CODEWITHSUNDEEP
c#
Cindy93
Cindy93

Reputation: 1300

C# error Unrecognized grouping construct

Need help..Why do i get an ArgumentException was Unhandle. the error shows Unrecognized grouping construct. Is my pattern wrong?

   WebClient client = new WebClient();
            string contents = client.DownloadString("http://site.com");

                string pattern =@"<td>\s*(?<no>\d+)\.\s*</td>\s*<td>\s*
                        <a class=""LN"" href=""[^""]*+"" 
                        onclick=""[^""]*+"">\s*+<b>(?<name>[^<]*+)
                        </b>\s*+</a>.*\s*</td>\s*+ 
                        <td align=""center"">[^<]*+</td>
                        \s*+<td>\s*+(?<locations>(?:<a href=""[^""]*+"">[^<]*+</a><br />\s*+)++)</td>";

            foreach (Match match in Regex.Matches(contents, pattern, RegexOptions.IgnoreCase))
            {
                string no = match.Groups["no"].Value;
                string name = match.Groups["name"].Value;
                string locations = match.Groups["locations"].Value;

                Console.WriteLine(no+" "+name+" "+locations);
            }

Upvotes: 1

Views: 1091

Answers (1)

porges
porges

Reputation: 30590

There's no such thing as ?P<name> in C#/.NET. The equivalent syntax is just ?<name>.

The P named group syntax is from PCRE/Python (and Perl allows it as an extension).

You'll also need to remove all nested quantifiers (i.e. change *+ to * and ++ to +). If you want to get the exact same behavior you can switch X*+ to (?>X*), and likewise with ++.

Here is your regex, modified. I've tried to comment it a bit too, but I can't guarantee I did so without breaking it.

new Regex(
@"<td>                   # a td element
    \s*(?<no>\d+)\.\s*   # containing a number captured as 'no'
  </td>\s*
  <td>\s*                # followed by another td, containing
                         # an <a href=... onclick=...> exactly
      <a class=""LN"" href=""(?>[^""]*)"" onclick=""(?>[^""]*)""> 
         (?>\s*)                   # which contains
         <b>(?<name>(?>[^<]*))</b> # some text in bold captured as 'name'
         (?>\s*)
      </a>
      .*                 # and anywhere later in the document
      \s*
  </td>                  # the end of a td, followed by whitespace
  (?>\s*)   
  <td align=""center"">  # after a <td align=center> containing no other elements
    (?>[^<]*)
  </td>
  (?>\s*)
  <td>                   # lastly 
    (?>\s*)
    (?<locations>        # a series of <a href=...>...</a><br/>
        (?>(?:           # captured as 'locations'
            <a href=""(?>[^""]*)"">(?>[^<]*)</a>
            <br />
            (?>\s*)
            )
        +))              # (containing at least one of these)
  </td>", RegexOptions.IgnorePatternWhitespace|RegexOptions.IgnoreCase)

But you really should use something like the HTML Agility Pack.

Upvotes: 1

Related Questions