private async Task <IEnumerable <Link> > GetRootPageLinksAsync()
{
IEnumerable <Link> links = null;
var html = await GetRootPageHtmlAsync();
links = CrawlerParser.GetPageLinks(this.RootUri, html);
if (links != null)
{
links = FilterLinks(links);
}
else
{
return(null);
}
foreach (var link in links)
{
try
{
if (link != null && !string.IsNullOrEmpty(link.Href))
{
SiteNode.AddChild(link);
}
}
catch (Exception ex)
{
}
}
return(links);
}
private static readonly string[] staticFileExtensions = { ".css", ".ico", ".pdf", ".json", ".xml", ".jpg", ".jpeg", ".bmp", ".gif", ".png", ".js" }; // etc
/// <summary>
/// Get a list of Link objects from html
/// </summary>
/// <param name="html">Page Html</param>
/// <returns>List of Link objects</returns>
public static IEnumerable <Link> GetPageLinks(string rootUrl, string html)
{
if (string.IsNullOrEmpty(html))
{
return(null);
}
var list = new List <Link>();
string strRegex = @"href\s*=\s*(?:\""(?<1>[^\""]*)\""|(?<1>\\S+))"; //regex: searches anchor tags
var regex = new Regex(strRegex, RegexOptions.IgnoreCase | RegexOptions.Multiline);
//On match found parse the link and store it as Link
if (regex != null && regex.IsMatch(html))
{
foreach (Match match in regex.Matches(html))
{
var link = CrawlerParser.ParseLink(rootUrl, match.Value);
//Only add if Link(:href) does not exist in the list
if (!list.Exists(x => x.Href.Equals(link.Href)))
{
list.Add(link);
}
}
}
return(list);
}
/// <summary>
/// Parse Link object from method Crawler.GetPageLinks() retunred html
/// </summary>
/// <param name="rootUri"></param>
/// <param name="raw"></param>
/// <returns></returns>
public static Link ParseLink(string rootUri, string raw)
{
//Parse URI from regex match value and perform some cleanups
var href = CrawlerParser.ParseTextFromQuote(raw);
// sets true if wipro site
bool isWipro = CrawlerParser.IsWipro(href);
// sets true if url is relative
bool isRelative = CrawlerParser.IsRelativeUrl(href);
// sets true javascript links
bool isJs = CrawlerParser.IsJavaScriptOrHashLink(raw, href);
// sets true if file is static
bool isStatic = CrawlerParser.IsStatic(href);
//get filename from url
string extension = string.Empty;
string fileName = null;
if (isWipro && !isJs && !isStatic)
{
fileName = Utilities.GetFileNameFromUrl(rootUri, href, out extension);
}
//get file-type from extension
var type = Utilities.GetHrefType(extension);
// group[1] value contains Title
//var title = match.Groups.Count > 0 ? match.Groups[1].Value : "";
//generate Link object
var link = new Link
{
Raw = raw, // store raw search
Href = href, // hyperlink
IsRelativeUrl = isRelative,
IsWipro = isWipro,
IsJavaScript = isJs,
IsStatic = isStatic,
Type = type,
FileName = fileName
};
return(link);
}
public IEnumerable <Link> GetPageImages(string html)
{
var list = new List <Link>();
var strRegex = @"<img.+?src=[""'](.+?)[""'].*?>"; //regex: searches image tags
var regex = new Regex(strRegex, RegexOptions.IgnoreCase | RegexOptions.Multiline);
//On match found parse the link and store it as Link
if (regex.IsMatch(html))
{
foreach (Match match in regex.Matches(html))
{
list.Add(new Link
{
Raw = match.Value, // store raw search
Href = match.Groups[1].Value, // hyperlink
IsRelativeUrl = CrawlerParser.IsRelativeUrl(match.Groups[1].Value), // set to true if wipro site
IsImage = true // javascript links
});
}
}
return(list);
}
public IEnumerable <Link> ExtractAll(string url, bool isRoot)
{
//sort, remove duplicates
IEnumerable <Link> links = null;
if (isRoot)
{
var html = GetRootPageHtml();
if (string.IsNullOrEmpty(html))
{
return(links);
}
links = FilterLinks(CrawlerParser.GetPageLinks(url, html));
foreach (var link in links)
{
if (!this.Links.Contains(link))
{
this.Links.Add(link);
}
SiteNode.AddChild(link);
ExtractAll(link.Href, false);
}
}
else
{
url = CrawlerParser.GetAbsoluteUri(this.RootUri, url);
var html = GetPageHtml(url);
if (string.IsNullOrEmpty(html))
{
return(links);
}
links = CrawlerParser.GetPageLinks(this.RootUri, html);
if (links != null)
{
links = FilterLinks(links);
}
else
{
return(links);
}
foreach (var link in links)
{
var node = this.SiteNode.FindTreeNode(x => x.Data.Href.Equals(link.Href));
if (node == null)
{
this.SiteNode.AddChild(link);
if (depthCounter2 < MAX_DEPTH_COUNT)
{
depthCounter2++;
this.Links.AddRange(ExtractAll(link.Href, false));
}
}
else
{
if (depthCounter2 < MAX_DEPTH_COUNT)
{
depthCounter2++;
node.AddChild(link);
link.Links = ExtractAll(link.Href, false);
}
}
}
}
return(links);
}
public async Task ExtractAllAsync(string url, bool isRoot)
{
//sort, remove duplicates
IEnumerable <Link> links = null;
if (isRoot)
{
var html = await GetRootPageHtmlAsync();
links = CrawlerParser.GetPageLinks(this.RootUri, html);
if (links != null)
{
links = links.Where(x => !x.IsJavaScript && !x.IsStatic && x.IsWipro);
}
else
{
return;
}
foreach (var link in links)
{
try
{
if (link != null && !string.IsNullOrEmpty(link.Href))
{
SiteNode.AddChild(link);
if (!blackListUrls.Contains(link.Href))
{
await ExtractAllAsync(link.Href, false);
}
}
}
catch (Exception ex)
{
}
}
}
else
{
url = CrawlerParser.GetAbsoluteUri(this.RootUri, url);
string html = string.Empty;
try
{
html = await GetPageAsync(url);
}
catch (Exception ex)
{
}
links = CrawlerParser.GetPageLinks(this.RootUri, html);
if (links != null)
{
links = links.Where(x => !x.IsJavaScript && !x.IsStatic && x.IsWipro);
}
else
{
return;
}
foreach (var link in links)
{
try
{
var node = this.SiteNode.FindTreeNode(x => x.Data.Href.Equals(link.Href));
if (node == null)
{
if (!this.SiteNode.Children.Any(x => x.Data.Href.Equals(link.Href)))
{
this.SiteNode.AddChild(link);
}
if (!blackListUrls.Contains(link.Href) && link.Href.ToLower() != this.RootUri.ToLower() && depthCounter2 < MAX_DEPTH_COUNT)
{
depthCounter2++;
await ExtractAllAsync(link.Href, false);
Console.WriteLine("Href: {0}\nChildren Count: {1}\nElements Index: {2}", link.Href, SiteNode.Children.Count, SiteNode.ElementsIndex.Count);
}
}
else
{
if (!blackListUrls.Contains(link.Href) && link.Href.ToLower() != this.RootUri.ToLower() && depthCounter2 < MAX_DEPTH_COUNT)
{
depthCounter2++;
if (!node.IsRoot && !node.Children.Any(x => x.Data.Href.Equals(link.Href)))
{
node.AddChild(link);
}
await ExtractAllAsync(link.Href, false);
}
}
}
catch (Exception ex)
{
}
}
}
}
private async Task <IEnumerable <Link> > GetChildPagesLinksAsync(string url)
{
IEnumerable <Link> links = null;
url = CrawlerParser.GetAbsoluteUri(this.RootUri, url);
string html = string.Empty;
try
{
html = await GetPageAsync(url);
}
catch (Exception ex)
{
}
links = CrawlerParser.GetPageLinks(this.RootUri, html);
if (links != null)
{
links = FilterLinks(links);
}
else
{
return(null);
}
Parallel.ForEach(links, async link =>
{
try
{
var node = this.SiteNode.FindTreeNode(x => x.Data.Href.Equals(link.Href));
if (node == null)
{
if (!this.SiteNode.Children.Any(x => x.Data.Href.Equals(link.Href)))
{
this.SiteNode.AddChild(link);
}
if (depthCounter2 < MAX_DEPTH_COUNT)
{
depthCounter2++;
link.Links = await GetChildPagesLinksAsync(link.Href);
Debug.WriteLine("Href: {0}\nChildren Count: {1}\nElements Index: {2}\nDepth: {3}", link.Href, SiteNode.Children.Count, SiteNode.ElementsIndex.Count, depthCounter2);
}
}
else
{
if (depthCounter2 < MAX_DEPTH_COUNT)
{
depthCounter2++;
if (!node.IsRoot && !node.Children.Any(x => x.Data.Href.Equals(link.Href)))
{
node.AddChild(link);
}
link.Links = await GetChildPagesLinksAsync(link.Href);
Debug.WriteLine("Href: {0}\nChildren Count: {1}\nElements Index: {2}\nDepth: {3}", link.Href, SiteNode.Children.Count, SiteNode.ElementsIndex.Count, depthCounter2);
}
}
}
catch (Exception ex)
{
}
});
return(links);
}
