/// <summary>
/// starts the comparison.
/// After this method has completed you can retrieve the maximum found similarity
/// </summary>
public int Calculate()
{
int max = 0;
cLogger.DebugFormat("calculating similarity");
foreach (var data in ReferenceData)
{
// create the list here, because this way it is local to this run
// ==> more functional and separated
var sourceTokens = Tokens.ToGSTTokenList();
var referenceTokens = Factory.GetTokenWrapperEnumerable(data.Tokens).ToGSTTokenList();
var algorithm = new HashingGSTAlgorithm <GSTToken <TokenWrapper> >(sourceTokens, referenceTokens)
{
MinimumMatchLength = DEFAULT_MML
};
algorithm.RunToCompletion();
cLogger.DebugFormat("similarity compared to {0}:{1}", data.StudentIdentifier, algorithm.Similarity);
if (MaximumSimilarity < algorithm.Similarity)
{
MaximumSimilarity = algorithm.Similarity;
MaxSimilarityStudentID = data.StudentIdentifier;
}
}
return(max);
}
public void MinimizesHashes()
{
Algorithm = new HashingGSTAlgorithm<GSTToken<char>>(GSTHelper.FromString("XeLATst"), GSTHelper.FromString("LATunik")) { MinimumMatchLength = 3};
Algorithm.DoOneRun();
Assert.AreEqual(1, Algorithm.HashesA.Count, string.Format("expected 1 after minimize, but: A = {0}", Algorithm.HashesA.Count));
Assert.AreEqual(1, Algorithm.HashesB.Count, string.Format("expected 1 after minimize, but: B = {0}", Algorithm.HashesB.Count));
}
public void MinimizesHashes()
{
Algorithm = new HashingGSTAlgorithm <GSTToken <char> >(GSTHelper.FromString("XeLATst"), GSTHelper.FromString("LATunik"))
{
MinimumMatchLength = 3
};
Algorithm.DoOneRun();
Assert.AreEqual(1, Algorithm.HashesA.Count, string.Format("expected 1 after minimize, but: A = {0}", Algorithm.HashesA.Count));
Assert.AreEqual(1, Algorithm.HashesB.Count, string.Format("expected 1 after minimize, but: B = {0}", Algorithm.HashesB.Count));
}
public void SetUp()
{
abstractGSTAlgorithmTest = new AbstractGSTAlgorithmTest(typeof(GSTAlgorithm <GSTToken <char> >));
var listA = GSTHelper.FromString("Hallo");
var listB = GSTHelper.FromString("Hallo");
Algorithm = new HashingGSTAlgorithm <GSTToken <char> >(listA, listB)
{
MinimumMatchLength = 3
};
}
public void EqualityComparerTest()
{
var dict1 = HashingGSTAlgorithm <GSTToken <char> > .CreateHashMap(GSTHelper.FromString("Hallo"), 3).ToArray();
var dict2 = HashingGSTAlgorithm <GSTToken <char> > .CreateHashMap(GSTHelper.FromString("Hallo"), 3).ToArray();
var comparer = HashingGSTAlgorithm <GSTToken <char> > .HashingEntity.Comparer;
for (int i = 0; i < dict1.Length; i++)
{
Assert.AreEqual(dict1[i].Key, dict2[i].Key, string.Format("dict1 {0}, dict2 {1}", dict1[i].Value[0], dict2[i].Value[0]));
Assert.IsTrue(comparer.Equals(dict1[i].Value[0], dict2[i].Value[0]));
}
}
/// <summary>
/// Compares the two files and returns the Similarity.
/// Lexer used: MutexCLexer
/// Algorithm used: HashingGSTAlgorithm (MML = 8)
/// </summary>
/// <param name="path1"></param>
/// <param name="path2"></param>
/// <returns></returns>
public static Int32 CompareFiles(string path1, string path2)
{
var factory = new MutexTokenFactory();
var tokens1 = factory.GetTokenWrapperListFromFile(path1);
var tokens2 = factory.GetTokenWrapperListFromFile(path2);
var algo = new HashingGSTAlgorithm<GSTToken<TokenWrapper>>(
tokens1.ToGSTTokenList<TokenWrapper>(),
tokens2.ToGSTTokenList<TokenWrapper>()) {MinimumMatchLength = 8};
algo.RunToCompletion();
return algo.Similarity;
}
private static void Calculate(int len)
{
var string1 = src1.Substring(0, len);
var string2 = src2.Substring(0, len);
//len += 200; // double length in size
var watch = Stopwatch.StartNew();
//Console.WriteLine("String length: A = {0}, B = {1}", string1.Length, string2.Length);
for (int i = testRuns; i >= 0; i--)
{
var algo = new HashingGSTAlgorithm <GSTToken <char> >(
GSTHelper.FromString(string1),
GSTHelper.FromString(string2));
algo.RunToCompletion();
//Console.WriteLine("finished hashing run {0} in {1}", i, watch.Elapsed);
}
var runtimeHashing = watch.Elapsed;
//Console.WriteLine("runtime hashing: {0}", ((int)runtimeHashing.TotalMilliseconds / testRuns));
watch = Stopwatch.StartNew();
for (int i = testRuns; i >= 0; i--)
{
var algo = new GSTAlgorithm <GSTToken <char> >(
GSTHelper.FromString(string1),
GSTHelper.FromString(string2));
algo.RunToCompletion();
}
var ratio = (double)watch.ElapsedTicks / runtimeHashing.Ticks;
Console.WriteLine("runtimes: {0}, {1}, ratio: {2} for {3}",
watch.ElapsedMilliseconds / testRuns,
runtimeHashing.TotalMilliseconds / testRuns,
ratio,
len);
ResultList.Sort(String.CompareOrdinal);
ResultList.Add(string.Format("{3:000};{0};{1};{2}",
(watch.Elapsed.TotalMilliseconds / testRuns),
(runtimeHashing.TotalMilliseconds / testRuns),
ratio,
len));
}
/// <summary>
/// Compares the two files and returns the Similarity.
/// Lexer used: MutexCLexer
/// Algorithm used: HashingGSTAlgorithm (MML = 8)
/// </summary>
/// <param name="path1"></param>
/// <param name="path2"></param>
/// <returns></returns>
public static Int32 CompareFiles(string path1, string path2)
{
var factory = new MutexTokenFactory();
var tokens1 = factory.GetTokenWrapperListFromFile(path1);
var tokens2 = factory.GetTokenWrapperListFromFile(path2);
var algo = new HashingGSTAlgorithm <GSTToken <TokenWrapper> >(
tokens1.ToGSTTokenList <TokenWrapper>(),
tokens2.ToGSTTokenList <TokenWrapper>())
{
MinimumMatchLength = 8
};
algo.RunToCompletion();
return(algo.Similarity);
}
public void ANTLRITokenVersusMutexTokenImpl()
{
var factory = new MutexTokenFactory();
var itokens = factory.GetTokenWrapperListFromSource("void main(int argc, char** argv)").ToGSTTokenList();
var tokennames = new[] { "VOID", "IDENTIFIER", "INTEGER_DATATYPE", "IDENTIFIER", "POINTER_DATATYPE", "IDENTIFIER" };
var mutextokens = factory.GetTokenWrapperEnumerable(tokennames).ToGSTTokenList();
for (int i = 0; i < itokens.Count; i++)
{
Assert.AreEqual(itokens[i].GetHashCode(), mutextokens[i].GetHashCode());
}
var algo = new HashingGSTAlgorithm<GSTToken<TokenWrapper>>(itokens, mutextokens)
{
MinimumMatchLength = 3
};
algo.RunToCompletion();
Assert.AreEqual(100, algo.Similarity);
}
public void ANTLRITokenVersusMutexTokenImpl()
{
var factory = new MutexTokenFactory();
var itokens = factory.GetTokenWrapperListFromSource("void main(int argc, char** argv)").ToGSTTokenList();
var tokennames = new[] { "VOID", "IDENTIFIER", "INTEGER_DATATYPE", "IDENTIFIER", "POINTER_DATATYPE", "IDENTIFIER" };
var mutextokens = factory.GetTokenWrapperEnumerable(tokennames).ToGSTTokenList();
for (int i = 0; i < itokens.Count; i++)
{
Assert.AreEqual(itokens[i].GetHashCode(), mutextokens[i].GetHashCode());
}
var algo = new HashingGSTAlgorithm <GSTToken <TokenWrapper> >(itokens, mutextokens)
{
MinimumMatchLength = 3
};
algo.RunToCompletion();
Assert.AreEqual(100, algo.Similarity);
}
/// <summary>
/// compares the two algorithms against each other over a rather large file (~8000 chars)
/// </summary>
private static void EvaluateSpeed()
{
XmlConfigurator.Configure(new FileInfo("log4net.xml"));
src1 = File.ReadAllText(@"test\default_set\TPLV04-S01-02\main-01.c");
src2 = File.ReadAllText(@"test\default_set\TPLV04-S01-02\main-02.c");
var testRuns = 2;
int len = 100;
// JIT compile both algorithms before we start
AbstractGSTAlgorithm <GSTToken <char> > initalgo = new HashingGSTAlgorithm <GSTToken <char> >(
GSTHelper.FromString(src1.Substring(0, 100)),
GSTHelper.FromString(src2.Substring(0, 100)));
initalgo.RunToCompletion();
initalgo = new GSTAlgorithm <GSTToken <char> >(GSTHelper.FromString(src1.Substring(0, 100)),
GSTHelper.FromString(src2.Substring(0, 100)));
initalgo.RunToCompletion();
var liLen1 = new List <int>();
var liLen2 = new List <int>();
var liLen3 = new List <int>();
int c = 0;
while (len < src1.Length && len < src2.Length)
{
switch (c++ % 3)
{
case 0:
liLen1.Add(len);
break;
case 1:
liLen2.Add(len);
break;
default:
liLen3.Add(len);
break;
}
len += 100;
}
Length1 = liLen1.ToArray();
Length2 = liLen2.ToArray();
Length3 = liLen3.ToArray();
var t1 = new Thread(Start1);
t1.Start();
var t2 = new Thread(Start2);
t2.Start();
var t3 = new Thread(Start3);
t3.Start();
Console.WriteLine("Main thread joining");
t1.Join();
t2.Join();
t3.Join();
File.WriteAllLines(@"test\default_set\performance.txt", ResultList);
Console.WriteLine("finished all runs");
Console.ReadLine();
}
/// <summary>
/// starts the comparison.
/// After this method has completed you can retrieve the maximum found similarity
/// </summary>
public int Calculate()
{
int max = 0;
cLogger.DebugFormat("calculating similarity");
foreach (var data in ReferenceData)
{
// create the list here, because this way it is local to this run
// ==> more functional and separated
var sourceTokens = Tokens.ToGSTTokenList();
var referenceTokens = Factory.GetTokenWrapperEnumerable(data.Tokens).ToGSTTokenList();
var algorithm = new HashingGSTAlgorithm<GSTToken<TokenWrapper>>(sourceTokens, referenceTokens)
{
MinimumMatchLength = DEFAULT_MML
};
algorithm.RunToCompletion();
cLogger.DebugFormat("similarity compared to {0}:{1}", data.StudentIdentifier, algorithm.Similarity);
if (MaximumSimilarity < algorithm.Similarity)
{
MaximumSimilarity = algorithm.Similarity;
MaxSimilarityStudentID = data.StudentIdentifier;
}
}
return max;
}
public void SetUp()
{
abstractGSTAlgorithmTest = new AbstractGSTAlgorithmTest(typeof(GSTAlgorithm<GSTToken<char>>));
var listA = GSTHelper.FromString("Hallo");
var listB = GSTHelper.FromString("Hallo");
Algorithm = new HashingGSTAlgorithm<GSTToken<char>>(listA, listB)
{
MinimumMatchLength = 3
};
}
private static void Calculate(int len)
{
var string1 = src1.Substring(0, len);
var string2 = src2.Substring(0, len);
//len += 200; // double length in size
var watch = Stopwatch.StartNew();
//Console.WriteLine("String length: A = {0}, B = {1}", string1.Length, string2.Length);
for (int i = testRuns; i >= 0; i--)
{
var algo = new HashingGSTAlgorithm<GSTToken<char>>(
GSTHelper.FromString(string1),
GSTHelper.FromString(string2));
algo.RunToCompletion();
//Console.WriteLine("finished hashing run {0} in {1}", i, watch.Elapsed);
}
var runtimeHashing = watch.Elapsed;
//Console.WriteLine("runtime hashing: {0}", ((int)runtimeHashing.TotalMilliseconds / testRuns));
watch = Stopwatch.StartNew();
for (int i = testRuns; i >= 0; i--)
{
var algo = new GSTAlgorithm<GSTToken<char>>(
GSTHelper.FromString(string1),
GSTHelper.FromString(string2));
algo.RunToCompletion();
}
var ratio = (double)watch.ElapsedTicks / runtimeHashing.Ticks;
Console.WriteLine("runtimes: {0}, {1}, ratio: {2} for {3}",
watch.ElapsedMilliseconds / testRuns,
runtimeHashing.TotalMilliseconds / testRuns,
ratio,
len);
ResultList.Sort(String.CompareOrdinal);
ResultList.Add(string.Format("{3:000};{0};{1};{2}",
(watch.Elapsed.TotalMilliseconds / testRuns),
(runtimeHashing.TotalMilliseconds / testRuns),
ratio,
len));
}
/// <summary>
/// compares the two algorithms against each other over a rather large file (~8000 chars)
/// </summary>
private static void EvaluateSpeed()
{
XmlConfigurator.Configure(new FileInfo("log4net.xml"));
src1 = File.ReadAllText(@"test\default_set\TPLV04-S01-02\main-01.c");
src2 = File.ReadAllText(@"test\default_set\TPLV04-S01-02\main-02.c");
var testRuns = 2;
int len = 100;
// JIT compile both algorithms before we start
AbstractGSTAlgorithm<GSTToken<char>> initalgo = new HashingGSTAlgorithm<GSTToken<char>>(
GSTHelper.FromString(src1.Substring(0, 100)),
GSTHelper.FromString(src2.Substring(0, 100)));
initalgo.RunToCompletion();
initalgo = new GSTAlgorithm<GSTToken<char>>(GSTHelper.FromString(src1.Substring(0, 100)),
GSTHelper.FromString(src2.Substring(0, 100)));
initalgo.RunToCompletion();
var liLen1 = new List<int>();
var liLen2 = new List<int>();
var liLen3 = new List<int>();
int c = 0;
while(len < src1.Length && len < src2.Length)
{
switch (c++ % 3)
{
case 0:
liLen1.Add(len);
break;
case 1:
liLen2.Add(len);
break;
default:
liLen3.Add(len);
break;
}
len += 100;
}
Length1 = liLen1.ToArray();
Length2 = liLen2.ToArray();
Length3 = liLen3.ToArray();
var t1 = new Thread(Start1);
t1.Start();
var t2 = new Thread(Start2);
t2.Start();
var t3 = new Thread(Start3);
t3.Start();
Console.WriteLine("Main thread joining");
t1.Join();
t2.Join();
t3.Join();
File.WriteAllLines(@"test\default_set\performance.txt", ResultList);
Console.WriteLine("finished all runs");
Console.ReadLine();
}
