正则表达式：句子末尾和URL中的区分句

Question

I am building an app that searches text for urls and adds the urls to a listbox. I have something working however what I have cannot pick up urls if they end the sentence (example: this is www.google.com.). Thanks in advance

Here is my code:

private void btnExtract_Click(object sender, EventArgs e)
        {
            StringBuilder taintedStr = new StringBuilder(txtInputText.Text);
            string cleanStr;

            taintedStr.Replace(",", "");
            taintedStr.Replace("!", "");
            taintedStr.Replace("(", "");
            taintedStr.Replace(")", "");
            taintedStr.Replace("[", "");
            taintedStr.Replace("]", "");
            taintedStr.Replace("http://", "");
            cleanStr = taintedStr.ToString();
            string[] wordlist = Regex.Split(cleanStr, @"\s");

            for (int i = 0; i < wordlist.Length; i++)
            {
                bool test = Regex.Match(wordlist[i], @"^[a-zA-Z0-9\-\.]+\.[a-zA-Z]{2,3}(/\S*)?$").Success;  
                if (test == true)
                {
                    lstWebsites.Items.Add("http://" + wordlist[i]);
                }
            } 
        }

Answer 1

Why not tweak your code by adding a line to remove ending punctuation from each word? For example:

for (int i = 0; i < wordlist.Length; i++)
{
  wordlist[i] = wordlist[i].Trim().TrimEnd('.').TrimEnd('!').TrimEnd('?');
  bool test = Regex.Match(wordlist[i], @"^[a-zA-Z0-9\-\.]+\.[a-zA-Z]{2,3}(/\S*)?$").Success;  
  if (test == true)
  {
    lstWebsites.Items.Add("http://" + wordlist[i]);
  }
} 

Alternatively, the following RegEx should catch the website:

^[a-zA-Z0-9\-\.]+\.[a-zA-Z]{2,3}[.!?]?(/\S*)?$

Answer 2

You'll have to decide how to work this into your code but basically you just want to add a special case for this. ".[a-zA-Z]{2,3}(/\\S*)?$.\\b" will match .*. . If this is the case then use do;

 myString = myString.TrimeEnd('.'); // remove the last character

/b matches on a word boundary. It will match returns, spaces, EOF, ect.

Answer 3

Periods at the end of a sentence are generally followed by whitespace in normal english. But if the period is at the end of a representation of english, it may be followed by other characters such as an EOF character, a "<", a quotation mark, etc.

The way to approach this problem is to recognize when the period is followed by a valid url character.