Obtaining our news with XMLHTTP

If you haven't used the XMLHTTP library before, you should have a quick read of this article first: http://www.devarticles.com/content.php?articleId=25. We will create a function named LoadNews. LoadNews will handle all of the background work (retrieving and displaying the news items), while still updating the progress bar on our actual ActiveX control.

Our LoadNews function will accept a reference to a string variable (byref), and return a Boolean value, indicating the success/failure of the function as a whole. If an error occurs, its description will be stored in the string variable passed as a reference. The declaration of the LoadNews function looks like this:

Private Function LoadNews(ByRef msg As String) As Boolean 

The LoadNews function is declared as private, so that when it comes time to compile and distribute our ActiveX control, only the control itself can call the LoadNews function. Several objects from the MSXML 4 library are also created, as well as two string variables and one integer variable. These will be used to hold the details of each news item:

Dim objXMLHTTP As New MSXML2.XMLHTTP40 
Dim objXML As New MSXML2.DOMDocument40 
Dim objXML_Root 
Dim strURL As String 
Dim strHeadLine As String 
Dim counter As Integer

Next, we set the URL of the XML news page we will be retrieving. If you visit http://w.moreover.com/categories/category_list_xml.html, you will see a list of news topics. At the time of writing, there were 334 available. I have created a constant string variable in the general declarations section to hold the name of our news category, like this:

Const NewsCategory As String = "Videogame news"

This is used to form the URL of the XML page to retrieve, as shown below:

strURL = "http://p.moreover.com/cgi-local/page?c=" & NewsCategory & "&o=xml" 

Next, we use the Open and Send methods of our XMLHTTP object to actually retrieve the page. We are using the "GET" HTTP protocol, and have set asynchronous mode to false, meaning that our XML document MUST be downloaded before we can continue:

objXMLHTTP.open "GET", strURL, False 
objXMLHTTP.send

We also update the loading message and progress bar:

lblLoading.Caption = "Loading XML news feed..." 
pbProgress.Value = pbProgress.Value + 1

Once the request method of the XMLHTTP object has returned, we try and load its text (available as the responseText variable) into our XML document object. If an error occurs, the function is aborted and the value of the string-referenced variable is set to the errors description:

objXML.loadXML (objXMLHTTP.responseText) 


'Make sure that valid XML was returned 
If objXML.parseError.errorCode <> 0 Then 
msg = objXML.parseError.reason 
LoadNews = False 
Exit Function 
End If 

At this point, if the function hasn't encountered any errors, then the XML has been loaded successfully, and we are ready to extract each news item. We obtain the root element of the XML document into the objXML_Root object, and use its getElementsByTagName function to make sure it only contains

Set objXML_Root = objXML.documentElement 
objXML_Root.getElementsByTagName ("article") 

Our ActiveX control will display the top five items in the XML document (The MoreOver.com XML document returns twenty by default). We use a simple "for" loop to extract them from our XML document. Once a loop is complete, the status message and progress bar are also updated:

For counter = 1 To 5 
lblLoading.Caption = "Loading news item " & counter & "..." 
pbProgress.Value = pbProgress.Value + 1

The URL of each item is extracted into the strURL variable. The headline and source of each news item are also extracted into the strHeadLine variable:

strURL = objXML_Root.childNodes(counter).childNodes(0).firstChild.Text
strHeadLine = objXML_Root.childNodes(counter).childNodes(1).firstChild.Text & " [Source: " & objXML_Root.childNodes(counter).childNodes(2).firstChild.Text & "]"

Under the general declarations section of our ActiveX control, I have created an array that will hold each of the links to the news items:

Dim arrLinks(5) As String

When a user clicks on a label containing the details of a news item, the items link (contained in the arrLinks string array) is executed using the ShellExecute API function (more on this later). We simply set the value of each items URL to its index in the arrLinks array, like this (we use "counter-1" because arrays are indexed from 0 onwards):

arrLinks(counter - 1) = strURL 

We also set the caption of the link to the headline:

Select Case counter 
Case 1 
lblLink1.Caption = strHeadLine 
Case 2 
lblLink2.Caption = strHeadLine 
Case 3 
lblLink3.Caption = strHeadLine 
Case 4 
lblLink4.Caption = strHeadLine 
Case 5 
lblLink5.Caption = strHeadLine 
End Select

Lastly, we make sure that all of the labels are visible, destroy all of the XML objects, and return true, indicating that the function succeeded:

'Hide the status label and progress bar 
lblLoading.Visible = False 
pbProgress.Visible = False 

'Make sure the labels are visible 
lblLink1.Visible = True 
lblLink2.Visible = True 
lblLink3.Visible = True 
lblLink4.Visible = True 
lblLink5.Visible = True 

'Kill the XML objects 
Set objXMLHTTP = Nothing 
Set objXML_Root = Nothing 
Set objXML = Nothing 

LoadNews = True 
Exit Function

The LoadNews function takes care of retrieving, extracting and displaying our news links. Let’s take a look at how the ActiveX control will actually display the link.