How to make a nongreedy RegEX expression?

I'm using regular expressions 5.5 and trying to javascripts that match a particular pattern. No problem doing that......it's just this VBScripting is so friggin greedy and when a match is found ALL scripts are removed.

What I need is some way to stop at the first </script> tag! Here is code I put together to illustrate:

-------------code-------------------
' Remove line breakes to create one line instead
' of multi lines

regEX.Pattern = "\n|\s+|\t"
sHTML = regEX.Replace(sHTML, " ")

' define the pattern I am matching
regEX.Pattern = "<script.*?>.*?</script>"
Set Matches = regEX.Execute(sHTML)   ' Execute search.
 
  For Each Match In Matches
If InStr(1, Match, "openwindow", vbTextCompare) = 0 Then
  sHTML = regEX.Replace(sHTML, "")
End If
   Next

-----------cut--------------

Like I said, above works fine if it wasn't for the fact it is tough as nails getting VBScript to stop matching at the first </script> tag like you easily can in perl. Here is 300 points for a solution!



LVL 2
BiffoAsked:
Who is Participating?
 
clockwatcherConnect With a Mentor Commented:
I wouldn't really call that a greedy match.  That's global matching.  If that's what you're getting then you've got global set on your regular expression.  Typically, the following is considered a greedy match.
 
  this <script>is</script> an example of a <script>greedy </script> match

leaving you with:

  this  match

Since you aren't trying to do a global replace (and you're not just trying to replace the first match -- which would be easy), you'll need something like the following -- which is very similar to what we've already posted.  

html = "<head>" & vbcrlf & _
    "<title>Sample</title>" & vbcrlf & _
    "<script language=""JavaScript1.1"" type=""text/javascript"">" & vbcrlf & _
    "<!--" & vbcrlf & _
    "function whatever... { " & vbcrlf & _
    "//-->" & vbcrlf & _
    "</script>" & vbcrlf & _
    "</head>" & vbcrlf & _
    "<body>" & vbcrlf & _
    "<SCRIPT Language=""Javascript"">" & vbcrlf & _
    "more script whatever" & vbcrlf & _
    "</SCRIPT>" & vbcrlf & _
    "<script language=""JavaScript1.1"">" & vbcrlf & _
    "<!--" & vbcrlf & _
    "openwindow.....stuff here" & vbcrlf & _
    "//-->" & vbcrlf & _
    "</script>" & vbcrlf & _
    "<script language=""JavaScript1.1"" src=""some menu perhaps""></script>" & vbcrlf & _
    "</body>" & vbcrlf & _
    "</html>" & vbcrlf

wscript.echo "BEFORE: " & vbcrlf & vbcrlf & html

set r = new RegExp
r.pattern = "<script(?:.|\n)*?</script>"
r.ignorecase = true
r.multiline = true
r.global = true
set matches = r.execute(html)

r.pattern = "openwindow"

for each m in matches
  if r.test(m) then html=replace(html,m,"")
next

wscript.echo "AFTER: " & vbcrlf & vbcrlf & html
0
 
rpaiCommented:
Suppose 'sGivenString' is the string  you have and you wish to stop matching at the first <script> tag.

sGivenString = "XXX YYY AAA <script> AAA BBB CCC DDD </script>"
s = Mid(sGivenString, 1, InStr(1, sGivenString, "<script>", vbTextCompare) - 1)
s = Replace(s, "AAA", "ZZZ", 1, , vbTextCompare)
Debug.Print s

Is this something that you are looking for?
0
 
rpaiCommented:
The above code would only replace the 'AAA' before the <script> tag with 'ZZZ'and not the one that exists within the script tag.
0
Never miss a deadline with monday.com

The revolutionary project management tool is here!   Plan visually with a single glance and make sure your projects get done.

 
BiffoAuthor Commented:
I want to replace the entire script...everything in between <script>...</script> including the tags and globally in case there is more than one script with that matches my search pattern.



0
 
rpaiCommented:
So maybe something like might help:-

sGivenString = "XXX YYY AAA <script> AAA BBB CCC DDD </script>"
s = Mid(sGivenString, InStr(1, sGivenString, "<script>", vbTextCompare) - 1, InStr(1, sGivenString, "</script>", vbTextCompare) - 1 )
s = Replace(s, s, "", 1, , vbTextCompare)
Debug.Print s
0
 
clockwatcherCommented:
For Each Match In Matches
If InStr(1, Match, "openwindow", vbTextCompare) = 0 Then
 sHTML = replace(sHTML,match,"")
End If
0
 
BiffoAuthor Commented:
All this isn't really helping me with my primary regex matching problems! And that problem is stop matching at the first </script> tag and not the last one encountered in the doc.

so I want change this:

---------------------------------------------
<head>
<title>Sample</title>
<script language="JavaScript1.1" type="text/javascript">
<!--

function whatever... {
//-->
</script>
</head>
<body>

<SCRIPT Language="Javascript">
more script whatever
</SCRIPT>

<script language="JavaScript1.1">
<!--
openwindow.....stuff here
//-->
</script>
<script language="JavaScript1.1" src="some menu perhaps"></script>
</body>
</html>
------------------------------------------------------------

To look like this:

-----------------------------------------------------
<head>
<title>Sample</title>
<script language="JavaScript1.1" type="text/javascript">
<!--

function whatever... {
//-->
</script>
</head>
<body>

<SCRIPT Language="Javascript">
more script whatever
</SCRIPT>

<script language="JavaScript1.1" src="some menu perhaps"></script>
</body>
</html>
---------------------------------------------------

And not like this which is the way current regex is doing it:

--------------------------------------------------
<head>
<title>Sample</title>
</head>
<body>


</body>
</html>
---------------------------------------------------


0
 
BiffoAuthor Commented:
Yah, clockwatcher, that seems to be my solution there. I was getting ready to split up the html and place scripts all on their own one line and then just match to end of line to keep from matching too much.

Job well done and here is your points......
0
All Courses

From novice to tech pro — start learning today.