Hi there,

I've signed up on this forum in the hope that one of the Reg Experts can help me though a problem.  I'm working on a site which is an exchange for MindManager maps. Users can upload their own maps and enter in title, description and tags etc.

I am currently using a regular expression to remove all traces of HTML from the desciption field before saving to the db when a user adds a new map. This works just fine, however, I would like people to be able to include a hyperlink or two in the description if they so desire.

My goal is to remove all attributes other than:

• href
• target

Target must for forced to "_blank".

I don't want any onclick, class, id etc attributes left in.

Here's an example:

User input - <a href="http://www.mycoolsite.com" onclick="sneakyFunction()" name="myLink" class="clsLink">Visit my cool site</a> 

Becomes -  <a href="http://www.mycoolsite.com" target="_blank">Visit my cool site</a>

I plan to use the RegExp replace function to achieve this (provided I can sort out the pattern!) 

I am using ASP VBScript on this site.

If anyone could offer some advice on this I would be most grateful!

All the best,

Nick

<%
function sanitizehtml(ByVal html)
  Set regEx = New RegExp
  regEx.Global = True
  regEx.IgnoreCase = True
  regEx.Pattern = "<[^>]*>|[\S\s]*?(?=<[^>]*>)"
  Set Matches = regEx.Execute(html)
  For z = 0 to Matches.Count-1
    regEx.Pattern = "^<[^>]*>$"
    If regEx.Test(Matches(z)) then
      regEx.Pattern = "^<a\s+[^>]*(href=(['""])[\S\s]*?\2)"
      If regEx.Test(Matches(z)) then
        Set Matches2 = regEx.Execute(Matches(z))
        text = "<a " & Matches2(0).SubMatches(0) & ">"
      Else
        regEx.Pattern = "^<\s*/a>$"
        If regEx.Test(Matches(z)) then
          text = Matches(z)
        Else
          text = ""
        End if
      End if
    Else
      text = Matches(z)
    End if
    sanitizehtml = sanitizehtml & text
  Next
end Function

input="text0<a test> text1 <a href=""http://www.mycoolsite.com"" onclick=""sneakyFunction()"" name=""myLink""

class=""clsLink"">Visit my cool site</a> text00 text01 <sometag> text2"
response.write Server.HTMLEncode(sanitizehtml(input))
%>

In your code, you wouldn't use:

response.write Server.HTMLEncode(sanitizehtml(input))

you would just use:

yourvar = sanitizehtml(input)

The HTMLEncode was just to display what the code is doing to the string.

Hi ddrudik,

This is great! You sir - are a God. 

I made one tweak to add in the target="_blank" attribute. Its done. I owe you a pint.

Many thanks,

Nick

Dear God (ddrudik)

I need to do something very similar to Nick but am having difficulty adapting your code.  I wonder if you could cast your knowledgable eyes over my requirements?

Input string
<div>test</div><p>Paragraph text <a href="someurl.com">link text</a></p>

Required output string
<div>test</div><p>Paragraph text link text</p>

So basically I need to remove the start and end anchor tags but leave the link text and any other html intact.

Thanks in advance for any help you can give me

Rob

Rob,

Sorry I'm not God/ddrudik, but I have a suggestion: try using the pattern

<a\s+[^>]*>([^<]*)</a>

with the replacement string of

$1

and the 'ignore case' option set if you are ever likely to have the tag name in upper case.

If you want to be more specific and only get tags with the 'href' attribute in them somewhere, then use the pattern:

<a\s+(?:(?=.*?href)[^>]*)>([^<]*)</a>

Susan

Thanks so much Susan, that works a treat.