Regex not working in C#

Question

Here is my regex:

href\\s*=\\s*(?:\"(?<1>[^\"]*)\"|(?<1>\\S+))

And here is what I have:

"<p>dfhdfh</p>\r\n<p><a href=\"/Content/blabla/345/344\">najnov</a></p>\r\n<p>&nbsp;</p>\r\n<p><a href=\"/Content/blabla/345/323:test 1\">test 1&nbsp;</a></p>"

But m.Groups are:

{href="/Content/blabla/345/344"}
{/Content/blabla/345/344}

How to get the second href in m?

Here is my code:

Match m = Regex.Match(myString, "href\\s*=\\s*(?:\"(?<1>[^\"]*)\"|(?<1>\\S+))", RegexOptions.IgnoreCase);
                if (m.Success)
                {
                    for (int ij = 0; ij < m.Groups.Count; ij++)
                        myString = myString.Replace(m.Groups[ij].Value.Substring(7), m.Groups[ij].Value.Substring(m.Groups[ij].Value.LastIndexOf("/") + 1));
                }

[You shouldn't try to parse HTML with regexes.](http://stackoverflow.com/a/1732454/41071) Use a HTML parser instead, like HTML Agility Pack. — svick, Apr 20 '12 at 09:47
Also, could you show us your code that actually uses your regex? — svick, Apr 20 '12 at 09:48
Maybe you should explain what you want to achieve, this replace using substring on a capturing group looks quite strange. — stema, Apr 20 '12 at 10:11

score 1 · Accepted Answer · answered Apr 20 '12 at 09:56

From testing this using RAD software RegEx designer.

This regex returns multiple matches, with one group within each match. So you shouldn't be trying to get your result from the Group (named "1"), you should be iterating over the collection of matches and retrieving the value of each (or the group from within each).

This is the result that gets output:

output from RAD RegEx designer

So you should be calling Regex.Matches in your code, and iterate through the results, not Regex.Match.

score 1 · Answer 2 · edited Apr 20 '12 at 09:59

1

Apart from the html/regex stuff, to get all results at once, use Matches, that method returns a MatchCollection that contains all found Match objects.

See The MatchCollection and Match Objects on msdn.

edited Apr 20 '12 at 09:59

svick

236,525
50
385
514

answered Apr 20 '12 at 09:56

stema

90,351
20
107
135

Alan Moore · Answer 3 · 2012-04-20T10:40:24.087

I'm going to assume the original string is this:

<p>dfhdfh</p>
<p><a href="/Content/blabla/345/344">najnov</a></p>
<p>&nbsp;</p>
<p><a href="/Content/blabla/345/323:test 1">test 1&nbsp;</a></p>

..and what you posted is the string literal you would use to create that string. Getting all the href attributes out of that is as simple as this:

Regex r = new Regex(@"href\s*=\s*(?:""(?<HREF>[^""]*)""|(?<HREF>\S+))");

foreach (Match m in r.Matches(htmlString))
{
  Console.WriteLine(m.Groups["HREF"].Value);
}

I changed the name of the capturing group to HREF to make it clear that we're retrieving the group by its name, not by its number.

As you can see, you're doing whole lotta work you don't need to do.

Regex not working in C#

3 Answers3