public void AddLinkToCrawl(LinkToCrawl link)
{
// if the link to add is NOT in the list of links to crawl or crawled links, then add it
Thread.Sleep(100);
var q = from l in LinksToCrawl.Values
where l.SessionId == link.SessionId &&
string.Compare(l.SourceUrl, link.SourceUrl, true) == 0 &&
string.Compare(l.TargetUrl, link.TargetUrl, true) == 0
select l;
if (!q.Any())
{
var q2 = from l in CrawledLinks.Values
where l.SessionId == link.SessionId &&
string.Compare(l.SourceUrl, link.SourceUrl, true) == 0 &&
string.Compare(l.TargetUrl, link.TargetUrl, true) == 0
select l;
if (!q2.Any())
{
link.Id = NextId;
LinksToCrawl.Add(link.Id, link);
}
}
}
private async Task RunCrawl(string hostname)
{
HostUrl = VerifyUrlIntegrity(hostname);
LinksToCrawl.Enqueue(HostUrl.OriginalString);
while (LinksToCrawl.Any())
{
List <Task> tasks = new List <Task>();
for (var thread = 1; thread <= ConcurrencyLimit && thread <= LinksToCrawl.Count; thread++)
{
var link = new Uri(LinksToCrawl.Dequeue());
if (WebsiteMap.ContainsKey(link.Host + link.AbsolutePath))
{
continue;
}
tasks.Add(CrawlPage(link.OriginalString));
}
await Task.WhenAll(tasks);
}
Console.WriteLine($"Found {WebsiteMap.Count} links");
}
private async Task CrawlPage(string link)
{
var pageResults = await _parser.ParsePage(link);
HandlePageResultLinks(pageResults);
WebsiteMap.TryAdd(pageResults.PageUrl.Host + pageResults.PageUrl.AbsolutePath, pageResults);
pageResults.Links.ForEach(x => LinksToCrawl.Enqueue(x));
}
public void ClearLinksToCrawl(int sessionId, string baseDomain)
{
var q = from l in LinksToCrawl.Values
where l.SessionId == sessionId &&
string.Compare(l.TargetBaseDomain, baseDomain, false) == 0
select l.Id;
foreach (var id in q.ToList())
{
LinksToCrawl.Remove(id);
}
}
public void DeleteLinkToCrawl(Guid id)
{
Thread.Sleep(100);
var q = from l in LinksToCrawl.Values
where l.Id == id
select l;
var link = q.FirstOrDefault();
if (link != null && LinksToCrawl.ContainsKey(link.Id))
{
LinksToCrawl.Remove(link.Id);
}
}
public void DeleteLinkToCrawl(int sessionId, string srcUrl, string targetUrl)
{
Thread.Sleep(100);
var q = from l in LinksToCrawl.Values
where l.SessionId == sessionId &&
l.SourceUrl == srcUrl &&
l.TargetUrl == targetUrl
select l;
var link = q.FirstOrDefault();
if (link != null && LinksToCrawl.ContainsKey(link.Id))
{
LinksToCrawl.Remove(link.Id);
}
}
public void GetLinksToCrawl()
{
for (int i = 1; i < NumPageToCrawl; i++)
{
var url = BaseUrl + "?p=" + i + "&q=" + UrlEncodedQueryString;
Driver.Navigate().GoToUrl(url);
var allLinks = Driver.FindElementsByTagName("a");
foreach (var link in allLinks)
{
if (link.Text.Contains(FileName))
{
var linkHref = link.GetAttribute("href");
if (!LinksToCrawl.Contains(linkHref))
{
LinksToCrawl.Add(linkHref);
}
}
}
}
}
public void ProcessLinks(Abot.Poco.CrawledPage page)
{
if (page.ParsedLinks == null || page.ParsedLinks.Count() == 0)
{
_logger.DebugFormat("CrawledPage contained 0 parsed links");
LinksToCrawl = new List <LinkToCrawl>();
LinksToByPass = new List <CrawledLink>();
return;
}
LinksToByPass = new List <CrawledLink>();
MapOfLinksToCrawl = new Dictionary <string, LinkToCrawl>();
using (var factory = _provider.GetInstanceOf <IModelFactory>())
{
var sessionId = page.PageBag.SessionId;
var crawlerId = page.PageBag.CrawlerId;
LinkToCrawl link = null;
CrawledLink bypassedLink = null;
foreach (var targetUri in page.ParsedLinks)
{
ProcessLink(page, factory, targetUri, sessionId, crawlerId);
}
LinksToCrawl = MapOfLinksToCrawl.Values.ToList();
MapOfLinksToCrawl.Clear();
MapOfLinksToCrawl = null;
if (_logger.IsDebugEnabled)
{
_logger.DebugFormat("TargetUrls of new LinksToCrawl: {0}",
String.Join("; ", LinksToCrawl.Select(o => o.TargetUrl)));
_logger.DebugFormat("TargetUrls of new LinksToByPass: {0}",
String.Join("; ", LinksToByPass.Select(o => o.TargetUrl)));
}
}
}
