/* fetchPage() - takes a _SpiderWorkItemDataWrapper object that will be cast from an object
* (because the work method of a C# ThreadPool work item has to take a single
* object argument, and be static/void), and fetches the _SpiderPageCandidate
* at the index specified by the _index field in the _SpiderWorkItemDataWrapper
* @o - the object argument to be cast into a _SpiderWorkItemDataWrapper
*/
static void fetchPage(object o)
{
// unpack the _SpiderWorkItemDataWrapper object
_SpiderWorkItemDataWrapper wi = (_SpiderWorkItemDataWrapper)o;
// get our spider object and our candidate_page to process
Spider spider_object = wi.getSpiderObject();
_SpiderPageCandidate candidate_page = wi.getCandidatePage();
List <string> pre_pages = new List <string>();
byte[] buf = new byte[8192];
StringBuilder sb = new StringBuilder();
HttpWebResponse resp = null;
try {
HttpWebRequest req = (HttpWebRequest)WebRequest.Create(candidate_page._candidate_getUrl());
//req.Timeout = 1000;
// sleep for the niceness time of this spider object
spider_object.acquireFetchLock();
resp = (HttpWebResponse)req.GetResponse();
}
catch (Exception e) {
candidate_page._candidate_setError();
spider_object.writeStatus("ERROR: " + e.Message);
spider_object.writeStatus("\tpage - " + candidate_page._candidate_getUrl() + "\n\t\treferred to by:");
List <SpiderLink> curr_refs = candidate_page.getReferredByLinks();
for (int i = 0; i < curr_refs.Count; i++)
{
spider_object.writeStatus("\t\t\t" + curr_refs.ElementAt(i).getReferringUrl());
}
}
if (resp != null)
{
// record the final Url after any redirects from this link
string normalized_final_url = spider_object.normalizeUrl(resp.ResponseUri.ToString(), "");
if (normalized_final_url.Count() < 1)
{
candidate_page._candidate_setError();
spider_object.writeStatus("fetchPage(): candidate page " + candidate_page._candidate_getUrl() +
" redirected to an illegal page.");
}
candidate_page.setUrl(normalized_final_url);
spider_object.writeStatus("thread id: " + Thread.CurrentThread.ManagedThreadId +
", fetchPage(): fetched " + candidate_page._candidate_getUrl() +
"\n\tfetchPage(): normalized final url - " + candidate_page.getUrl());
if (!candidate_page._candidate_isError())
{
// read in the content of the page
Stream resp_stream = resp.GetResponseStream();
string temp_string = null;
int count = 0;
do
{
count = resp_stream.Read(buf, 0, buf.Length);
if (count != 0)
{
temp_string = Encoding.ASCII.GetString(buf, 0, count);
sb.Append(temp_string);
}
}while (count > 0);
// add the source into the candidate page object
candidate_page.setPageContent(sb.ToString());
// parse the page for links
HtmlDocument doc = new HtmlDocument();
doc.LoadHtml(sb.ToString());
var linksOnPage = from lnks in doc.DocumentNode.Descendants()
where lnks.Name == "a" &&
lnks.Attributes["href"] != null &&
lnks.InnerText.Trim().Length > 0
select new {
Url = lnks.Attributes["href"].Value,
};
foreach (var link in linksOnPage)
{
pre_pages.Add(link.Url);
}
;
// parse out the distinct links on this page, removing any duplicates, and marking illegal links as such
List <string> distinct_pre_pages = pre_pages.Distinct().ToList();
for (int m = 0; m < distinct_pre_pages.Count; m++)
{
string new_url = distinct_pre_pages.ElementAt(m);
SpiderLink new_link = new SpiderLink(new_url, spider_object.normalizeUrl(new_url, candidate_page.getUrl()),
candidate_page.getUrl());
if (new_link.getNormalizedUrl().Count() < 1)
{
new_link.setIllegalLink();
}
candidate_page.addLinkingToLink(new_link);
}
// set this candidate page as processed
candidate_page._candidate_setDone();
}
}
// mark this thread as done in _thread_status
spider_object.removeThreadStatus();
}
