public static WebPageDocument crawl(string url)
{
HttpWebRequest request = (HttpWebRequest)HttpWebRequest.Create(url);
request.UserAgent = "A .NET Web Crawler";
try
{
WebPageDocument webPage = new WebPageDocument();
HttpWebResponse response = (HttpWebResponse)request.GetResponse();
Stream stream = response.GetResponseStream();
StreamReader reader = new StreamReader(stream);
string rawHtmlString = reader.ReadToEnd();
webPage.Links = Detector.linkFinder(rawHtmlString);
Detector.setWebPageInformation(rawHtmlString, webPage);
webPage.Url = url;
return(webPage);
}
catch
{
return(null);
}
}
public static void setWebPageInformation(string rawHtmlString, WebPageDocument webPage)
{
Match match1 = Regex.Match(rawHtmlString, @"<meta name=\""description\"" content=\""(.*?)\""", RegexOptions.Singleline);
if (match1.Success)
{
webPage.Description = match1.Groups[1].Value;
}
Match match2 = Regex.Match(rawHtmlString, @"<meta name=\""keywords\"" content=\""(.*?)\""", RegexOptions.Singleline);
if (match2.Success)
{
webPage.Keywords = webPage.buildFeatures(Tokenizer.tokenize(match2.Groups[1].Value));
}
Match match3 = Regex.Match(rawHtmlString, @"<title>(.*?)</title>", RegexOptions.Singleline);
if (match3.Success)
{
webPage.Title = match3.Groups[1].Value;
}
MatchCollection matchCollection = Regex.Matches(rawHtmlString, @"<h\d.*?>(.*?)</h1>");
webPage.Headings = new List <Feature>();
foreach (Match m in matchCollection)
{
string pureText = Regex.Replace(m.Groups[1].Value, "<.*?>", String.Empty);
List <string> tokens = Tokenizer.tokenize(pureText);
webPage.Headings = webPage.buildFeatures(tokens);
}
MatchCollection matchCollection2 = Regex.Matches(rawHtmlString, @"<p.*?>(.*?)</p>", RegexOptions.Multiline);
webPage.Paragraphs = new List <Feature>();
foreach (Match m in matchCollection2)
{
string pureText = Regex.Replace(m.Groups[1].Value, "<.*?>", String.Empty);
List <string> tokens = Tokenizer.tokenize(pureText);
webPage.Paragraphs = webPage.buildFeatures(tokens);
}
}
static void Main(string[] args)
{
Console.WriteLine("Alireza Khanshan - A Web Crawler");
Console.Write("How many pages? ");
int ALL_PAGES = Convert.ToInt32(Console.ReadLine());
using (StreamWriter writer =
new StreamWriter("C:/Crawler/Data/pages.txt"))
{
writer.Write(ALL_PAGES);
}
Console.WriteLine("crawling...");
int count = 1;
List <Link> frontier = Crawler.crawl("http://espn.go.com/nba").Links;
List <string> blackList = new List <string>();
List <Document> documents = new List <Document>();
List <WebPageDocument> webPageDcuments = new List <WebPageDocument>();
Console.WriteLine(count);
Console.WriteLine("http://espn.go.com/nba : " + frontier.Count);
count++;
while (true)
{
if (frontier.Count > 0 && count < ALL_PAGES)
{
//
foreach (string href in blackList)
{
frontier.RemoveAll(item => item.Href != null && item.Href.Equals(href));
}
//
Link crawlMe = frontier[0];
if (crawlMe.Href == null)
{
frontier.Remove(crawlMe);
}
else if (crawlMe.Href.ToLower().Contains("mailto"))
{
frontier.Remove(crawlMe);
}
else if (crawlMe.Href.ToLower().StartsWith("http://espn.go.com/nba"))
{
frontier.RemoveAll(item => item.Href == crawlMe.Href);
blackList.Add(crawlMe.Href);
WebPageDocument webPage = Crawler.crawl(crawlMe.Href);
List <Link> newLinks = webPage != null ? webPage.Links : null;
if (newLinks != null)
{
Console.Write(count + ". ");
Console.WriteLine(webPage.Url);
documents.Add(new Document()
{
Id = count,
Description = webPage.Description,
Links = newLinks.Count,
Title = webPage.Title,
Url = webPage.Url
});
webPage.Id = count;
webPageDcuments.Add(webPage);
count++;
foreach (Link l in newLinks)
{
frontier.Add(l);
}
}
}
else if (crawlMe.Href.ToLower().StartsWith("/nba/"))
{
frontier.RemoveAll(item => item.Href == crawlMe.Href);
blackList.Add(crawlMe.Href);
WebPageDocument webPage = Crawler.crawl("http://espn.go.com" + crawlMe.Href);
List <Link> newLinks = webPage != null ? webPage.Links : null;
if (newLinks != null)
{
Console.Write(count + ". ");
Console.WriteLine(webPage.Url);
documents.Add(new Document()
{
Id = count,
Description = webPage.Description,
Links = newLinks.Count,
Title = webPage.Title,
Url = webPage.Url
});
webPage.Id = count;
webPageDcuments.Add(webPage);
count++;
foreach (Link l in newLinks)
{
frontier.Add(l);
}
}
}
else
{
frontier.Remove(crawlMe);
}
}
else
{
break;
}
}
List <InvertedListItem> list = IndexBuilder.build(webPageDcuments);
using (var db = new LiteDatabase(@"C:/Crawler/Data/crawler.db"))
{
var invertedIndex = db.GetCollection <InvertedListItem>("inverted_index");
foreach (InvertedListItem i in list)
{
invertedIndex.Insert(i);
}
var documentDataStore = db.GetCollection <Document>("document_store");
foreach (Document d in documents)
{
documentDataStore.Insert(d);
}
}
Console.WriteLine("*** done ***");
Console.ReadKey();
}
