Program()
{
this.Size = new Size(300, 200);
var b = new Button();
b.Parent = this;
b.Text = "Cancel";
b.Anchor = AnchorStyles.None;
b.Click += (object sender, EventArgs e) =>
{
tokenSource.Cancel();
lock (Lock)
{
PoisonTheWater();
Monitor.PulseAll(Lock);
}
this.Close();
};
b.Left = b.Parent.ClientSize.Width / 2 - b.Width / 2;
b.Top = b.Parent.ClientSize.Height / 2 - b.Height / 2;
this.Show();
webLink rootLink = new webLink();
rootLink.link = rootAddress;
rootLink.origin = rootAddress;
rootLink.depth = 0;
//start thread pool and wait till finish
Console.WriteLine("Working...\n");
setTimer();
sw.Start();
startThreadTask(rootLink);
return;
}
public static void startThreadTask(webLink link)
{
lock (Lock)
{
currentWorkers++;
}
ThreadPool.QueueUserWorkItem((si) =>
{
Consumer(link);
lock (Lock)
{
currentWorkers--;
if (currentWorkers == 0) //last worker poisons everyone else
{
PoisonTheWater();
return;
}
}
});
}
public static void startThreadTask(webLink link)
{
lock (Lock)
{
currentWorkers++;
}
Task.Run(() =>
{
Consumer(link);
lock (Lock)
{
currentWorkers--;
if (currentWorkers == 0) //last worker poisons everyone else
{
PoisonTheWater();
return;
}
}
}, tokenSource.Token);
}
public static void Consumer(webLink linkToTry)
{
WebClient Client = new WebClient();
int num = 0;
try
{
lock (Lock)
{
if (poison)
{
return;
}
num = fileNum++; //increment file name for link
}
Client.DownloadFile(linkToTry.link.AbsoluteUri, num.ToString()); //Try to download, throws exception if link is dead or doesnt work
if (linkToTry.depth < maxDepth) //start producing more consumer tasks if maxDepth not reached
{
Producer(new Tuple <string, webLink>(num.ToString(), linkToTry));
}
}
catch (Exception e) //link is dead or doesnt respond, add to deadlinks and incriment death count.
{
lock (Lock)
{
if (!deadLinks.ContainsKey(linkToTry.link.AbsoluteUri))
{
linkToTry.exception = e;
deadLinks.Add(linkToTry.link.AbsoluteUri, linkToTry);
numDeadLinks++;
}
}
}
}
public static void Producer()
{
//string regMatch = "<\\s*a\\s+[^>]*href\\s*=\\s*['\"][^'\"]*['\"]";
string regMatch = "href\\s*=\\s*(?:[\"'](?<1>[^\"']*)[\"']|(?<1>\\S+))"; //this regex is provided on the microsoft C# documentation website
string data, address = "";
Regex URLmatch = new Regex(regMatch);
MatchCollection MC;
Tuple <string, webLink> link = null; //fileName, weblink struct
webLink newLink;
Uri newAddress;
bool poisoned = false;
while (!poisoned)
{
try
{
//Check the Queue for contents, else wait or die if poisoned
lock (Lock)
{
while (producerLinks.Count == 0 && poison == false)
{
Monitor.Wait(Lock);
}
if (poison)
{
return;
}
link = producerLinks.Dequeue();
if (visitedLinks.ContainsKey(link.Item2.link.AbsoluteUri))
{
throw new Exception("Error: Link already added to Dictionary!!! Why is it in Queue!?");
}
visitedLinks.Add(link.Item2.link.AbsoluteUri, link.Item2);
currentWorkers++;
//Console.WriteLine("Producer Working on: {0}, depth: {1}", link.Item2.link.AbsoluteUri, link.Item2.depth);
}
//read the entire document and match all instances of an address to a collention
data = File.ReadAllText(link.Item1);
if (data.Length > 0)
{
MC = URLmatch.Matches(data);
foreach (Match m in MC)
{
int testEqPos = m.Value.IndexOf('=');
if (testEqPos > 0)
{
address = m.Value.Substring(m.Value.IndexOf('=') + 1).Trim(); //get the address after the = and href
if (address.Length > 3)
{
address = address.Substring(1, address.Length - 2); //remove the " " from the address
}
if (address.Contains(link.Item2.link.Host) || address.StartsWith("http") || address.StartsWith("https"))
{
newAddress = new Uri(address);
}
else
{
newAddress = new Uri(link.Item2.link, address);
}
if (newAddress.Host == link.Item2.link.Host)
{
newLink = new webLink();
newLink.link = newAddress;
newLink.origin = link.Item2.link;
newLink.depth = link.Item2.depth + 1;
lock (Lock)
{
//Console.WriteLine("match address: '{0}'", address);
if (!visitedLinks.ContainsKey(newLink.link.AbsoluteUri) && !deadLinks.ContainsKey(newLink.link.AbsoluteUri) && !quedLinks.ContainsKey(newLink.link.AbsoluteUri))
{
quedLinks.Add(newLink.link.AbsoluteUri, newLink);
clientLinks.Enqueue(newLink);
Monitor.PulseAll(Lock);
}
}
}
}
}
}
}
catch (Exception e)
{
lock (Lock)
{
Console.WriteLine("Error This shouldnt happen, Trying to Read from File:'{0}' for : '{1}'\nAddress:'{2}', Origin:'{3}'\nException: '{4}'\n"
, link.Item1, link.Item2.link.AbsoluteUri, address, link.Item2.link.AbsoluteUri, e.Message);
poison = true;
Monitor.PulseAll(Lock);
return;
}
}
lock (Lock)
{
currentWorkers--;
}
}
}
static void Main(string[] args)
{
Uri rootAddress = null;
bool gotDistance = false;
if (args.Length != 0)
{
if (args[0].Length > 0)
{
try
{
rootAddress = new Uri(args[0]);
}
catch (Exception e)
{
ErrorMessageExit("Issue with creating root Uri", e);
}
gotDistance = Int32.TryParse(args[1], out maxDepth);
if (gotDistance == false)
{
maxDepth = 1;
//ErrorMessageExit("Please Input a integer for distance!");
}
}
}
else if (rootAddress == null)
{
OpenFileDialog dlg = new OpenFileDialog();
dlg.Filter = "All files|*.*";
dlg.ShowDialog();
rootAddress = new Uri(dlg.FileName.Trim());
if (rootAddress == null)
{
return;
}
dlg.Dispose();
maxDepth = 1;
}
else
{
ErrorMessageExit("Please Input a URL string!");
}
if (gotDistance == false)
{
maxDepth = 1;
//ErrorMessageExit("Please Input a integer for distance!");
}
if (maxDepth < 0)
{
ErrorMessageExit("Please Input a positive integer for distance!");
}
//Console.WriteLine("Args0: {0}, Args1: {1}", args[0], args[1]);
//Console.WriteLine("Address: {0}, maxDepth: {1}", rootAddress.AbsoluteUri, maxDepth);
//INITIALLIZE EVERYTHING
visitedLinks = new Dictionary <string, webLink>();
deadLinks = new Dictionary <string, webLink>();
quedLinks = new Dictionary <string, webLink>();
clientLinks = new Queue <webLink>();
producerLinks = new Queue <Tuple <string, webLink> >();
Threads = new List <Thread>();
Lock = new object();
webLink rootLink = new webLink();
setTimer();
numDeadLinks = 0;
currentWorkers = 0;
fileNum = 0;
rootLink.link = rootAddress;
rootLink.origin = rootAddress;
rootLink.depth = 0;
clientLinks.Enqueue(rootLink);
Console.WriteLine("Working...\n");
//create 4 producers and consumers
for (int t = 0; t < 4; t++)
{
Thread newConsumer = new Thread(() => Consumer());
Thread newProducer = new Thread(() => Producer());
newConsumer.Start();
newProducer.Start();
Threads.Add(newConsumer);
Threads.Add(newProducer);
}
//wait for the work to be done
foreach (Thread t in Threads)
{
t.Join();
}
//get ride of timer
onTheClock.Stop();
onTheClock.Dispose();
//print out the bad links
if (visitedLinks.Count > 0)
{
foreach (KeyValuePair <string, webLink> key in deadLinks)
{
Console.WriteLine("DeadLinkOrigin: {0}\nDeadLink: {1}\nDepth: {2}\nException: {3}\n", key.Value.origin, key.Key, key.Value.depth, key.Value.exception.Message);
}
Console.WriteLine("number of links visited: {0}\nnumber of deadLinks: {1}\n", visitedLinks.Count, numDeadLinks);
}
Console.WriteLine("Done");
Console.Read();
}
public static void Consumer()
{
WebClient Client = new WebClient();
webLink linkToTry = new webLink();
Tuple <string, webLink> linkToAdd = null;
int num = 0;
bool poisoned = false;
while (!poisoned)
{
try
{
lock (Lock)
{
if (clientLinks.Count == 0 && producerLinks.Count == 0 && currentWorkers == 0) //last worker poisons everyone else
{
poison = true;
Monitor.PulseAll(Lock);
return;
}
while (clientLinks.Count == 0 && poison == false) //no work, no poison? wait
{
Monitor.Wait(Lock);
}
if (poison) //Is there poison to partake of and leave?
{
return;
}
linkToTry = clientLinks.Dequeue(); //no snack, just work, get item to work on
currentWorkers++; //incriment workers
num = fileNum++; //increment file name for link
//Console.WriteLine("Consumer working on: {0}\nDepth: {1}", linkToTry.link.AbsoluteUri, linkToTry.depth);
}
Client.DownloadFile(linkToTry.link.AbsoluteUri, num.ToString()); //Try to download, throws exception if link is dead or doesnt work
if (linkToTry.depth < maxDepth) //add links file to queue for more work, and wake everyone up
{
if (linkToTry.link.Host == linkToTry.origin.Host)
{
linkToAdd = new Tuple <string, webLink>(num.ToString(), linkToTry);
lock (Lock)
{
producerLinks.Enqueue(linkToAdd);
Monitor.PulseAll(Lock);
}
}
}
}
catch (Exception e) //link is dead or doesnt respond, add to deadlinks and incriment death count.
{
lock (Lock)
{
if (!deadLinks.ContainsKey(linkToTry.link.AbsoluteUri))
{
linkToTry.exception = e;
deadLinks.Add(linkToTry.link.AbsoluteUri, linkToTry);
numDeadLinks++;
}
}
}
lock (Lock)
{
currentWorkers--;
}
}
}
public static void Producer(Tuple <string, webLink> link)
{
//string regMatch = "<\\s*a\\s+[^>]*href\\s*=\\s*['\"][^'\"]*['\"]";
string regMatch = "href\\s*=\\s*(?:[\"'](?<1>[^\"']*)[\"']|(?<1>\\S+))"; //this regex is provided on the microsoft C# documentation website
string data, address = "";
Regex URLmatch = new Regex(regMatch);
webLink newLink;
Uri newAddress;
try
{
//Check the Queue for contents, else wait or die if poisoned
lock (Lock)
{
if (poison)
{
return;
}
}
//read the entire document and match all instances of an address to a collention
data = File.ReadAllText(link.Item1);
if (data.Length > 0)
{
foreach (Match m in URLmatch.Matches(data))
{
int testEqPos = m.Value.IndexOf('=');
address = m.Value.Substring(m.Value.IndexOf('=') + 1).Trim(); //get the address after the = and href
address = address.Substring(1, address.Length - 2); //remove the " " from the address
if (address.Contains(link.Item2.link.Host) || address.StartsWith("http") || address.StartsWith("https"))
{
newAddress = new Uri(address);
}
else
{
newAddress = new Uri(link.Item2.link, address);
}
if (newAddress.Host == link.Item2.link.Host)
{
newLink = new webLink();
newLink.link = newAddress;
newLink.origin = link.Item2.link;
newLink.depth = link.Item2.depth + 1;
lock (Lock)
{
if (!visitedLinks.ContainsKey(newAddress.AbsoluteUri) && !deadLinks.ContainsKey(newLink.link.AbsoluteUri))
{
visitedLinks.Add(newAddress.AbsoluteUri, newLink);
startThreadTask(newLink);
}
}
}
}
}
}
catch (Exception e)
{
lock (Lock)
{
Console.WriteLine("Error!!! This shouldnt happen; Trying to Read from File:'{0}' for : '{1}'\nAddress:'{2}', Origin:'{3}'\nException: '{4}'\n"
, link.Item1, link.Item2.link.AbsoluteUri, address, link.Item2.link.AbsoluteUri, e.Message);
PoisonTheWater();
return;
}
}
}
static void Main(string[] args)
{
Uri rootAddress = null;
bool gotDistance = false;
if (args.Length != 0)
{
if (args[0].Length > 0)
{
try
{
rootAddress = new Uri(args[0]);
}
catch (Exception e)
{
ErrorMessageExit("Issue with creating root Uri", e);
}
gotDistance = Int32.TryParse(args[1], out maxDepth);
if (gotDistance == false)
{
ErrorMessageExit("Please Input a integer for distance!");
}
}
}
else if (rootAddress == null)
{
OpenFileDialog dlg = new OpenFileDialog();
dlg.Filter = "All files|*.*";
dlg.ShowDialog();
rootAddress = new Uri(dlg.FileName.Trim());
if (rootAddress == null)
{
return;
}
dlg.Dispose();
maxDepth = 1;
}
else
{
ErrorMessageExit("Please Input a URL string!");
}
if (gotDistance == false)
{
maxDepth = 1;
//ErrorMessageExit("Please Input a integer for distance!");
}
if (maxDepth < 0)
{
ErrorMessageExit("Please Input a positive integer for distance!");
}
//INITIALLIZE EVERYTHING
visitedLinks = new Dictionary <string, webLink>();
deadLinks = new Dictionary <string, webLink>();
Lock = new object();
webLink rootLink = new webLink();
Stopwatch sw = new Stopwatch();
numDeadLinks = 0;
currentWorkers = 0;
fileNum = 0;
rootLink.link = rootAddress;
rootLink.origin = rootAddress;
rootLink.depth = 0;
setTimer();
sw.Start();
//start thread pool and wait till finish
Console.WriteLine("Working...\n");
visitedLinks.Add(rootAddress.AbsoluteUri, rootLink);
startThreadTask(rootLink);
lock (Lock)
{
while (!poison)
{
Monitor.Wait(Lock);
}
}
//get ride of timer and stop stopwatch
sw.Stop();
onTheClock.Stop();
onTheClock.Dispose();
//print out the bad links
if (visitedLinks.Count > 0)
{
foreach (KeyValuePair <string, webLink> key in deadLinks)
{
Console.WriteLine("DeadLinkOrigin: {0}\nDeadLink: {1}\nDepth: {2}\nException: {3}\n", key.Value.origin, key.Key, key.Value.depth, key.Value.exception.Message);
}
Console.WriteLine("number of links visited: {0}\nnumber of deadLinks: {1}\nElapsedTime: {2}", visitedLinks.Count, numDeadLinks, sw.Elapsed);
}
else
{
Console.WriteLine("No Links Visited, RootAddressGiven: '{0}', DepthGiven: '{1}'", rootAddress, maxDepth);
}
Console.WriteLine("Done");
Console.Read();
}
public static void Consumer(webLink linkToTry)
{
HttpClient Client = new HttpClient();
FileStream newFile;
int num;
string FileName;
try
{
lock (Lock)
{
if (poison)
{
return;
}
num = fileNum++; //increment file name for link
}
var result = Client.GetAsync(linkToTry.link.AbsoluteUri, tokenSource.Token); //Try to download, throws exception if link is dead or doesnt work
if (tokenSource.IsCancellationRequested)
{
tokenSource.Token.ThrowIfCancellationRequested();
}
if (result.Result.IsSuccessStatusCode) //Http connection success
{
if (linkToTry.link.IsFile)
{
FileName = Path.GetFileName(linkToTry.link.LocalPath) + '_' + num.ToString();
}
else
{
FileName = num.ToString();
}
newFile = new FileStream(FileName, FileMode.Create);
byte[] webData = result.Result.Content.ReadAsByteArrayAsync().Result;
newFile.Write(webData, 0, webData.Length);
newFile.Close();
if (linkToTry.depth < maxDepth) //start producing more consumer tasks if maxDepth not reached
{
Producer(new Tuple <string, webLink>(num.ToString(), linkToTry));
}
}
else //link is dead or doesnt respond, add to deadlinks and incriment death count.
{
lock (Lock)
{
if (!deadLinks.ContainsKey(linkToTry.link.AbsoluteUri))
{
linkToTry.exception = result.Result.StatusCode.ToString();
deadLinks.Add(linkToTry.link.AbsoluteUri, linkToTry);
numDeadLinks++;
}
}
}
}
catch (OperationCanceledException e)
{
PoisonTheWater();
return;
}
catch (Exception e)
{
lock (Lock)
{
if (!deadLinks.ContainsKey(linkToTry.link.AbsoluteUri))
{
linkToTry.exception = e.Message.ToString();
deadLinks.Add(linkToTry.link.AbsoluteUri, linkToTry);
numDeadLinks++;
}
}
}
}
