Find specific pattern in string using C#

I am trying t开发者_运维知识库o find, and remove, a specific pattern inside a string with C#.

The pattern is an asterisk, followed by any number of numbers, followed by .txt

Example strings:

1. test*123.txt
2. test2*1.txt
3. test*1234.txt3
4. test4*12.txt123

Given these examples, the desired results would be:

1. test ("*123.txt" was removed)
2. test2 ("*1.txt" was removed)
3. test3 ("*1234.txt" was removed)
4. test4123 ("*12.txt" was removed)

How can this be accomplished?

---

string pattern = @"\*\d*\.txt";
Regex rgx = new Regex(pattern)
input = rgx.Replace(input, "");

---

If you build a regular expression and replace its matches with an empty string, you're effectively removing that pattern. Here's what you'll need for your pattern:

1. An asterisk has a special meaning in a regular expression (zero or more of the previous item), so you'll have to escape it with a backslash (\*).

2. You can match a digit with the digit character class (\d) or with an explicit class that includes all of them ([0-9]). There are differences between them because of culture settings: \d can match things like eastern arabic numerals (٠.١.٢.٣.٤.٥.٦.٧.٨.٩), while [0-9] will match only the hindu-arabic numerals (0, 1, 2, 3, 4, 5, 6, 7, 8, 9).

3. You can use a + quantifier to match one or more of the previous item: \d+ will match one or more digits.

4. A dot is another special character (it matches any single character except for newlines). It will also need escaping (\.).

5. You can match text without special characters with the text itself: txt matches exactly txt.

Putting everything together we get:

string purged = Regex.Replace(input, @"\*[0-9]+\.txt", "");

---

I would use RegEx to solve this. I recommend an online editor to help you do this. It's called Rubular and can be found at http://www.rubular.com/