//The Content Type supported by the parser
#endregion
#region Constructor and Singleton Instance members
/// <summary>
/// The constructor is private so that only the class itself can create an instance.
/// </summary>
private TextParser()
{
//Initialize the synchronization mechanism
mutex = new Mutex();
//Initialize the Regular Expressions
hrefRegex = new Regex(@"(http|https)://[\w]+(\.[\w]+)+([\w\-\.,@?^=%&:/~\+#]*[\w\-\@?^=%&/~\+#])?", RegexOptions.CultureInvariant | RegexOptions.Multiline | RegexOptions.IgnoreCase | RegexOptions.Compiled);
//use "(http|ftp|https)://[\w]+(\.[\w]+)+([\w\-\.,@?^=%&:/~\+#]*[\w\-\@?^=%&/~\+#])?" to enable ftp urls
sessionIDRegex = new Regex(@"([0-9a-fA-F]{40,64})|([\{|\(]?[0-9a-fA-F]{8}[-]?([0-9a-fA-F]{4}[-]?){3}[0-9a-fA-F]{12}[\)|\}]?)$", RegexOptions.CultureInvariant | RegexOptions.Multiline | RegexOptions.IgnoreCase | RegexOptions.Compiled); //@"^([0-9a-f]{32})|(\{?[0-9a-f]{8}-([0-9a-f]{4}-){3}-[0-9a-f]{12}\}?)$"
spacesRegex = new Regex(@"\s+", RegexOptions.CultureInvariant | RegexOptions.Multiline | RegexOptions.IgnoreCase | RegexOptions.Compiled);
//Initialize the filters
robotsFilter = RobotsFilter.Instance();
domainFilter = DomainFilter.Instance();
//Get a reference to the global variables and application settings
globals = Globals.Instance();
}
