正则表达式从字符串中替换第n个元素并添加自定义HTML标记

Question

I have a list of words "this", "be able", "it" that I want to find inside a paragraph so I can replace preserving their capitalization.

Having this paragraph:

This is my text and this is why I want to match it! As this is just a text, I would like to be able to solve it. This is the final phrase of this paragraph.

"this" is found 5 times and if I decide to replace the 4th one ( "This" ) I want to still be able to keep the T capital. Now you will see that's not actually a replace but more of an adding problem as the actual replace would be from this to This

so my final paragraph would be:

This is my text and this is why I want to match it! As this is just a text, I would like to be able to solve it. This is the final phrase of this paragraph.

My code so far:

    List<string> words = new List<string>(new string[] { "this", "be able", "it"});
    var paragraph = "This is my text and this is why I want to match it! As this is just a text, I would like to be able to solve it. This is the final phrase of this paragraph.";
    //List<string> 
    for (int w = 0; w < words.Count; w++)
    {
        var foudItems = Regex.Matches(paragraph, @"\b" + words[w] + "\\b", RegexOptions.IgnoreCase);

        if (foudItems.Count != 0)
        {
            Random rnd = new Random();
            int rndWord = rnd.Next(0, foudItems.Count);
            Regex.Replace(paragraph, @"\b" + words[w] + "\\b", "<strong>" + foudItems[rndWord] + "</strong>");
            Console.WriteLine(paragraph);
        }

        //Regex.Replace()
        Console.WriteLine(foudItems[0] + " " + foudItems[1]);
}

The main problem is that I don't know how to replace only the n'th word using regex. Another issue would be the complicated approach in solving this so I'm open to new suggestions.

Answer 1

If you want to replace nth occurrence of something, you can use MatchEvaluator delegate which checks current occurrence index and returns unmodified matched value if index match is not one you want to replace. To track current index you can capture local variable:

int occurrenceToReplace = 4;
int index = 0;
MatchEvaluator evaluator = m => (++index == occurrenceToReplace)
    ? $"<strong>{m.Value}</strong>"
    : m.Value;

text = Regex.Replace(text, @"\bthis\b", evaluator, RegexOptions.IgnoreCase);

---

Now back to your problem - you can write method which wraps nth occurrence of given word into html tag:

private static string MakeStrong(string text, string word, int occurrence)
{
    int index = 0;
    MatchEvaluator evaluator = m => (++index == occurrence)
         ? $"<strong>{m.Value}</strong>"
         : m.Value;
    return Regex.Replace(text, $@"\b{word}\b", evaluator, RegexOptions.IgnoreCase);
}

And if you want to randomly replace one of the occurrences of each word, then just use this method in a loop:

string[] words = { "this", "be able", "it"};   
var paragraph = @"This is my text and this is why I want to match it! As this is just
a text, I would like to be able to solve it. This is the final phrase of this paragraph.";

var random = new Random();
foreach(var word in words)
{
    int count = Regex.Matches(paragraph, $@"\b{word}\b", RegexOptions.IgnoreCase).Count;
    int occurrence = random.Next(count + 1);
    paragraph = MakeStrong(paragraph, word, occurrence);
}

Sample output:

This is my text and this is why I want to match it ! As this is just a text, I would like to be able to solve it. This is the final phrase of this paragraph.

Answer 2

If you want to keep the regex side quite simple, you can use this algo:

List<string> words = new List<string>(new string[] { "this", "be able", "it" });
var paragraph = "This is my text and this is why I want to match it! As this is just a text, I would like to be able to solve it. This is the final phrase of this paragraph.";
//List<string> 
foreach (string word in words)
{
    var foundItems = Regex.Matches(paragraph, @"\b" + word + @"\b", RegexOptions.IgnoreCase);
    if (foundItems.Count != 0)
    {
        var count = 0;
        var toReplace = 3;
        foreach (Match foudItem in foundItems)
        {
            count++;
            if(count != toReplace)
                continue;

            var regex = $"(^.{{{foudItem.Index}}}){foudItem.Value}(.*)";
            paragraph = Regex.Replace(paragraph, regex, $"$1<strong>{foudItem.Value}</strong>$2");
        }
        Console.WriteLine(paragraph);
    }
    Console.WriteLine(foundItems[0] + " " + foundItems[1]);
}