[RESOLVED] Highlight text in WebBrowser
I'm using a WebBrowser control and I'd like to highlight text that the user enters. It's supposed to be kind of like what you get from Ctrl+F in Firefox. I initially tried this:
Me.webHelp.DocumentText = Me.strOriginal.Replace(Me.txtSearch.Text, "<span style=""background-color:#44F; color:#FFF"">" & Me.txtSearch.Text & "</span>")
However, if the user types "body" it replaces every instance of "body", even ones inside tags, which hoses the HTML. How could I separate the page text from the tags(and everything else) and just do a replace on that while maintaining the integrity of the HTML?
Thanks
Re: Highlight text in WebBrowser
You may have to use regular expressions to pull the text between html tags and do a replace on them, then replace the old tag with the new one in the document.
Try something like this:
Code:
Dim txt2highlight As String = "Yahoo" '<<< this is the text you want to search and highlight in the document"
Dim pattern As String = "<(?<tag>\w*)>[\n\s]*(?<text>.*" & txt2highlight & ".*)[\n\s]*</\k<tag>>"
Dim htmlsrc As String = WebBrowser1.DocumentText()
Dim txt As String = String.Empty
Dim tag As String = String.Empty
For Each m As Match In Regex.Matches(htmlsrc, pattern)
txt = m.Groups("text").Value.Replace(txt2highlight, "<span style=""background-color:#44F; color:#FFF"">" & txt2highlight & "</span>")
tag = m.Value.Replace(m.Groups("text").Value, txt)
htmlsrc = htmlsrc.Replace(m.Value, tag)
Next
WebBrowser1.DocumentText = htmlsrc
Re: Highlight text in WebBrowser
I've been working on other problems, so I've just come back to this one now. Your code almost works. The problem is the last line. When I change the WebBrowser.DocumentText, I get an error on the page. The simplest example I can think of is this. It's just a page with a centered image. When I reset the DocumentText, the image no longer shows(I get a broken link image). Using a msgbox I printed WebBrowser1.DocumentText before doing anything, htmlsrc after the for loop, and WebBrowser1.DocumentText again after it was set. They all displayed exactly the same way and I also printed the String length of each and they all matched up.
Here is the html:
<html>
<head>
<title>This is the title</title>
</head>
<body style="background-color:#FFFFFF;">
<div align="center">
<center>
<table border="0" height="100%">
<tr>
<td>
<br />
<img src="placeholder.png">
</td>
</tr>
</table>
</center>
</div>
</body>
</html>
Like I said, that displays fine, then when I replace .DocumentText with the exact same string, it breaks.
Any thoughts?
Thanks
Re: Highlight text in WebBrowser
The link in my signature about manipulating HTML with the browser control has a "highlight text" feature. You may want to take a look at that and see if you can adapt it to your code. It uses the COM browser though.
Not sure which version of .NET you are using since your sig says 1.1/2003 but your thread title says 2008 which would imply .NET 2.0-3.5
Re: Highlight text in WebBrowser
kleinma,
I knew something was wrong when I went to download that and the link was purple. I've downloaded that before, but it doesn't work with VS 2008, which I've been using for the past year or so(fixed my sig, thanks).
Re: Highlight text in WebBrowser
What problem did you have with 2008? I have used it fine. If you can give me some more info, I can probably tell you what needs to be done to get it working.
Re: Highlight text in WebBrowser
It doesn't open. Or rather, when it does there isn't anything. I get this warning upon opening.
The application for project
'C:\Users\me\Desktop\WebpageManipulation\WebpageManipulation\WebpageManipulation.vbproj' is not installed.
Make sure the application for the project type (.vbproj) is installed.
I've never encountered this error message before. I just chalked it up to incompatibility. I've had problems with some of my older projects converting to 2008, and I figured that was the case.
Re: Highlight text in WebBrowser
There should be a 'sln' file you should be opening, not a vbproj file.
Re: Highlight text in WebBrowser
I'm trying to open the sln. I attached a screenshot.
Re: Highlight text in WebBrowser
I was testing this using a very simply html page that had a centered image and a title. Trying to replace text on the source produced an identical page(so it should look exactly the same). However, the image wouldn't display(just a small empty box).
The image was displayed like this:
<img src="placeholder.png">
So, I got the idea that perhaps when I was setting the DocumentText, it was accidentally throwing away WebBrowser1.Document.Domain. So, I checked the domain right after setting the html, and it was fine. The program then jumped to the Navigating event of the webbrowser. It was still fine. However, once it jumped to the Navigated event, trying to check the Domain in the Immediate Window resulted in this error:
Error HRESULT E_FAIL has been returned from a call to a COM component.
So, I feel like I might be onto something, but I don't know what to do to fix it. Any thoughts anybody?
Thanks
Re: Highlight text in WebBrowser
I'm still having zero luck with this.
Kleinma,
Could you possibly open your project and paste the pertinent code? I don't wanna be a burden, but I can't get that project to open. I just get an error and a blank screen.
Re: Highlight text in WebBrowser
I'm getting closer!
I changed the last line of Stanav's code to read:
WebBrowser1.Document.Body.InnerHtml = htmlsrc
and that did the trick.
However, I think there is a problem with the RegEx string now. It seems that if I type single characters that exist in tags, it performs the operation on them. So, if I type 'a' it'll pick up the 'a' in 'span', and I get some garbage on the screen. If I continue with 'ab' it snaps back to the original page perfectly formatted, but now the 'ab's are highlighted. The RegEx string doesn't seem to like single characters. Anybody know how to fix it?
Thanks
Re: Highlight text in WebBrowser
A co-worker of mine decided to take a crack at it and he decided to ditch RegEx and go with a loop. I edited it after he sent it to me. It's case insensitive, though it wouldn't require much to make it case sensitive. I'm thinking of giving them that option and also giving a "Match Whole Word" option as well. I think that will be pretty easy. Here's the code. Would anyone mind looking over this just to see if I have any glaring mistakes? It seems safe and flaw free from the little bit of testing I've done. Thanks to everyone. Every post helped push me a little closer to the correct answer.
Code:
Public Sub HighlightSearch()
Dim NextOpen As Integer = 0
Dim CurrPos As Integer = 0
Dim HTMLPart As String
Dim NewHTML As String = String.Empty
Dim HTML As String = String.Empty
Dim Index As Integer
Dim strSearchTerm As String = Me.txtSearch.Text
If Me.strOriginal = "" Then Me.strOriginal = Me.webHelp.DocumentText
HTML = Me.strOriginal
If strSearchTerm <> String.Empty Then
While CurrPos < HTML.Length
If HTML.Substring(CurrPos, 6) = "<head>" Then
NewHTML &= HTML.Substring(CurrPos, HTML.IndexOf("</head>", CurrPos) - CurrPos + 7)
CurrPos = HTML.IndexOf("</head>", CurrPos) + 7
ElseIf HTML.Substring(CurrPos, 1) = "<" Then
NewHTML &= HTML.Substring(CurrPos, HTML.IndexOf(">", CurrPos) - CurrPos + 1)
CurrPos = HTML.IndexOf(">", CurrPos) + 1
Else
If HTML.IndexOf("<", CurrPos) <> -1 Then
HTMLPart = HTML.Substring(CurrPos, HTML.IndexOf("<", CurrPos) - CurrPos)
Else
HTMLPart = HTML.Substring(CurrPos)
End If
CurrPos += HTMLPart.Length
While HTMLPart <> String.Empty
If HTMLPart.ToLower.IndexOf(strSearchTerm.ToLower) > -1 Then
Index = HTMLPart.ToLower.IndexOf(strSearchTerm.ToLower)
NewHTML &= HTMLPart.Substring(0, Index) & _
"<span style=""background-color:#44F; color:#FFF"">" & _
HTMLPart.Substring(Index, strSearchTerm.Length) & _
"</span>"
HTMLPart = HTMLPart.Substring(Index + strSearchTerm.Length)
Else
NewHTML &= HTMLPart
HTMLPart = String.Empty
End If
End While
End If
End While
Else
NewHTML = HTML
End If
Me.webHelp.Document.Body.InnerHtml = NewHTML
End Sub
