/// <summary>
/// Creates an array of links based on the post's body. Then walks through each link and attempts to send trackbacks/pingbacks
/// </summary>
public void CheckPost(object state)
{
List <string> links = GetLinks(_postBody, _baseUrl);
foreach (string externalUrl in links)
{
if (string.IsNullOrEmpty(externalUrl))
{
continue;
}
try
{
WebHeaderCollection headers = null;
string pageText = null;
using (HttpWebResponse response = GRequest.GetResponse(externalUrl, _postLink))
{
headers = response.Headers;
pageText = GRequest.GetPageText(response);
response.Close();
}
if (!string.IsNullOrEmpty(pageText))
{
bool pingSent = false;
// First try to send a TrackBack if the required RDF info was embedded in the HTML
TrackBackSender trackBack = new TrackBackSender(pageText, externalUrl, _siteName, _postTitle, _postLink, _excerpt);
pingSent = trackBack.SendTrackBackPing();
// If the Trackback attempt failed, try to do a Pingback
// (if the required PingBack XML-RPC service URL exists in the HTTP Header or HTML Head)
if (!pingSent)
{
PingBackSender pingBack = new PingBackSender(headers, pageText, externalUrl, _postTitle, _postLink);
pingSent = pingBack.SendPingbackPing();
}
}
}
catch (System.Exception ex)
{
string message = String.Format("Trackback/Pingback attempt to the url [{0}] failed for post {1} while retrieving the remote document. Error message returned was: {2}.", externalUrl, this._postTitle, ex.Message);
Log.Warn("Trackback/Pingback Error", message);
}
}
}
/// <summary>
/// Retrieves a remote html page and determines if it contains a link to a given target url. Also gets the page title.
/// </summary>
internal static bool SourceContainsTarget(string sourceURI, string targetURI, out string pageTitle)
{
pageTitle = string.Empty;
Uri baseUri = new Uri(sourceURI);
string baseUrl = string.Format("{0}://{1}{2}", baseUri.Scheme, baseUri.Host, baseUri.Port == 80 ? string.Empty : ":" + baseUri.Port.ToString());
string page = null;
HttpWebRequest request = GRequest.CreateSafeRequest(sourceURI, targetURI);
request.MaximumAutomaticRedirections = 3;
using (HttpWebResponse response = request.GetResponse() as HttpWebResponse)
{
if (response != null)
{
if (GRequest.IsContentTypeValid(response.ContentType, GRequest.ContentTypeCategory.HTML) && response.ContentLength < 300000)
{
page = GRequest.GetPageText(response);
}
response.Close();
}
}
if (string.IsNullOrEmpty(page))
{
return(false);
}
if (page.IndexOf(targetURI) < 0)
{
// The direct link to the Graffiti post was not found in the body of the source post
// BUT, the source may have linked to a url that redirected to the Graffiti post (i.e. Feedburner url)
List <string> links = LinkParser.GetLinks(page, baseUrl);
bool targetLinkFound = false;
foreach (string externalUrl in links)
{
if (string.IsNullOrEmpty(externalUrl))
{
continue;
}
try
{
string responseUrl = null;
HttpWebRequest request2 = GRequest.CreateSafeRequest(externalUrl, targetURI);
request2.MaximumAutomaticRedirections = 3;
using (HttpWebResponse response2 = request2.GetResponse() as HttpWebResponse)
{
if (response2 != null)
{
if (GRequest.IsContentTypeValid(response2.ContentType, GRequest.ContentTypeCategory.HTML) && response2.ContentLength < 300000)
{
responseUrl = response2.ResponseUri.ToString();
}
response2.Close();
}
}
if (responseUrl != null && string.Compare(targetURI, responseUrl, StringComparison.InvariantCultureIgnoreCase) == 0)
{
targetLinkFound = true;
break;
}
}
catch { }
}
if (!targetLinkFound)
{
return(false);
}
}
// Look for the HTML Title of the remote page
string pat = @"<head.*?>.*<title.*?>(.*)</title.*?>.*</head.*?>";
Regex reg = new Regex(pat, RegexOptions.IgnoreCase | RegexOptions.Singleline);
Match m = reg.Match(page);
if (m.Success)
{
pageTitle = HttpUtility.HtmlDecode(Graffiti.Core.Util.RemoveHtml(m.Result("$1").Replace("\r", "").Replace("\n", "").Replace("\t", "").Trim(), 250));
}
return(true);
}