/// <summary>
/// Posts the process data.
/// </summary>
/// <param name="page">The page.</param>
/// <returns></returns>
private ScrapedPage PostProcessData(ScrapedPage page)
{
if (FollowEquivRefreshes)
{
bool followed = false;
// See if we can find an http-equiv refresh
List<ScreenScraperTag> metaTags = ScrapedPage.ConvertToTagList(page.FindChildlessTags("meta", false), true);
// Now, we have all META tags. Try to find one with HTTP-EQUIV="refresh"
ScreenScraperTag refreshTag = null;
foreach (ScreenScraperTag metaTag in metaTags)
{
string httpEquivValue = metaTag.FindAttributeValue("http-equiv");
if (httpEquivValue != null && httpEquivValue.Equals("refresh"))
{
refreshTag = metaTag;
break;
}
}
if (refreshTag != null)
{
// It's a refresh. Try to figure out the URL we have to go to.
string contentValue = refreshTag.FindAttributeValue("content");
if (contentValue != null)
{
// First, split it by semicolon
string[] refreshPieces = contentValue.Split(';');
string url = null;
int time = 0;
foreach (string refreshPiece in refreshPieces)
{
if (refreshPiece.ToLower().Trim().StartsWith("url"))
{
// found the URL. Just take everything after the =
int equalPos = refreshPiece.IndexOf('=');
if (equalPos != -1)
{
url = refreshPiece.Substring(equalPos + 1).Trim();
break;
}
}
else if (time == 0)
{
int.TryParse(refreshPiece.Trim(), out time);
}
}
if (time == 0 && url != null)
{
// We have a refresh url, so we need to update the page
// If it is a relative url, make it absolute.
if (!Uri.IsWellFormedUriString(url, UriKind.Absolute))
{
url = LastProcessResponseUri.GetLeftPart(UriPartial.Authority) + "/" + (url.StartsWith("/") ? url.Substring(1) : url);
}
if (!url.Equals(LastMetaFollow))
{
page = Scrape(MetaRefreshScrapeType == null ? page.ScrapeType : MetaRefreshScrapeType.Value, url, page.QueryParameters);
LastMetaFollow = url;
followed = true;
}
else
{
throw new Exception("Appears to be a recursive loop of http-equiv redirects to the same page ('" + url + "').");
}
}
}
}
if (!followed)
{
LastMetaFollow = null;
}
}
return page;
}
/// <summary>
/// Scrapes the specified type.
/// </summary>
/// <param name="type">The type.</param>
/// <param name="uri">The URI.</param>
/// <param name="query">The query.</param>
/// <returns></returns>
public ScrapedPage Scrape(ScrapeType type, string uri, NameValueCollection query)
{
ScrapedPage page = new ScrapedPage();
string qs = BuildQueryString(query);
page.QueryParameters = query;
page.ScrapeType = type;
switch (type)
{
case ScrapeType.GET:
uri = uri.Contains("?") ? (uri + "&" + qs) : (uri + "?" + qs);
page.RawStream = HttpGet(uri);
break;
case ScrapeType.POST:
page.RawStream = HttpPost(uri, qs);
break;
default:
throw new NotImplementedException();
}
if (page.RawStream == null)
{
throw new Exception("No data for " + uri);
}
else
{
page.Url = new Uri(uri);
Referer = uri;
page = PostProcessData(page);
}
return page;
}