to becomes visible and enters the navigation stack. Fires after a navigation to a link is completed on either a window or frameSet element.


FeBrowser.EventNavigateComplete2 (long pDisp, string pURL))

  • pDisp
    Object that evaluates to the top-level or frame WebBrowser object corresponding to the event.
  • URL
    String expression that evaluates to the URL, UNC file name, or PIDL that was navigated to. Note that this URL can be different from the URL that the browser was told to navigate to. One reason is that this URL is the canonicalized and qualified URL; for example, if an application specified a URL of "" in a call to the Navigate or Navigate2 method, the URL passed by Navigate2 will be "". Also, if the server has redirected the browser to a different URL, the redirected URL will be reflected here.

This event replaces the NavigateComplete and FrameNavigateComplete events, which should no longer be used. Microsoft Internet Explorer 4.0 and Internet Explorer 5 continue to fire the NavigateComplete and FrameNavigateComplete events for compatibility with Internet Explorer 3.0.

The document might still be downloading (and in the case of HTML, images might still be downloading), but at least part of the document has been received from the server, and the viewer for the document has been created.
NewWindowfe:WBEvt_NewWindowFires when a new window is to be created.


This event is obsolete. Use the NewWindow2 event instead.
NewWindow2fe:WBEvt_NewWindow2 Fires when a new window is to be created. Create a new, hidden, nonnavigated WebBrowser.




In Microsoft Internet Explorer, the NewWindow2 event is not fired when the user selects Window from the New command on the File menu. This event precedes the creation of a new window from within the WebBrowser.

The NewWindow2 event is fired when a window is about to be created, such as during the following actions:

The user clicks a link while pressing the SHIFT key.
The user right-clicks a link and selects Open In New Window.
The user selects New Window from the File menu.
There is a targeted navigation to a frame name that does not yet exist.

Your browser application can also trigger this event by calling the Navigate or Navigate2 method with the navOpenInNewWindow flag. The WebBrowser control has an opportunity to handle the new window creation itself. If it does not, a top-level Internet Explorer window is created as a separate (nonhosted) process.

Sent when the address bar should be shown/hidden.

Sent when fullscreen / kiosk mode should be on/off

Sent when the menubar should be shown/hidden

Fired when the application is quiting

Sent when the statusbar should be shown/hidden

Sent when theater mode should be on/off

Sent when the toolbar should be shown/hidden

Sent when the window goes visible/hidden

Sent when download progress is updated

Sent when the PutProperty method is called

Fired to indicate the security level of the current web page contents

Statusbar text changed

Sent when the document title changes
WindowActivatefe:WBEvt_WindowActivateSent when main window has been activated

Fired when the WebBrowser is about to be closed by script
WindowMovefe:WBEvt_WindowMoveSent when main window has been moved
WindowResizefe:WBEvt_WindowResizeSent when main window has been sized

Fired when the host window should change its height

Fired when the host window should change its Left coordinate

Fired when the host window should allow/disallow resizing

Fired when the host window should change its Top coordinate

Fired when the host window should change its width
Other FileExplorer Specific Event Callbacks
feBrowser.EventClick()Fired when the user clicks in the control. Only traps the click event, no other information is provided (such as what the user clicked on etc.) The FeBrowser.EventBeforeNavigate() event callback can be used to trap mouse clicks that cause a navigation command to execute.
feBrowser.EventDblClick()Fired when the user double clicks in the control.

UseFul Tip:How To Load and Print Web Pages in the Background

Using FileExplorer you can load a web page, print it, and close the window that the page is loaded with, without the user having to see anything. You can do this simply by adding a procedure with a FileExplore feBrowser control, and passing it a string as a parameter. This string contains the URL of the website page to print.

This is demonstrated in the BackgroundPrint example in the Clarion\3rdparty\Examples\FileExplorer\BackgroundPrint\ folder. The example allows a file to be selected that is printed using FE in the background. While the example allows a file to be selected and a button to be pressed to print the file, this can be automated using exactly the same technique. Note that you can also specify a web page address (URL) instead of a file name and FE will print it. You can also specify any file type that IE can open - HTML, EML, MHT, PDF etc.

The example uses the following technique to implement background printing, which involves embedding a small amount of code in four embed points.

  1. Hide the Window

    In the ThisWindow.Init method, after the window has opened, hide it (this is in the embed below SELF.Open(Window)):
    ! Open the window
        Open(Window)                ! Open window
        ! [Priority 8030]
        Window{prop:hide} = 1
  2. Load the page. After the call to ThisViewer.Init(), call the .Load() method to load the page that you want to print. The example application passes a parameter to the procedure to specify the page to load.
  3. Print the page after it has been loaded in the ThisViewer.EventDocumentComplete() procedure, after the parent call add the following code:
        if Clip(pUrl) <> 'about:blank'     ! Ignore the blank page
    if Message('Would you like to print the page ' & Clip(pURL) & '?', | 'Print this page?', Icon:Question, Button:Yes + Button:No) = Button:Yes ThisViewer1.PrintMe(false) ! print the document end
  4. Close the window once the print is completed, in the ThisViewer1.EventPrintTemplateTeardown() method, again after the parent call, add the following:
    And that is it - you now have a procedure that will print any file or page that Internet Explorer can load, all in the background!

UseFul Tip: Finding which Radio button is selected.

998111[/.yWhen using Radio button on a web page, all the buttons that need to be in the same "group" should be name the same, and each given a unique value:
<input type="radio" name="R1" value="V1" checked>
<input type="radio" name="R1" value="V2">
<input type="radio" name="R1" value="V3">  
The problem is determining which radio button is selected before the form is submitted, unfortunately the Checked attribute is not set for radio buttons when getting the value from the IE HTML DOM (although it is used to determine the default selection). However a small amount of JavaScript can be used to assign the selected value to a hidden entry in the form:
<p><input type="radio" name="R1" value="V1" onMouseUp="rSelect.value='V1'" checked></p>
<p><input type="radio" name="R1" value="V2" onMouseUp="rSelect.value='V2'"></p>
<p><input type="radio" name="R1" value="V3" onMouseUp="rSelect.value='V3'"></p>
<input type="hidden" value="" name="rSelect">
The above code assigns the value to the hidden control (rSelect) when the radio button is selected. Using GetElementsQ() will list the selected value for the rRelected element. JavaScript and FileExplorer are an excellent combination for providing additional functionality and providing a rich, dynamic user interface.

Leveraging the power of the HTML DOM using JavaScript and FileExplorer

Using JavaScript in conjunction with FileExplorer allows you to create powerful, dynamic user interfaces and web pages to display in your application. You use JavaScript with FE in the same manner as using it in any other web page - simply add the required JavaScript to the HTML file that is being loaded. You can add an 'onload="MyFunction();"' to the <body> tag to run JavaScript when the application loads, you can call JS code when a control is click on, selected, moused over, moused out etc. You have full control over the document and the DOM, and there are loads of JavaScript examples and documentation available online.

Below there is some JavaScript from an recent application that uses FileExplorer to display dynamic forms (an example of which will ship with FE in the near future). In the <head> section of the HTML declare the JavaScript functions:

<script type="text/javascript">
function autoTab(input, len, e)
    var keyCode = (isNN) ? e.which : e.keyCode;
    var filter = (isNN) ? [0,8,9] : [0,8,9,16,17,18,37,38,39,40,46];
    if(input.value.length >= len && !containsElement(filter,keyCode))
        input.value = input.value.slice(0, len);
        input.form[(getIndex(input)+1) % input.form.length].focus();

    function containsElement(arr, ele)
        var found = false, index = 0;
        while(!found && index < arr.length )
            if(arr[index] == ele)
                found = true;

        return found;

    function getIndex(input)
        var index = -1, i = 0, found = false;

        while (i < input.form.length && index == -1)
            if (input.form[i] == input)
                index = i;

        return index;
    return true;

function expandTA(ta) {
    var maxChars = ta.cols;
    var theRows = ta.value.split("\n");

    var numNewRows = 0;

    for ( var i = 0; i < theRows.length; i++ ) {
        numNewRows += ( Math.ceil( theRows[i].length / maxChars ) - 1 );
    if (numNewRows + theRows.length == 0)
        ta.rows = 1
        ta.rows = numNewRows + theRows.length;

function padout(number) { return (number < 10) ? '0' + number : number; }

function SetDate(dayInput, monthInput, yearInput)
    var now = new Date();

    dayInput.value = padout(now.getDate());
    monthInput.value = padout(now.getMonth() + 1); // return from 0 to 11.
    yearInput.value = now.getYear();


The autoTab() function automatically moves to the next field when the user completes the current field and the expandTA() function expands a Text box as the user types, so each time they press return for a new line the box expands to fit the text (very useful for keeping forms compact and neat, but still functional and clear). SetDate() sets the date of the passed fields to the current date, and the PadOut() function just adds a zero to the front of the number if it is a single digit so that dates are always in the format 01/01/2007 instead of 1/1/2007 etc.

For example in the <body> tag call SetDate() to fill fields with the current date:

<body onload="SetDate(document.ConsultationForm.dayField, document.ConsultationForm.monthField,

The syntax for a fields is document.formname.fieldname. The document object is the current document and contains a number of properties and methods (any JS reference will list then for you). You declare the form as:

<form name="ConsultationForm" id="ConsulationForm">

And you can then reference it as document.ConsulationForm. I have set the ID to ConsultationForm here because I used the ID when I read the fields in using the FE GetElementsQueue() method. Similarly you can the fields to use them:

<input id="DOB_d" size="3" maxlength="2" onKeyUp="return autoTab(this, 2, event);"
       onFocus=" this.value='';" />
<input id="DOB_m" size="3" maxlength="2" onKeyUp="return autoTab(this, 2, event);"
       onFocus=" this.value='';" />
<input id="DOB_y" size="5" maxlength="4" onKeyUp="return autoTab(this, 4, event);"
       onFocus=" this.value='';" />

This declares fields for day, month and year. When the user fills in a field it jumps to the next field in the form automatically, and when the user clicks in a field it clears it. The onKeyUp and onFocus events are used.

JavaScript is immensely flexible and powerful, and you can provide some great interfaces for your user by leveraging the combination of FileExplorer and JavaScript. We will be expanding on this documentation and providing a dynamic form example with FE in the near future.

How to remove parts of a loaded page before displaying it to the user

When a page is loaded using Internet Explorer (or any other web browser) you don't have any control over which parts of the page is displayed, whatever the web server provides is displayed by the web browser. FileExplorer allows you to remove parts of web pages as well as modifying values (such as filling in forms automatically etc.).
  1. It is often helpful to hide the control using the feBrowser.HideControl() method while the content is loading. If you want to make it appear as if the control is still there then use a Box control, with a white fill that is displayed in place of the FE control while it is loading. This also allows you to add a progress control or animation to show the user that the loading of the page etc. This isn't required but means that it won't be displayed until you have finished modifying it, which may be better visually.
  2. Turn on callbacks in the feBrowser extension (or right click on the FileExplorer control in the Window Editor and choose Actions, the Callbacks setting is on the second tab).
  3. Load the URL by calling the Load() method
  4. In the EventDocumentComplete() method embed code to process the received page:
    • Check that the pUrl parameter is the page that you have just loaded (this method is called a number of times, for example it is initially called for 'about:blank' before you load any other pages).
    • Call GetElementQ() to get a list of all elements on the page.
    • For each element in the queue that you want to remove call SelectElementByID() or SelectElementByName() to select the object on the page.
    • Call Edit(fe:Browser_Delete) to delete the selected element

How to browse html content stored in a BLOB

Declare a dynamic cstring:

MyCString        &cstring

then in your code, create a new cstring for the size of the blob, and then fill the cstring with the blob's contents. You can then use the ThisViewer.LoadFromStream() method to load the cstring into the file explorer control.

Enabling HTML 5 Support in IE 9.

The IE9 "version" of the WebBrowser control, like the IE8 version, is actually several browsers in one. Unlike the IE8 version, you do have a little more control over the rendering mode inside the page by changing the DOCTYPE. Of course, to change the browser mode you have to change the registry. Here is the location of FEATURE_BROWSER_EMULATION:

            Internet Explorer
                            MyApplication.exe = (DWORD) 000090000

Here is the complete set of codes:

The full docs:

The other option is to add the following tag to the beginning of the HEAD section of the HTML document:

<META http-equiv="X-UA-Compatible" content="IE=edge" />

Note that it must be before all HEAD elements except TITLE and any other META tags in order to work.

You can use the same approach to get and set the values of elements on the page by calling GetValueByID(), GetValueByName(), SetValueByID(), SetValueByName() and elements can be "clicked" on using ClickByRecordNumber().