[RESOLVED] Prevent WebBrowser From Loading Pictures

[shirazamod]

I need my app to load webpages very quickly, so how can I prevent it from downloading images? The specific images that I am trying to stop are from a certain domain, so maybe its possible to prevent downloading from that domain?
I need to make sure it doesnt change any settings in internet explorer.

I heard somewhere that this can be done by removing the img src=, but how do I do this?

[penagate]

You can remove the <img> tags as follows:

Private Sub WB1_DocumentComplete(ByVal pDisp As Object, URL As Variant)
    Dim source As String
 
    source = WB1.Document.body.parentElement.innerHTML
 
    Do
        Dim tagPos As Long, tagEnd As Long
 
        tagPos = InStr(1, source, "<img")
        If (tagPos) Then
            tagEnd = InStr(tagPos, source, ">")
 
            If (tagEnd) Then
                Mid$(source, tagPos, 4) = "<!--"
                Mid$(source, tagEnd - 2, 3) = "-->"
                tagPos = 0
            End If
        End If
 
    Loop While (tagPos > 0)
 
    WB1.Document.Clear
    WB1.Document.write source
End Sub

Hacky I know, but still...

[penagate]

On second thoughts, you may want to use an Inet control to download the page source first, run it through my parsing loop above, then feed it to the WebBrowser control AFTER the img tags have been removed. Because my code above probably doesn't stop them from being downloaded.

You can feed HTML source in to the control by navigating to "about:blank" and then using the .Document.Write method, passing the parsed HTML source code.

[shirazamod]

Thanx penagate, but I dont think that'll work 4 me as I've already developed my entire app and I'll have to change it to make this work.
Is there any other way to do it? Can't I prevent it from downloading anything from certain URLs?

[penagate]

OK, Mr. HackySolution™ to the rescue

Change "WB1" to the name of your browser control. Add an Internet Transfer Control to your form and change "Inet1" to its name.

You can then use the WebBrowser just as you had been doing before - the added code doesn't affect how you call it or anything, it just strips images from the HTML source as you go.

Private nextURL As String
Private inParse As Boolean
 
Private Sub Form_Load()
    WB1.Navigate2 "www.vbforums.com"
End Sub
 
Private Sub WB1_BeforeNavigate2(ByVal pDisp As Object, URL As Variant, Flags As Variant, TargetFrameName As Variant, PostData As Variant, Headers As Variant, Cancel As Boolean)
    If ((pDisp Is WB1.Object) And (Not inParse)) Then _
        nextURL = CStr(URL)
End Sub
 
Private Sub WB1_DownloadBegin()
    Static parsed As Boolean
    
    If (Not parsed) Then
        inParse = True
 
        Dim buffer As String
        WB1.Stop
        buffer = Inet1.OpenURL(nextURL)
 
        Do
            Dim tagPos As Long
            Dim srcPos As Long, srcEnd As Long
 
            tagPos = InStr(tagPos + 1, buffer, "<img")
            If (tagPos) Then
                srcPos = InStr(tagPos, buffer, "src=""") + 5
                If (srcPos) Then
                    srcEnd = InStr(srcPos, buffer, """")
                    
                    If (srcEnd) Then
                        Mid$(buffer, srcPos, srcEnd - srcPos) = Space$(srcEnd - srcPos)
                    End If
                End If
            End If
 
        Loop While (tagPos > 0)
 
        parsed = True
        WB1.Navigate2 "about:blank"
        WB1.Document.write buffer
 
        inParse = False
    End If
 
    nextURL = vbNullString
End Sub

Shout out if you want me to explain any of the code for ya

[shirazamod]

wow, thanx penagate!
That looks great. I'll try it out after I have dinner (abt 30 min) & let you know how it goes.

[penagate]

I've made some changes. It now replaces the src attribute of every <img> tag with spaces (of the same length as the original address, so that the buffer string is not re-created - faster this way). I also fixed the unfortunate infinite loop

The downside is that you will get Javascript error messages on pages that use JS to manipulate the images, because the images no longer exist. And it is a bit slow. A better alternative might be to have a small 1x1 pixel GIF image in the application directory and replace the src attribute of all images with the address of that GIF file (e.g. "file://" & App.Path & "/blank1x1.gif"). That would probably fix the problem with JS error messages.

[penagate]

OK, encountered more problems - relative links don't work because of the need to navigate to "about:blank" in order to dump in the source - hence the relativity is broken.

[shirazamod]

That's fine, I will be navigating to a page with lyrics so I will only need to get the text from the page and that is all