public static string ObtainImageInfo(Uri url, string text)
{
try
{
var client = new CookieWebClient();
client.Headers[HttpRequestHeader.UserAgent] = FakeUserAgent;
// alibi-visit the image search page to get the cookies
client.Headers[HttpRequestHeader.Referer] = GoogleHomepageUrl.ToString();
client.DownloadData(GoogleImageSearchUrl);
// fetch the actual info
var searchUrl = new Uri(string.Format(
GoogleImageSearchByImageUrlPattern,
Util.UrlEncode(url.ToString(), Util.Utf8NoBom, true)
));
client.Headers[HttpRequestHeader.Referer] = GoogleImageSearchUrl.ToString();
var responseBytes = client.DownloadData(searchUrl);
var parseMe = EncodingGuesser.GuessEncodingAndDecode(responseBytes, null, null);
var htmlDoc = new HtmlDocument();
htmlDoc.LoadHtml(parseMe);
var foundHints = htmlDoc.DocumentNode.QuerySelectorAll(".qb-bmqc .qb-b");
foreach (var hint in foundHints)
{
return(string.Format("{0} ({1})", text, HtmlEntity.DeEntitize(hint.InnerText)));
}
return(text);
}
catch (Exception ex)
{
Logger.Warn("image info", ex);
return(text);
}
}
public static string RealObtainLinkInfo(Uri link)
{
var lowerUrl = link.ToString().ToLowerInvariant();
if (!lowerUrl.StartsWith("http://") && !lowerUrl.StartsWith("https://"))
{
return("(I only access HTTP and HTTPS URLs)");
}
// check URL blacklist
var addresses = Dns.GetHostAddresses(link.Host);
if (addresses.Length == 0)
{
return("(cannot resolve)");
}
if (addresses.Any(IPAddressBlacklist.IsIPAddressBlacklisted))
{
return("(I refuse to access this IP address)");
}
var request = WebRequest.Create(link);
using (var respStore = new MemoryStream())
{
var contentType = "application/octet-stream";
string contentTypeHeader = null;
string responseCharacterSet = null;
request.Timeout = 5000;
try
{
var resp = request.GetResponse();
// find the content-type
contentTypeHeader = resp.Headers[HttpResponseHeader.ContentType];
if (contentTypeHeader != null)
{
contentType = contentTypeHeader.Split(';')[0];
}
var webResp = resp as HttpWebResponse;
responseCharacterSet = (webResp != null) ? webResp.CharacterSet : null;
// copy
resp.GetResponseStream().CopyTo(respStore);
}
catch (WebException we)
{
var httpResponse = we.Response as HttpWebResponse;
return(string.Format("(HTTP {0})", httpResponse != null ? httpResponse.StatusCode.ToString() : "error"));
}
switch (contentType)
{
case "application/octet-stream":
return("(can't figure out the content type, sorry)");
case "text/html":
case "application/xhtml+xml":
// HTML? parse it and get the title
var respStr = EncodingGuesser.GuessEncodingAndDecode(respStore.ToArray(), responseCharacterSet,
contentTypeHeader);
var htmlDoc = new HtmlDocument();
htmlDoc.LoadHtml(respStr);
var titleElement = htmlDoc.DocumentNode.SelectSingleNode(".//title");
if (titleElement != null)
{
return(HtmlEntity.DeEntitize(titleElement.InnerText));
}
var h1Element = htmlDoc.DocumentNode.SelectSingleNode(".//h1");
if (h1Element != null)
{
return(HtmlEntity.DeEntitize(h1Element.InnerText));
}
return("(HTML without a title O_o)");
case "image/png":
return(ObtainImageInfo(link, "PNG image"));
case "image/jpeg":
return(ObtainImageInfo(link, "JPEG image"));
case "image/gif":
return(ObtainImageInfo(link, "GIF image"));
case "application/json":
return("JSON");
case "text/xml":
case "application/xml":
return("XML");
default:
return(string.Format("file of type {0}", contentType));
}
}
}
