Trying to replace strings in an HTML document and adding extra characters

Question

What I'm trying to do is open up a file and search for "searchText". I want to replace all the instances of it in the file with a new link, which is actually just the filename with an achor link so instead of opening up javascript it just goes to another point in the page.

So far what I have is this:

private void writeNotes(){
        StreamReader reader = new StreamReader(openFileDialog1.FileName);
        string content = reader.ReadToEnd();

        reader.Close();

        string fileName = openFileDialog1.SafeFileName;
        string searchText = "<a class=\"x-fn\" href=\"javascript:void(0);\">";
        string replaceText = "<a class=\"x-fn\" href=\"" + fileName + "#fn" + "\">";

        content = Regex.Replace(content, searchText, replaceText);

        StreamWriter writer = new StreamWriter(openFileDialog1.FileName);
        writer.Write(content);
        writer.Close();

However, after writing and closing... I open up the file and no changes were made. Besides that, what I want to do is add a number that counts up after "#fn" for every instance of replacement. So, basically, for every time I replace the javascript link with another, I want it to be:

<a class="x-fn" href="fileName#fn1">

And then when I replace the second instance of javascript, it reads

<a class="x-fn" href="fileName#fn2">

and so on...

I imagine I would have to count the instances of how many times the javascript appears, replace it, and use a for loop to iterate throughout all of the new links and add the #fn(n) at the end?

You know you can use HtmlAgilityPack for HTML editing as well, right? — Marcel N.
– Marcel N., Commented Jul 9, 2012 at 18:20
Why are you using regex? A simple string replace will do.. especially since your search string is NOT a regex!!! — banging
– banging, Commented Jul 9, 2012 at 18:28
use the debugger. Maybe your regex isn't finding a match. I'd also suggest using string replace as banging said. Don't have to worry about special regex symbols that way. — Sam I am says Reinstate Monica
– Sam I am says Reinstate Monica, Commented Jul 9, 2012 at 18:31

Ethan Brown · Accepted Answer · 2012-07-10 19:47:49Z

2

Your problem is that you have metacharacters (in this case, the open and close paren in "void(0)"). Instead of representing literal open and close parens, that is creating a regular expression group, which is causing your match to fail. If you escape the parens with a backslash, it will work as expected.

However, since you're just matching a string literal, you don't need to use regular expressions at all; it'll be faster to use string.Replace.

As for the second part of your question, that's a little trickier. There's no easy way to do it that I know of, so the best approach is to look for your search text in the input, and then build up a StringBuilder as you go along, incrementing a count variable. In the following example, the word "the" is replaced by "(0)" and "(1)" for simplicity's sake, but you can adapt it to your problem easily enough.

var content = "the quick brown fox jumped over the lazy dog";
var searchText = "the";
var sb = new StringBuilder();
var idx = -1; // will hold the index of our search text
var uncopiedIdx = 0; // the start index of what hasn't been copied yet
var replacementCount = 0;
while( (idx = content.IndexOf( searchText, idx+1 )) != -1 ) {
    // copy everything leading up to our search text
    sb.Append( content.Substring( uncopiedIdx, idx-uncopiedIdx ) );
    // copy the replacement text, with the replacement count
    var replacement = "(" + replacementCount++ + ")";
    sb.Append( replacement );
    // skip over the search text
    uncopiedIdx = idx + searchText.Length;
}
// copy everything after the last match
sb.Append( content.Substring( uncopiedIdx ) );

edited Jul 10, 2012 at 19:47

answered Jul 9, 2012 at 18:30

Ethan Brown

27.4k5 gold badges84 silver badges95 bronze badges

Sign up to request clarification or add additional context in comments.

7 Comments

Servy Over a year ago

You can also use Regex.Escape to ensure that a string is treated as all literals in a regular expression.

JMarsh Over a year ago

Alright, I'm removed Regex and instead just used string.Replace. However, after I run it, the changes still aren't being made. Am I saving it incorrectly by chance? Thanks for your help!

Ethan Brown Over a year ago

You're opening the file with SafeFileName, which does not include the path, but then you're saving it with FileName, which does include the path. So you may be saving the file in a different location from which you're opening it....

JMarsh Over a year ago

Hey buddy, it's working fine now! But still, do you have any idea how to add a number at the end of each string that is replaced that counts up for every instance?

Ethan Brown Over a year ago

Glad to hear it's working. Your second problem, though, is going to require a little more work. Let me think about it and update my answer.

|

user557597 · Accepted Answer · 2012-07-09 20:14:13Z

You could use the MatchEvaluator which should do it all at once.
Something like this -

C#

string content = 
 @"
     <a class=""x-fn"" href=""javascript:void(0);"">
     <a class='x-fn' href = ""javascript:void(0); "">
     <a href='javascript:void(0);' class=x-fn >
     <a class=""x-fn"" href=javascript:void(0); >
     <a 'hello' href=javascript:void(0); world class=x-fn >
  ";
 string fileName = "FILE";

 Regex jsRx =  new Regex(
  @"
     <a 
         (?=\s) 
         (?= 
             (?: [^>""']|""[^""]*""|'[^']*')*? (?<=\s)
             class \s*=
             (?:
                 (?> \s* (['""])  \s* x-fn  \s* \1 )         # (1) quote
               | (?> (?!\s*['""]) \s* x-fn  (?=\s|>)   )
             )
         )
         (?= 
             ( (?: [^>""']|""[^""]*""|'[^']*')*? ) (?<=\s)       # (2) - before 'href'
             href \s*=
             (?:
                 (?> \s* (['""])  \s* javascript:void\(0\); \s* \3   )   # (3) quote
               | (?> (?!\s*['""]) \s* javascript:void\(0\); (?=\s|>) )
             )
             ( (?> (?:"".*?""|'.*?'|[^>]?)+ ) )                  # (4) - after 'href'
         )
         (?> \s+ (?:"".*?""|'.*?'|[^>]*?)+ 
     >        
         ) (?<! /> )
  ", RegexOptions.IgnoreCase | RegexOptions.IgnorePatternWhitespace | RegexOptions.Singleline);

 int counter = 0;
 string contentNew = jsRx.Replace( content, delegate(Match match)
 {
     return "<a" + match.Groups[2] + "href=\"" + fileName + "#fn" + (counter++) + "\"" + match.Groups[4] + ">";
 });

 Console.WriteLine( contentNew );

Output

<a class="x-fn" href="FILE#fn0">
<a class='x-fn' href="FILE#fn1">
<a href="FILE#fn2" class=x-fn >
<a class="x-fn" href="FILE#fn3" >
<a 'hello' href="FILE#fn4" world class=x-fn >

Your welcome! Its a solution thats taken a year to perfect. No longer an idea.

Collectives™ on Stack Overflow

Trying to replace strings in an HTML document and adding extra characters

2 Answers 2

7 Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

7 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related