C# RegEx string extraction

I have a string:

"ImageDimension=655x0;ThumbnailDimension=0x0".

I have to extract first number ("655" string) coming in between "ImageDimension=" and first occurrence of "x" ; and need extract second number ("0" string) coming after first "x" occurring after "ImageDimension=" string. Similar with third and fourth numbers.

Can this be done with regex ("ImageDimension=? x ?;ThumbnailDimension=? x ?") and how ? Instead of clumsy substrings and indexof ? Thank you!

My solution which is not nice :

String configuration = "ImageDimension=655x0;ThumbnailDimension=0x0";
String imageDim = configuration.Substring(0, configuration.IndexOf(";"));
int indexOfEq = imageDim.IndexOf("=");
int indexOfX = imageDim.IndexOf("x");

String width1 = imageDim.Substring(indexOfEq+1, indexOfX-indexOfEq-1);
String height1 = imageDim.Substring(imageDim.IndexOf("x") + 1);

String thumbDim = configuration.Substring(configuration.IndexOf(";") + 1);
indexOfEq = thumbDim.IndexOf("=");
indexOfX = thumbDim.IndexOf("x");

String width2 = imageDim.Substring(indexOfEq + 1, indexOfX - indexOfEq-1);
String height2 = imageDim.Substring(imageDim.IndexOf("x") + 1);

This will get each of the values into separate ints for you:

string text = "ImageDimension=655x0;ThumbnailDimension=0x0";
Regex pattern = new Regex(@"ImageDimension=(?<imageWidth>\d+)x(?<imageHeight>\d+);ThumbnailDimension=(?<thumbWidth>\d+)x(?<thumbHeight>\d+)");
Match match = pattern.Match(text);
int imageWidth = int.Parse(match.Groups["imageWidth"].Value);
int imageHeight = int.Parse(match.Groups["imageHeight"].Value);
int thumbWidth = int.Parse(match.Groups["thumbWidth"].Value);
int thumbHeight = int.Parse(match.Groups["thumbHeight"].Value);

var groups = Regex.Match(input,@"ImageDimension=(\d+)x(\d+);ThumbnailDimension=(\d+)x(\d+)").Groups;
var x1= groups[1].Value;
var y1= groups[2].Value;
var x2= groups[3].Value;
var y2= groups[4].Value;

var m = Regex.Match(str,@"(\d+).(\d+).*?(\d+).(\d+)");
m.Groups[1].Value; // 655 ....

(\d+) 

Get the first set of one or more digits. and store it as the first captured group after the entire match

.

Match any character

(\d+)

Get the next set of one or more digits. and store it as the second captured group after the entire match

.*? 

match and number of any characters in a non greedy fashion.

(\d+)

Get the next set of one or more digits. and store it as the third captured group after the entire match

(\d+)

Get the next set of one or more digits. and store it as the fourth captured group after the entire match

Since a lot of people already gave you what you wanted, I will contribute with something else. Regexes are hard to read and error prone. Maybe a little less verbose than your implementation but more straightforward and friendly than using regex:

private static Dictionary<string, string> _extractDictionary(string str)
{
    var query = from name_value in str.Split(';')   // Split by ;
                let arr = name_value.Split('=')     // ... then by =
                select new {Name = arr[0], Value = arr[1]};

    return query.ToDictionary(x => x.Name, y => y.Value);
}

public static void Main()
{
    var str = "ImageDimension=655x0;ThumbnailDimension=0x0";
    var dic = _extractDictionary(str);

    foreach (var key_value in dic)
    {
        var key = key_value.Key;
        var value = key_value.Value;
        Console.WriteLine("Value of {0} is {1}.", key, value.Substring(0, value.IndexOf("x")));
    }
}