Merge pull request #175 from jplag/107-testcases

Add more initial test cases

jplag/src/test/java/jplag/NormalComparisonTest.java

@@ -2,10 +2,15 @@ package jplag;
 
 import org.junit.Test;
 
+import java.util.Optional;
+
 import static org.junit.Assert.assertEquals;
 
 public class NormalComparisonTest extends TestBase {
 
+    /**
+     * The simple duplicate contains obvious plagiarism.
+     */
     @Test
     public void testSimpleDuplicate() throws ExitException {
         JPlagResult result = runJPlagWithDefaultOptions("SimpleDuplicate");
@@ -14,6 +19,79 @@ public class NormalComparisonTest extends TestBase {
         assertEquals(1, result.getComparisons().size());
         assertEquals(1, result.getComparisons().get(0).matches.size());
         assertEquals(1, result.getSimilarityDistribution()[6]);
+        assertEquals(62.07f, result.getComparisons().get(0).percent(), 0.1f);
     }
 
+    /**
+     * The classes in no duplicate have nearly nothing in common.
+     */
+    @Test
+    public void testNoDuplicate() throws ExitException {
+        JPlagResult result = runJPlagWithDefaultOptions("NoDuplicate");
+
+        assertEquals(3, result.getNumberOfSubmissions());
+        assertEquals(3, result.getComparisons().size());
+
+        result.getAllComparisons().forEach(comparison -> {
+            assertEquals(0f, comparison.percent(), 0.1f);
+        });
+    }
+
+    /**
+     * This case is more complex and consists out of 5 submissions with different plagiarism.
+     * A is the original code (coming from an older JPlag version)
+     * B is a partial copy of that code
+     * C is a full copy of that code
+     * D is dumb plagiarism, e.g., changed variable names, additional unneeded code, ...
+     * E is just a Hello World Java program
+     */
+    @Test
+    public void partialPlagiarism() throws ExitException {
+        JPlagResult result = runJPlagWithDefaultOptions("PartialPlagiarism");
+
+        assertEquals(5, result.getNumberOfSubmissions());
+        assertEquals(10, result.getAllComparisons().size());
+
+        // All comparisons with E shall have no matches
+        result.getAllComparisons()
+                .stream()
+                .filter(comparison ->
+                        comparison.secondSubmission.name.equals("E") ||
+                                comparison.firstSubmission.name.equals("E"))
+                .forEach(comparison ->
+                        assertEquals(0f, comparison.percent(), 0.1f)
+                );
+
+        // Hard coded assertions on selected comparisons
+        assertEquals(24.6f, getSelectedPercent(result, "A", "B"), 0.1f);
+        assertEquals(99.7f, getSelectedPercent(result, "A", "C"), 0.1f);
+        assertEquals(77.9f, getSelectedPercent(result, "A", "D"), 0.1f);
+        assertEquals(24.6f, getSelectedPercent(result, "B", "C"), 0.1f);
+        assertEquals(28.3f, getSelectedPercent(result, "B", "D"), 0.1f);
+        assertEquals(77.9f, getSelectedPercent(result, "C", "D"), 0.1f);
+
+        // More detailed assertions for the plagiarism in A-D
+        var biggestMatch = getSelectedComparison(result, "A", "D");
+        assertEquals(96.4f, biggestMatch.get().percentMaxAB(), 0.1f);
+        assertEquals(65.3f, biggestMatch.get().percentMinAB(), 0.1f);
+        assertEquals(12, biggestMatch.get().matches.size());
+
+    }
+
+    // TODO SH: Methods like this should be moved to the API and also should accept wildcards
+    private float getSelectedPercent(JPlagResult result, String nameA, String nameB) {
+        return getSelectedComparison(result, nameA, nameB)
+                .map(JPlagComparison::percent)
+                .orElse(-1f);
+    }
+
+    private Optional<JPlagComparison> getSelectedComparison(JPlagResult result, String nameA, String nameB) {
+        return result.getAllComparisons().stream()
+                .filter(comparison ->
+                        comparison.firstSubmission.name.equals(nameA) &&
+                                comparison.secondSubmission.name.equals(nameB) ||
+                                comparison.firstSubmission.name.equals(nameB) &&
+                                        comparison.secondSubmission.name.equals(nameA))
+                .findFirst();
+    }
 }

jplag/src/test/java/jplag/OptionsTest.java

@@ -0,0 +1,5 @@
+package jplag;
+
+public class OptionsTest extends TestBase {
+    // TODO SH: There should be more tests testing with different combinations of JPlagOptions, e.g., base code
+}

jplag/src/test/resources/samples/NoDuplicate/A/AlsoNoDuplicate.java

@@ -0,0 +1,30 @@
+/**
+ * This is code from the TagParser class of Jplag
+ */
+public class AlsoNoDuplicate {
+    public static String parse(String message, String[] params) {
+        String[] tokens = message.split("[{}]", -1);
+        String result = tokens[0];
+
+        for (int i = 1; i < tokens.length; i += 2)    // Go to next tag position
+        {
+            try {
+                int ind = tokens[i].indexOf('_');
+                String num = (ind == -1) ? tokens[i] : tokens[i].substring(0, ind);
+                result += params[Integer.parseInt(num) - 1];
+            } catch (Exception ex) {
+                if (ex instanceof NumberFormatException || ex instanceof IndexOutOfBoundsException) {
+                    ex.printStackTrace();
+                    result += "{ILLEGAL PARAMETER INDEX \"" + tokens[i] + "\"}";
+                } else {
+                    throw (RuntimeException) ex;
+                }
+            }
+            if (i + 1 < tokens.length) {
+                result += tokens[i + 1];
+            }
+        }
+
+        return result;
+    }
+}

jplag/src/test/resources/samples/NoDuplicate/B/NoDuplicate.java

@@ -0,0 +1,5 @@
+public class NoDuplicate {
+    public static void main(String[] args) {
+        System.out.println("Hello World!");
+    }
+}

jplag/src/test/resources/samples/NoDuplicate/C/AnotherClass.java

@@ -0,0 +1,30 @@
+/**
+ * This is so pointless, I hope nobody looks at test cases, e.g., in 25 years from now
+ */
+public class AnotherClass {
+
+    public String getName() {
+        return name;
+    }
+
+    public void setName(String name) {
+        this.name = name;
+    }
+
+    public String getId() {
+        return id;
+    }
+
+    public void setId(String id) {
+        this.id = id;
+    }
+
+    private String name;
+    private String id;
+
+    public AnotherClass(String name, String id) {
+        this.name = name;
+        this.id = id;
+    }
+
+}

jplag/src/test/resources/samples/NoDuplicate/C/CertainlyNoDuplicate.java

@@ -0,0 +1,27 @@
+import java.util.Locale;
+import java.util.MissingResourceException;
+import java.util.ResourceBundle;
+
+/**
+ * This is code from the Messages class of Jplag
+ */
+public class CertainlyNoDuplicate {
+
+    private final ResourceBundle resourceBundle;
+
+    /**
+     * @param countryTag may be "de", "en", "fr", "es", "pt" or "ptbr"
+     */
+    public CertainlyNoDuplicate(String countryTag) {
+        String bundleName = "jplag.messages";
+        resourceBundle = ResourceBundle.getBundle(bundleName, new Locale(countryTag));
+    }
+
+    public String getString(String key) {
+        try {
+            return resourceBundle.getString(key);
+        } catch (MissingResourceException e) {
+            return '!' + key + '!';
+        }
+    }
+}

jplag/src/test/resources/samples/PartialPlagiarism/A/GSTiling.java

@@ -0,0 +1,247 @@
+/*
+ * Diese Klasse implementiert den GSTiling Algorithmus
+ * Allerdings ist sie sehr spezifisch auf die Klassen "Structure", "Token",
+ * sowie "Matches" und "Match" ausgelegt
+ */
+public class GSTiling implements TokenConstants {
+	private Matches matches = new Matches();
+
+	/*
+	 * creating hashes.
+	 * The hash-code will be written in every Token for the next <hash_length> token.
+	 * (includes the Token itself)
+	 * Das Ganze laeuft in linearer Zeit.
+	 * condition: 1 < hashLength < 26   !!!
+	 */
+
+	public void create_hashes(Structure s, int hashLength, boolean makeTable) {
+		// Hier wird die obere Grenze der Hash-Laenge festgelegt.
+		// Sie ist bestimmt durch die Bitzahl des 'int' Datentyps und der Anzahl
+		// der Token.
+		if (hashLength<1) hashLength = 1;
+		hashLength = (hashLength<26 ? hashLength : 25);
+
+		if (s.size()<hashLength) return;
+
+		int modulo = ((1 << 6) - 1);   // Modulo 64!
+
+		int loops = s.size()-hashLength;
+		s.table = (makeTable ? new Table(3*loops) : null);
+		int hash = 0;
+		int i;
+		int hashedLength = 0;
+		for (i=0; i<hashLength; i++) {
+			hash = (2*hash) + (s.tokens[i].type & modulo);
+			hashedLength++;
+			if (s.tokens[i].marked)
+				hashedLength = 0;
+		}
+		int factor = (hashLength != 1 ? (2<<(hashLength-2)) : 1);
+
+		if (makeTable) {
+			for (i=0; i<loops; i++) {
+				if (hashedLength >= hashLength) {
+					s.tokens[i].hash = hash;
+					s.table.add(hash, i);   // add into hashtable
+				} else
+					s.tokens[i].hash = -1;
+				hash -= factor * (s.tokens[i].type & modulo);
+				hash = (2*hash) + (s.tokens[i+hashLength].type & modulo);
+				if (s.tokens[i+hashLength].marked)
+					hashedLength = 0;
+				else
+					hashedLength++;
+			}
+		} else {
+			for (i=0; i<loops; i++) {
+				s.tokens[i].hash = (hashedLength >= hashLength) ? hash : -1;
+				hash -= factor * (s.tokens[i].type & modulo);
+				hash = (2*hash) + (s.tokens[i+hashLength].type & modulo);
+				if (s.tokens[i+hashLength].marked)
+					hashedLength = 0;
+				else
+					hashedLength++;
+			}
+		}
+		s.hash_length = hashLength;
+	}
+
+	public final void compare(Submission subA, Submission subB) {
+		Submission A, B, tmp;
+		if (subA.struct.size() > subB.struct.size()) {
+			A = subB;  B = subA;
+		} else {
+			A = subB;  B = subA;
+		}
+		// if hashtable exists in first but not in second structure: flip around!
+		if (B.struct.table == null && A.struct.table != null) {
+			tmp = A;
+			A = B;
+			B = tmp;
+		}
+	}
+
+	// first parameter should contain the smaller sequence!!! 
+	private final void compare(Submission subA, Submission subB, int mml) {
+		Structure structA = subA.struct;
+		Structure structB = subB.struct;
+
+		// FILE_END used as pivot
+
+		// init
+		Token[] A = structA.tokens;
+		Token[] B = structB.tokens;
+		int lengthA = structA.size()-1;  // minus pivots!
+		int lengthB = structB.size()-1;  // minus pivots!
+
+
+		if(true) {
+			for(int i = 0; i <= lengthA; i++)
+				A[i].marked = A[i].type == FILE_END || A[i].type == SEPARATOR_TOKEN;
+
+			for(int i = 0; i <= lengthB; i++)
+				B[i].marked = B[i].type == FILE_END || B[i].type == SEPARATOR_TOKEN;
+		} else {
+			for(int i = 0; i <= lengthA; i++)
+				A[i].marked = A[i].type == FILE_END || A[i].type == SEPARATOR_TOKEN || A[i].basecode;
+
+			for(int i = 0; i <= lengthB; i++)
+				B[i].marked = B[i].type == FILE_END || B[i].type == SEPARATOR_TOKEN || B[i].basecode;
+		}
+
+
+		int maxmatch;
+		int[] elemsB;
+
+		do {
+			maxmatch = mml;
+			matches.clear();
+			for (int x = 0; x <= lengthA - maxmatch; x++) {
+				if (A[x].marked || A[x].hash == -1
+						|| (elemsB = structB.table.get(A[x].hash)) == null)
+					continue;
+inner:			for (int i = 1; i <= elemsB[0]; i++) { // elemsB[0] contains the length of the Array
+					int y = elemsB[i];
+					if (B[y].marked || maxmatch > lengthB - y) continue;
+
+					int j, hx, hy;
+					for (j = maxmatch - 1; j >= 0; j--) { //begins comparison from behind
+						if (A[hx = x + j].type != B[hy = y + j].type || A[hx].marked || B[hy].marked)
+							continue inner;
+					}
+
+					// expand match
+					j = maxmatch;
+					while(A[hx = x + j].type == B[hy = y + j].type && !A[hx].marked && !B[hy].marked)
+						j++;
+
+					if (j > maxmatch) {  // new biggest match? -> delete current smaller
+						matches.clear();
+						maxmatch = j;
+					}
+					matches.addMatch(x, y, j);  // add match
+				}
+			}
+			for (int i = matches.size() - 1; i >= 0; i--) {
+				int x = matches.matches[i].startA;  // begining of sequence A
+				int y = matches.matches[i].startB;  // begining of sequence B
+
+				//in order that "Match" will be newly build     (because reusing)
+				for (int j = matches.matches[i].length; j > 0; j--)
+					A[x++].marked = B[y++].marked = true;   // mark all Token!
+			}
+
+		} while (maxmatch != mml);
+	}
+
+	public final void compareWithBasecode(Submission subA, Submission subB) {
+		Submission A, B, tmp;
+		if (subA.struct.size() > subB.struct.size()) {
+			A = subB;  B = subA;
+		} else {
+			A = subB;  B = subA;
+		}
+		// if hashtable exists in first but not in second structure: flip around!
+		if (B.struct.table == null && A.struct.table != null) {
+			tmp = A;
+			A = B;
+			B = tmp;
+		}
+	}
+
+	private final void compareWithBasecode(Submission subA, Submission subB, int mml) {
+		Structure structA = subA.struct;
+		Structure structB = subB.struct;
+
+		// FILE_END used as pivot
+
+		// init
+		Token[] A = structA.tokens;
+		Token[] B = structB.tokens;
+		int lengthA = structA.size()-1;  // minus pivots!
+		int lengthB = structB.size()-1;  // minus pivots!
+
+
+		// Initialize
+		for(int i = 0; i <= lengthA; i++)
+			A[i].marked = A[i].type == FILE_END || A[i].type == SEPARATOR_TOKEN;
+
+		for(int i = 0; i <= lengthB; i++)
+			B[i].marked = B[i].type == FILE_END || B[i].type == SEPARATOR_TOKEN;
+
+
+
+
+		int maxmatch;
+		int[] elemsB;
+
+		do {
+			maxmatch = mml;
+			matches.clear();
+			for (int x = 0; x <= lengthA - maxmatch; x++) {
+				if (A[x].marked || A[x].hash == -1 || (elemsB = structB.table.get(A[x].hash)) == null)
+					continue;
+inner:			for (int i = 1; i <= elemsB[0]; i++) {// elemsB[0] contains the length of the Array
+					int y = elemsB[i];
+					if (B[y].marked || maxmatch > lengthB - y) continue;
+
+					int j,hx,hy;
+					for (j = maxmatch - 1; j >= 0; j--) { // begins comparison from behind
+						if (A[hx = x + j].type != B[hy = y + j].type || A[hx].marked || B[hy].marked)
+							continue inner;
+					}
+					// expand match
+					j = maxmatch;
+					while(A[hx = x + j].type == B[hy = y + j].type && !A[hx].marked && !B[hy].marked)
+						j++;
+
+					if (j != maxmatch) {  // new biggest match? -> delete current smaller
+						matches.clear();
+						maxmatch = j;
+					}
+					matches.addMatch(x, y, j);  // add match
+				}
+			}
+			for (int i = matches.size() - 1; i >= 0; i--) {
+				int x = matches.matches[i].startA;  // beginning in sequence A
+				int y = matches.matches[i].startB;  // beginning in sequence B
+
+				//in order that "Match" will be newly build     (because reusing)
+				for (int j = matches.matches[i].length; j > 0; j--){
+					A[x].marked = B[y].marked = true;   // mark all Token!
+					A[x].basecode = B[y].basecode= true;
+					x++; y++;
+				}
+			}
+		} while (maxmatch != mml);
+
+	}
+
+	public void resetBaseSubmission(Submission sub){
+		Structure tmpStruct = sub.struct;
+		Token[] tok = tmpStruct.tokens;
+		for (int z = 0; z < tmpStruct.size()-1;z++){
+			tok[z].basecode = false;
+		}
+	}
+}

jplag/src/test/resources/samples/PartialPlagiarism/A/Match.java

@@ -0,0 +1,52 @@
+public class Match {
+  public int startA;
+  public int startB;
+  public int length;
+
+  public Match(int startA, int startB, int length) {
+    this.startA = startA;
+    this.startB = startB;
+    this.length = length;
+  }
+  public Match() {}
+
+  public void set(int startA, int startB, int length) {
+    this.startA = startA;
+    this.startB = startB;
+    this.length = length;
+  }
+
+  public final boolean contains(int index, int sub) {
+    int start = (sub==0 ? startA : startB);
+    return (start <= index && index < (start+length));
+  }
+
+  public final boolean overlap(Match other) {
+    if (startA < other.startA) {
+      if ((other.startA - startA) < length) return true;
+    } else {
+      if ((startA - other.startA) < other.length) return true;
+    }
+    if (startB < other.startB) {
+      if ((other.startB - startB) < length) return true;
+    } else {
+      if ((startB - other.startB) < other.length) return true;
+    }
+    return false;
+  }
+
+  public final boolean overlap(int oStartA, int oStartB, int oLength) {
+    if (startA < oStartA) {
+      if ((oStartA - startA) < length) return true;
+    } else {
+      if ((startA - oStartA) < oLength) return true;
+    }
+    if (startB < oStartB) {
+      if ((oStartB - startB) < length) return true;
+    } else {
+      if ((startB - oStartB) < oLength) return true;
+    }
+    return false;
+  }
+}
+

jplag/src/test/resources/samples/PartialPlagiarism/A/Matches.java

@@ -0,0 +1,57 @@
+/* Minimal class that stores "Match" objects.
+ */
+public class Matches {
+  public Match[] matches;
+
+  private int anzahl;
+  private final int increment = 20;
+  
+  public Matches() {
+    matches = new Match[10];
+    for (int i=0; i<10; i++) matches[i] = new Match();
+    anzahl = 0;
+  }
+
+  public final int size() {
+    return anzahl;
+  }
+
+  public final void ensureCapacity(int minCapacity) {
+    int oldCapacity = matches.length;
+    if (minCapacity > oldCapacity) {
+      Match[] oldMatches = matches;
+      int newCapacity = (oldCapacity + increment);
+      if (newCapacity < minCapacity) {
+	newCapacity = minCapacity;
+      }
+      matches = new Match[newCapacity];
+      System.arraycopy(oldMatches, 0, matches, 0, oldCapacity);
+      for (int i=oldCapacity; i<newCapacity; i++)
+	matches[i] = new Match();
+    }
+  }
+  /*
+  public final void addMatch(Match match) {
+    for (int i=0; i<anzahl; i++)
+      if (match.overlap(matches[i])) return;  // do not allow overlaps
+    ensureCapacity(anzahl + 1);
+    matches[anzahl++] = match;
+  }
+  */
+  public final void addMatch(int startA, int startB, int length) {
+    for (int i=anzahl-1; i>=0; i--) { // starting at the end is better(?)
+      if (matches[i].overlap(startA,startB,length)) return;
+                                      // no overlaps!
+    }
+    ensureCapacity(anzahl + 1);
+    //if (matches[anzahl] != null)  // object recycling...
+    matches[anzahl].set(startA,startB,length);
+    //else
+    //matches[anzahl ] = new Match(startA,startB,length);
+    anzahl++;
+  }
+
+  public final void clear() {
+    anzahl=0;
+  }
+}

jplag/src/test/resources/samples/PartialPlagiarism/A/Structure.java

@@ -0,0 +1,118 @@
+import java.io.*;
+
+/** The tokenlist */
+public class Structure implements TokenConstants {
+	public Token[] tokens = new Token[0];
+	Table table = null;
+	int hash_length = -1;
+
+	int files; // number of END_FILE tokens
+	private int anzahl;
+
+	//private final int increment = 300;
+
+	public Structure() {
+		tokens = new Token[400];
+		files = anzahl = 0;
+	}
+
+	public final int size() {
+		return anzahl;
+	}
+
+	public final void ensureCapacity(int minCapacity) {
+		int oldCapacity = tokens.length;
+		if (minCapacity > oldCapacity) {
+			Token[] oldTokens = tokens;
+			int newCapacity = (2 * oldCapacity);
+			if (newCapacity < minCapacity) {
+				newCapacity = minCapacity;
+			}
+			tokens = new Token[newCapacity];
+			System.arraycopy(oldTokens, 0, tokens, 0, anzahl);
+		}
+	}
+
+	public final void addToken(Token token) {
+		ensureCapacity(anzahl + 1);
+		if (anzahl > 0 && tokens[anzahl - 1].file.equals(token.file))
+			token.file = tokens[anzahl - 1].file; // To save memory ...
+		if ((anzahl > 0) && (token.getLine() < tokens[anzahl - 1].getLine()) && (token.file.equals(tokens[anzahl - 1].file)))
+			token.setLine(tokens[anzahl - 1].getLine());
+		// just to make sure
+
+		tokens[anzahl++] = token;
+		if (token.type == FILE_END)
+			files++;
+	}
+
+	public final String toString() {
+		StringBuffer buf = new StringBuffer();
+
+		try {
+			for (int i = 0; i < anzahl; i++) {
+				String s = tokens[i].toString();
+				buf.append(i);
+				buf.append("\t");
+				buf.append(s);
+				if (i < anzahl - 1) {
+					buf.append("\n");
+				}
+			}
+		} catch (OutOfMemoryError e) {
+			return "Tokenlist to large for output: " + (anzahl) + " Tokens";
+		}
+		return buf.toString();
+	}
+
+	public void save(File file) {
+
+		try {
+			ObjectOutputStream p = new ObjectOutputStream(/* new GZIPOutputStream */(new FileOutputStream(file)));
+
+			p.writeInt(anzahl);
+			p.writeInt(hash_length);
+			p.writeInt(files);
+
+			for (int i = 0; i < anzahl; i++)
+				p.writeObject(tokens[i]);
+			p.flush();
+			p.close();
+		} catch (IOException e) {
+			System.out.println("Error writing file: " + file.toString());
+		}
+	}
+
+	/* returns "true" when successful */
+	public boolean load(File file) {
+		try {
+			ObjectInputStream p = new ObjectInputStream(/* new GZIPInputStream */(new FileInputStream(file)));
+
+			int newAnzahl = p.readInt();
+			hash_length = p.readInt();
+			files = p.readInt();
+			ensureCapacity(newAnzahl);
+			anzahl = newAnzahl;
+			for (int i = 0; i < anzahl; i++) {
+				tokens[i] = (Token) p.readObject();
+				// special case for text tokens:
+				// 	if (tokens[i] instanceof jplag.text.TextToken) {
+				// 	  jplag.text.TextToken token = (jplag.text.TextToken)tokens[i];
+				// 	  jplag.text.TextToken.put(token.getText(), token.type);
+				// 	}
+			}
+			p.close();
+			table = null;
+		} catch (FileNotFoundException e) {
+			System.out.println("File not found: " + file.toString());
+			return false;
+		} catch (IOException e) {
+			System.out.println("Error reading file: " + file.toString() + " (" + e + ")");
+			return false;
+		} catch (ClassNotFoundException e) {
+			System.out.println("Class not found in file: " + file.toString());
+			return false;
+		}
+		return true;
+	}
+}

jplag/src/test/resources/samples/PartialPlagiarism/A/Submission.java

@@ -0,0 +1,218 @@
+import java.io.*;
+import java.net.URL;
+import java.text.DecimalFormat;
+import java.util.Vector;
+
+/*
+ * Everything about a single submission is stored in this object. (directory,
+ * files, ...)
+ */
+public class Submission implements Comparable<Submission> {
+	public String name;
+
+	private boolean readSubDirs;
+
+	public File dir;
+
+	public String[] files = null; // = new String[0];
+
+	public Structure struct;
+
+	public int structSize = 0;
+
+	// public long structMem;
+	boolean exact_match = false; // used for fallback
+
+	public boolean errors = false;
+
+	public DecimalFormat format = new DecimalFormat("0000");
+
+	public Submission(String name, File dir, boolean readSubDirs) {
+		this.dir = dir;
+		this.name = name;
+		this.readSubDirs = readSubDirs;
+		try {
+			lookupDir(dir, "");
+		} catch (Throwable b) {
+		}
+	}
+
+	public Submission(String name, File dir) {
+		this.dir = dir;
+		this.name = name;
+		this.readSubDirs = false;
+
+		files = new String[1];
+		files[0] = name;
+
+	}
+
+	// recursively read in all the files
+	private void lookupDir(File dir, String subDir) throws Throwable {
+		File aktDir = new File(dir, subDir);
+		if (!aktDir.isDirectory())
+			return;
+		if (readSubDirs) {
+			String[] dirs = aktDir.list();
+			if (!subDir.equals(""))
+				for (int i = 0; i < dirs.length; i++)
+					lookupDir(dir, subDir + File.separator + dirs[i]);
+			else
+				for (int i = 0; i < dirs.length; i++)
+					lookupDir(dir, dirs[i]);
+		}
+		String[] newFiles = aktDir.list(new FilenameFilter() {
+			public boolean accept(File dir, String name) {
+					return false;
+			}
+		});
+		if (files != null) {
+			String[] oldFiles = files;
+			files = new String[oldFiles.length + newFiles.length];
+			if (subDir != "")
+				for (int i = 0; i < newFiles.length; i++)
+					files[i] = subDir + File.separator + newFiles[i];
+			else
+				System.arraycopy(newFiles, 0, files, 0, newFiles.length);
+			System.arraycopy(oldFiles, 0, files, newFiles.length, oldFiles.length);
+		} else {
+			if (subDir != "") {
+				files = new String[newFiles.length];
+				for (int i = 0; i < newFiles.length; i++)
+					files[i] = subDir + File.separator + newFiles[i];
+			} else
+				files = newFiles;
+		}
+	}
+
+	/* parse all the files... */
+	public boolean parse(){
+		return false;
+	}
+
+	/*
+	 * This method is used to copy files that can not be parsed to a special
+	 * folder: jplag/errors/java old_java scheme cpp /001/(...files...)
+	 * /002/(...files...)
+	 */
+	private void copySubmission() {
+		File errorDir = null;
+		try {
+			URL url = Submission.class.getResource("errors");
+			errorDir = new File(url.getFile());
+		} catch (NullPointerException e) {
+			return;
+		}
+
+		if (!errorDir.exists())
+			errorDir.mkdir();
+		int i = 0;
+		File destDir;
+		while ((destDir = new File(errorDir, format.format(i))).exists())
+			i++;
+		destDir.mkdir();
+		for (i = 0; i < files.length; i++)
+			copyFile(new File(dir, files[i]), new File(destDir, files[i]));
+	}
+
+	/* Physical copy. :-) */
+	private void copyFile(File in, File out) {
+		byte[] buffer = new byte[10000];
+		try {
+			FileInputStream dis = new FileInputStream(in);
+			FileOutputStream dos = new FileOutputStream(out);
+			int count;
+			do {
+				count = dis.read(buffer);
+				if (count != -1)
+					dos.write(buffer, 0, count);
+			} while (count != -1);
+			dis.close();
+			dos.close();
+		} catch (IOException e) {
+
+		}
+	}
+
+	public int size() {
+		if (struct != null)
+			return structSize = struct.size();
+		return structSize;
+	}
+
+	/*
+	 * Used by the "Report" class. All source files are returned as an array of
+	 * an array of strings.
+	 */
+	public String[][] readFiles(String[] files) {
+		String[][] result = new String[files.length][];
+		String help;
+
+		Vector<String> text = new Vector<String>();
+		for (int i = 0; i < files.length; i++) {
+			text.removeAllElements();
+			try {
+				/* file encoding = "UTF-8" */
+				FileInputStream fileInputStream = new FileInputStream(new File(dir, files[i]));
+				InputStreamReader inputStreamReader = new InputStreamReader(fileInputStream, "UTF-8");
+				BufferedReader in = new BufferedReader(inputStreamReader);
+				while ((help = in.readLine()) != null) {
+					help = help.replaceAll("&", "&amp;");
+					help = help.replaceAll("<", "&lt;");
+					help = help.replaceAll(">", "&gt;");
+					help = help.replaceAll("\"", "&quot;");
+					text.addElement(help);
+				}
+				in.close();
+				inputStreamReader.close();
+				fileInputStream.close();
+			} catch (FileNotFoundException e) {
+				System.out.println("File not found: " + ((new File(dir, files[i])).toString()));
+			} catch (IOException e) {
+
+			}
+			result[i] = new String[text.size()];
+			text.copyInto(result[i]);
+		}
+		return result;
+	}
+
+	/*
+	 * Used by the "Report" class. All source files are returned as an array of
+	 * an array of chars.
+	 */
+	public char[][] readFilesChar(String[] files) {
+		char[][] result = new char[files.length][];
+
+		for (int i = 0; i < files.length; i++) {
+			try {
+				File file = new File(dir, files[i]);
+				int size = (int) file.length();
+				char[] buffer = new char[size];
+
+				FileReader fis = new FileReader(file);
+
+				if (size != fis.read(buffer)) {
+					System.out.println("Not right size read from the file, " + "but I will still continue...");
+				}
+
+				result[i] = buffer;
+				fis.close();
+			} catch (FileNotFoundException e) {
+				// TODO: Should an ExitException be thrown here?
+				System.out.println("File not found: " + ((new File(dir, files[i])).toString()));
+			} catch (IOException e) {
+
+			}
+		}
+		return result;
+	}
+
+	public int compareTo(Submission other) {
+		return name.compareTo(other.name);
+	}
+
+	public String toString() {
+		return name;
+	}
+}

jplag/src/test/resources/samples/PartialPlagiarism/A/Token.java

@@ -0,0 +1,47 @@
+import java.io.Serializable;
+
+abstract public class Token implements TokenConstants, Serializable {
+	private static final long serialVersionUID = 862935679966383302L;
+	public int type;
+	public String file;
+	
+	protected boolean marked;
+	protected boolean basecode = false;
+	protected int hash = -1;// hash-value. set and used by main algorithm (GSTiling)
+	
+	public Token(int type, String file, int line) {
+		this(type, file, line, -1, -1);
+	}
+	
+	public Token(int type, String file, int line, int column, int length) {
+		this.type = type;
+		this.file = file;
+		setLine(line>0 ? line : 1);
+		setColumn(column);
+		setLength(length);
+	}
+	
+	// abstract members
+	abstract public int getLine();
+	abstract public int getColumn();
+	abstract public int getLength();
+	protected void setLine(int line) {}
+	protected void setColumn(int line) {}
+	protected void setLength(int line) {}
+	
+	// this is made to distinguish the character front end.
+	// maybe other front ends can use it too?
+	protected int getIndex() { return -1; }
+	
+	public static String type2string(int type) {
+		return "<abstract>";
+	}
+	
+	public String toString() {
+		return type2string(type);
+	}
+	
+	public static int numberOfTokens() { 
+		return 1;
+	} 
+}

jplag/src/test/resources/samples/PartialPlagiarism/A/TokenConstants.java

@@ -0,0 +1,8 @@
+public interface TokenConstants {
+  final static int FILE_END = 0;
+	/**
+	 * Used to optionally separate methods from each other with an always marked
+	 * token
+	 */
+  final static int SEPARATOR_TOKEN = 1;
+}

jplag/src/test/resources/samples/PartialPlagiarism/B/Token.java

@@ -0,0 +1,47 @@
+import java.io.Serializable;
+
+abstract public class Token implements TokenConstants, Serializable {
+	private static final long serialVersionUID = 862935679966383302L;
+	public int type;
+	public String file;
+	
+	protected boolean marked;
+	protected boolean basecode = false;
+	protected int hash = -1;// hash-value. set and used by main algorithm (GSTiling)
+	
+	public Token(int type, String file, int line) {
+		this(type, file, line, -1, -1);
+	}
+	
+	public Token(int type, String file, int line, int column, int length) {
+		this.type = type;
+		this.file = file;
+		setLine(line>0 ? line : 1);
+		setColumn(column);
+		setLength(length);
+	}
+	
+	// abstract members
+	abstract public int getLine();
+	abstract public int getColumn();
+	abstract public int getLength();
+	protected void setLine(int line) {}
+	protected void setColumn(int line) {}
+	protected void setLength(int line) {}
+	
+	// this is made to distinguish the character front end.
+	// maybe other front ends can use it too?
+	protected int getIndex() { return -1; }
+	
+	public static String type2string(int type) {
+		return "<abstract>";
+	}
+	
+	public String toString() {
+		return type2string(type);
+	}
+	
+	public static int numberOfTokens() { 
+		return 1;
+	} 
+}

jplag/src/test/resources/samples/PartialPlagiarism/B/TokenConstants.java

@@ -0,0 +1,8 @@
+public interface TokenConstants {
+  final static int FILE_END = 0;
+	/**
+	 * Used to optionally separate methods from each other with an always marked
+	 * token
+	 */
+  final static int SEPARATOR_TOKEN = 1;
+}

jplag/src/test/resources/samples/PartialPlagiarism/C/GSTiling.java

@@ -0,0 +1,247 @@
+/*
+ * Diese Klasse implementiert den GSTiling Algorithmus
+ * Allerdings ist sie sehr spezifisch auf die Klassen "Structure", "Token",
+ * sowie "Matches" und "Match" ausgelegt
+ */
+public class GSTiling implements TokenConstants {
+	private Matches matches = new Matches();
+
+	/*
+	 * creating hashes.
+	 * The hash-code will be written in every Token for the next <hash_length> token.
+	 * (includes the Token itself)
+	 * Das Ganze laeuft in linearer Zeit.
+	 * condition: 1 < hashLength < 26   !!!
+	 */
+
+	public void create_hashes(Structure s, int hashLength, boolean makeTable) {
+		// Hier wird die obere Grenze der Hash-Laenge festgelegt.
+		// Sie ist bestimmt durch die Bitzahl des 'int' Datentyps und der Anzahl
+		// der Token.
+		if (hashLength<1) hashLength = 1;
+		hashLength = (hashLength<26 ? hashLength : 25);
+
+		if (s.size()<hashLength) return;
+
+		int modulo = ((1 << 6) - 1);   // Modulo 64!
+
+		int loops = s.size()-hashLength;
+		s.table = (makeTable ? new Table(3*loops) : null);
+		int hash = 0;
+		int i;
+		int hashedLength = 0;
+		for (i=0; i<hashLength; i++) {
+			hash = (2*hash) + (s.tokens[i].type & modulo);
+			hashedLength++;
+			if (s.tokens[i].marked)
+				hashedLength = 0;
+		}
+		int factor = (hashLength != 1 ? (2<<(hashLength-2)) : 1);
+
+		if (makeTable) {
+			for (i=0; i<loops; i++) {
+				if (hashedLength >= hashLength) {
+					s.tokens[i].hash = hash;
+					s.table.add(hash, i);   // add into hashtable
+				} else
+					s.tokens[i].hash = -1;
+				hash -= factor * (s.tokens[i].type & modulo);
+				hash = (2*hash) + (s.tokens[i+hashLength].type & modulo);
+				if (s.tokens[i+hashLength].marked)
+					hashedLength = 0;
+				else
+					hashedLength++;
+			}
+		} else {
+			for (i=0; i<loops; i++) {
+				s.tokens[i].hash = (hashedLength >= hashLength) ? hash : -1;
+				hash -= factor * (s.tokens[i].type & modulo);
+				hash = (2*hash) + (s.tokens[i+hashLength].type & modulo);
+				if (s.tokens[i+hashLength].marked)
+					hashedLength = 0;
+				else
+					hashedLength++;
+			}
+		}
+		s.hash_length = hashLength;
+	}
+
+	public final void compare(Submission subA, Submission subB) {
+		Submission A, B, tmp;
+		if (subA.struct.size() > subB.struct.size()) {
+			A = subB;  B = subA;
+		} else {
+			A = subB;  B = subA;
+		}
+		// if hashtable exists in first but not in second structure: flip around!
+		if (B.struct.table == null && A.struct.table != null) {
+			tmp = A;
+			A = B;
+			B = tmp;
+		}
+	}
+
+	// first parameter should contain the smaller sequence!!! 
+	private final void compare(Submission subA, Submission subB, int mml) {
+		Structure structA = subA.struct;
+		Structure structB = subB.struct;
+
+		// FILE_END used as pivot
+
+		// init
+		Token[] A = structA.tokens;
+		Token[] B = structB.tokens;
+		int lengthA = structA.size()-1;  // minus pivots!
+		int lengthB = structB.size()-1;  // minus pivots!
+
+
+		if(true) {
+			for(int i = 0; i <= lengthA; i++)
+				A[i].marked = A[i].type == FILE_END || A[i].type == SEPARATOR_TOKEN;
+
+			for(int i = 0; i <= lengthB; i++)
+				B[i].marked = B[i].type == FILE_END || B[i].type == SEPARATOR_TOKEN;
+		} else {
+			for(int i = 0; i <= lengthA; i++)
+				A[i].marked = A[i].type == FILE_END || A[i].type == SEPARATOR_TOKEN || A[i].basecode;
+
+			for(int i = 0; i <= lengthB; i++)
+				B[i].marked = B[i].type == FILE_END || B[i].type == SEPARATOR_TOKEN || B[i].basecode;
+		}
+
+
+		int maxmatch;
+		int[] elemsB;
+
+		do {
+			maxmatch = mml;
+			matches.clear();
+			for (int x = 0; x <= lengthA - maxmatch; x++) {
+				if (A[x].marked || A[x].hash == -1
+						|| (elemsB = structB.table.get(A[x].hash)) == null)
+					continue;
+inner:			for (int i = 1; i <= elemsB[0]; i++) { // elemsB[0] contains the length of the Array
+					int y = elemsB[i];
+					if (B[y].marked || maxmatch > lengthB - y) continue;
+
+					int j, hx, hy;
+					for (j = maxmatch - 1; j >= 0; j--) { //begins comparison from behind
+						if (A[hx = x + j].type != B[hy = y + j].type || A[hx].marked || B[hy].marked)
+							continue inner;
+					}
+
+					// expand match
+					j = maxmatch;
+					while(A[hx = x + j].type == B[hy = y + j].type && !A[hx].marked && !B[hy].marked)
+						j++;
+
+					if (j > maxmatch) {  // new biggest match? -> delete current smaller
+						matches.clear();
+						maxmatch = j;
+					}
+					matches.addMatch(x, y, j);  // add match
+				}
+			}
+			for (int i = matches.size() - 1; i >= 0; i--) {
+				int x = matches.matches[i].startA;  // begining of sequence A
+				int y = matches.matches[i].startB;  // begining of sequence B
+
+				//in order that "Match" will be newly build     (because reusing)
+				for (int j = matches.matches[i].length; j > 0; j--)
+					A[x++].marked = B[y++].marked = true;   // mark all Token!
+			}
+
+		} while (maxmatch != mml);
+	}
+
+	public final void compareWithBasecode(Submission subA, Submission subB) {
+		Submission A, B, tmp;
+		if (subA.struct.size() > subB.struct.size()) {
+			A = subB;  B = subA;
+		} else {
+			A = subB;  B = subA;
+		}
+		// if hashtable exists in first but not in second structure: flip around!
+		if (B.struct.table == null && A.struct.table != null) {
+			tmp = A;
+			A = B;
+			B = tmp;
+		}
+	}
+
+	private final void compareWithBasecode(Submission subA, Submission subB, int mml) {
+		Structure structA = subA.struct;
+		Structure structB = subB.struct;
+
+		// FILE_END used as pivot
+
+		// init
+		Token[] A = structA.tokens;
+		Token[] B = structB.tokens;
+		int lengthA = structA.size()-1;  // minus pivots!
+		int lengthB = structB.size()-1;  // minus pivots!
+
+
+		// Initialize
+		for(int i = 0; i <= lengthA; i++)
+			A[i].marked = A[i].type == FILE_END || A[i].type == SEPARATOR_TOKEN;
+
+		for(int i = 0; i <= lengthB; i++)
+			B[i].marked = B[i].type == FILE_END || B[i].type == SEPARATOR_TOKEN;
+
+
+
+
+		int maxmatch;
+		int[] elemsB;
+
+		do {
+			maxmatch = mml;
+			matches.clear();
+			for (int x = 0; x <= lengthA - maxmatch; x++) {
+				if (A[x].marked || A[x].hash == -1 || (elemsB = structB.table.get(A[x].hash)) == null)
+					continue;
+inner:			for (int i = 1; i <= elemsB[0]; i++) {// elemsB[0] contains the length of the Array
+					int y = elemsB[i];
+					if (B[y].marked || maxmatch > lengthB - y) continue;
+
+					int j,hx,hy;
+					for (j = maxmatch - 1; j >= 0; j--) { // begins comparison from behind
+						if (A[hx = x + j].type != B[hy = y + j].type || A[hx].marked || B[hy].marked)
+							continue inner;
+					}
+					// expand match
+					j = maxmatch;
+					while(A[hx = x + j].type == B[hy = y + j].type && !A[hx].marked && !B[hy].marked)
+						j++;
+
+					if (j != maxmatch) {  // new biggest match? -> delete current smaller
+						matches.clear();
+						maxmatch = j;
+					}
+					matches.addMatch(x, y, j);  // add match
+				}
+			}
+			for (int i = matches.size() - 1; i >= 0; i--) {
+				int x = matches.matches[i].startA;  // beginning in sequence A
+				int y = matches.matches[i].startB;  // beginning in sequence B
+
+				//in order that "Match" will be newly build     (because reusing)
+				for (int j = matches.matches[i].length; j > 0; j--){
+					A[x].marked = B[y].marked = true;   // mark all Token!
+					A[x].basecode = B[y].basecode= true;
+					x++; y++;
+				}
+			}
+		} while (maxmatch != mml);
+
+	}
+
+	public void resetBaseSubmission(Submission sub){
+		Structure tmpStruct = sub.struct;
+		Token[] tok = tmpStruct.tokens;
+		for (int z = 0; z < tmpStruct.size()-1;z++){
+			tok[z].basecode = false;
+		}
+	}
+}

jplag/src/test/resources/samples/PartialPlagiarism/C/Match.java

@@ -0,0 +1,52 @@
+public class Match {
+  public int startA;
+  public int startB;
+  public int length;
+
+  public Match(int startA, int startB, int length) {
+    this.startA = startA;
+    this.startB = startB;
+    this.length = length;
+  }
+  public Match() {}
+
+  public void set(int startA, int startB, int length) {
+    this.startA = startA;
+    this.startB = startB;
+    this.length = length;
+  }
+
+  public final boolean contains(int index, int sub) {
+    int start = (sub==0 ? startA : startB);
+    return (start <= index && index < (start+length));
+  }
+
+  public final boolean overlap(Match other) {
+    if (startA < other.startA) {
+      if ((other.startA - startA) < length) return true;
+    } else {
+      if ((startA - other.startA) < other.length) return true;
+    }
+    if (startB < other.startB) {
+      if ((other.startB - startB) < length) return true;
+    } else {
+      if ((startB - other.startB) < other.length) return true;
+    }
+    return false;
+  }
+
+  public final boolean overlap(int oStartA, int oStartB, int oLength) {
+    if (startA < oStartA) {
+      if ((oStartA - startA) < length) return true;
+    } else {
+      if ((startA - oStartA) < oLength) return true;
+    }
+    if (startB < oStartB) {
+      if ((oStartB - startB) < length) return true;
+    } else {
+      if ((startB - oStartB) < oLength) return true;
+    }
+    return false;
+  }
+}
+

jplag/src/test/resources/samples/PartialPlagiarism/C/Matches.java

@@ -0,0 +1,57 @@
+/* Minimal class that stores "Match" objects.
+ */
+public class Matches {
+  public Match[] matches;
+
+  private int anzahl;
+  private final int increment = 20;
+  
+  public Matches() {
+    matches = new Match[10];
+    for (int i=0; i<10; i++) matches[i] = new Match();
+    anzahl = 0;
+  }
+
+  public final int size() {
+    return anzahl;
+  }
+
+  public final void ensureCapacity(int minCapacity) {
+    int oldCapacity = matches.length;
+    if (minCapacity > oldCapacity) {
+      Match[] oldMatches = matches;
+      int newCapacity = (oldCapacity + increment);
+      if (newCapacity < minCapacity) {
+	newCapacity = minCapacity;
+      }
+      matches = new Match[newCapacity];
+      System.arraycopy(oldMatches, 0, matches, 0, oldCapacity);
+      for (int i=oldCapacity; i<newCapacity; i++)
+	matches[i] = new Match();
+    }
+  }
+  /*
+  public final void addMatch(Match match) {
+    for (int i=0; i<anzahl; i++)
+      if (match.overlap(matches[i])) return;  // do not allow overlaps
+    ensureCapacity(anzahl + 1);
+    matches[anzahl++] = match;
+  }
+  */
+  public final void addMatch(int startA, int startB, int length) {
+    for (int i=anzahl-1; i>=0; i--) { // starting at the end is better(?)
+      if (matches[i].overlap(startA,startB,length)) return;
+                                      // no overlaps!
+    }
+    ensureCapacity(anzahl + 1);
+    //if (matches[anzahl] != null)  // object recycling...
+    matches[anzahl].set(startA,startB,length);
+    //else
+    //matches[anzahl ] = new Match(startA,startB,length);
+    anzahl++;
+  }
+
+  public final void clear() {
+    anzahl=0;
+  }
+}

jplag/src/test/resources/samples/PartialPlagiarism/C/Structure.java

@@ -0,0 +1,118 @@
+import java.io.*;
+
+/** The tokenlist */
+public class Structure implements TokenConstants {
+	public Token[] tokens = new Token[0];
+	Table table = null;
+	int hash_length = -1;
+
+	int files; // number of END_FILE tokens
+	private int anzahl;
+
+	//private final int increment = 300;
+
+	public Structure() {
+		tokens = new Token[400];
+		files = anzahl = 0;
+	}
+
+	public final int size() {
+		return anzahl;
+	}
+
+	public final void ensureCapacity(int minCapacity) {
+		int oldCapacity = tokens.length;
+		if (minCapacity > oldCapacity) {
+			Token[] oldTokens = tokens;
+			int newCapacity = (2 * oldCapacity);
+			if (newCapacity < minCapacity) {
+				newCapacity = minCapacity;
+			}
+			tokens = new Token[newCapacity];
+			System.arraycopy(oldTokens, 0, tokens, 0, anzahl);
+		}
+	}
+
+	public final void addToken(Token token) {
+		ensureCapacity(anzahl + 1);
+		if (anzahl > 0 && tokens[anzahl - 1].file.equals(token.file))
+			token.file = tokens[anzahl - 1].file; // To save memory ...
+		if ((anzahl > 0) && (token.getLine() < tokens[anzahl - 1].getLine()) && (token.file.equals(tokens[anzahl - 1].file)))
+			token.setLine(tokens[anzahl - 1].getLine());
+		// just to make sure
+
+		tokens[anzahl++] = token;
+		if (token.type == FILE_END)
+			files++;
+	}
+
+	public final String toString() {
+		StringBuffer buf = new StringBuffer();
+
+		try {
+			for (int i = 0; i < anzahl; i++) {
+				String s = tokens[i].toString();
+				buf.append(i);
+				buf.append("\t");
+				buf.append(s);
+				if (i < anzahl - 1) {
+					buf.append("\n");
+				}
+			}
+		} catch (OutOfMemoryError e) {
+			return "Tokenlist to large for output: " + (anzahl) + " Tokens";
+		}
+		return buf.toString();
+	}
+
+	public void save(File file) {
+
+		try {
+			ObjectOutputStream p = new ObjectOutputStream(/* new GZIPOutputStream */(new FileOutputStream(file)));
+
+			p.writeInt(anzahl);
+			p.writeInt(hash_length);
+			p.writeInt(files);
+
+			for (int i = 0; i < anzahl; i++)
+				p.writeObject(tokens[i]);
+			p.flush();
+			p.close();
+		} catch (IOException e) {
+			System.out.println("Error writing file: " + file.toString());
+		}
+	}
+
+	/* returns "true" when successful */
+	public boolean load(File file) {
+		try {
+			ObjectInputStream p = new ObjectInputStream(/* new GZIPInputStream */(new FileInputStream(file)));
+
+			int newAnzahl = p.readInt();
+			hash_length = p.readInt();
+			files = p.readInt();
+			ensureCapacity(newAnzahl);
+			anzahl = newAnzahl;
+			for (int i = 0; i < anzahl; i++) {
+				tokens[i] = (Token) p.readObject();
+				// special case for text tokens:
+				// 	if (tokens[i] instanceof jplag.text.TextToken) {
+				// 	  jplag.text.TextToken token = (jplag.text.TextToken)tokens[i];
+				// 	  jplag.text.TextToken.put(token.getText(), token.type);
+				// 	}
+			}
+			p.close();
+			table = null;
+		} catch (FileNotFoundException e) {
+			System.out.println("File not found: " + file.toString());
+			return false;
+		} catch (IOException e) {
+			System.out.println("Error reading file: " + file.toString() + " (" + e + ")");
+			return false;
+		} catch (ClassNotFoundException e) {
+			System.out.println("Class not found in file: " + file.toString());
+			return false;
+		}
+		return true;
+	}
+}

jplag/src/test/resources/samples/PartialPlagiarism/C/Submission.java

@@ -0,0 +1,218 @@
+import java.io.*;
+import java.net.URL;
+import java.text.DecimalFormat;
+import java.util.Vector;
+
+/*
+ * Everything about a single submission is stored in this object. (directory,
+ * files, ...)
+ */
+public class Submission implements Comparable<Submission> {
+	public String name;
+
+	private boolean readSubDirs;
+
+	public File dir;
+
+	public String[] files = null; // = new String[0];
+
+	public Structure struct;
+
+	public int structSize = 0;
+
+	// public long structMem;
+	boolean exact_match = false; // used for fallback
+
+	public boolean errors = false;
+
+	public DecimalFormat format = new DecimalFormat("0000");
+
+	public Submission(String name, File dir, boolean readSubDirs) {
+		this.dir = dir;
+		this.name = name;
+		this.readSubDirs = readSubDirs;
+		try {
+			lookupDir(dir, "");
+		} catch (Throwable b) {
+		}
+	}
+
+	public Submission(String name, File dir) {
+		this.dir = dir;
+		this.name = name;
+		this.readSubDirs = false;
+
+		files = new String[1];
+		files[0] = name;
+
+	}
+
+	// recursively read in all the files
+	private void lookupDir(File dir, String subDir) throws Throwable {
+		File aktDir = new File(dir, subDir);
+		if (!aktDir.isDirectory())
+			return;
+		if (readSubDirs) {
+			String[] dirs = aktDir.list();
+			if (!subDir.equals(""))
+				for (int i = 0; i < dirs.length; i++)
+					lookupDir(dir, subDir + File.separator + dirs[i]);
+			else
+				for (int i = 0; i < dirs.length; i++)
+					lookupDir(dir, dirs[i]);
+		}
+		String[] newFiles = aktDir.list(new FilenameFilter() {
+			public boolean accept(File dir, String name) {
+					return false;
+			}
+		});
+		if (files != null) {
+			String[] oldFiles = files;
+			files = new String[oldFiles.length + newFiles.length];
+			if (subDir != "")
+				for (int i = 0; i < newFiles.length; i++)
+					files[i] = subDir + File.separator + newFiles[i];
+			else
+				System.arraycopy(newFiles, 0, files, 0, newFiles.length);
+			System.arraycopy(oldFiles, 0, files, newFiles.length, oldFiles.length);
+		} else {
+			if (subDir != "") {
+				files = new String[newFiles.length];
+				for (int i = 0; i < newFiles.length; i++)
+					files[i] = subDir + File.separator + newFiles[i];
+			} else
+				files = newFiles;
+		}
+	}
+
+	/* parse all the files... */
+	public boolean parse(){
+		return false;
+	}
+
+	/*
+	 * This method is used to copy files that can not be parsed to a special
+	 * folder: jplag/errors/java old_java scheme cpp /001/(...files...)
+	 * /002/(...files...)
+	 */
+	private void copySubmission() {
+		File errorDir = null;
+		try {
+			URL url = Submission.class.getResource("errors");
+			errorDir = new File(url.getFile());
+		} catch (NullPointerException e) {
+			return;
+		}
+
+		if (!errorDir.exists())
+			errorDir.mkdir();
+		int i = 0;
+		File destDir;
+		while ((destDir = new File(errorDir, format.format(i))).exists())
+			i++;
+		destDir.mkdir();
+		for (i = 0; i < files.length; i++)
+			copyFile(new File(dir, files[i]), new File(destDir, files[i]));
+	}
+
+	/* Physical copy. :-) */
+	private void copyFile(File in, File out) {
+		byte[] buffer = new byte[10000];
+		try {
+			FileInputStream dis = new FileInputStream(in);
+			FileOutputStream dos = new FileOutputStream(out);
+			int count;
+			do {
+				count = dis.read(buffer);
+				if (count != -1)
+					dos.write(buffer, 0, count);
+			} while (count != -1);
+			dis.close();
+			dos.close();
+		} catch (IOException e) {
+
+		}
+	}
+
+	public int size() {
+		if (struct != null)
+			return structSize = struct.size();
+		return structSize;
+	}
+
+	/*
+	 * Used by the "Report" class. All source files are returned as an array of
+	 * an array of strings.
+	 */
+	public String[][] readFiles(String[] files) {
+		String[][] result = new String[files.length][];
+		String help;
+
+		Vector<String> text = new Vector<String>();
+		for (int i = 0; i < files.length; i++) {
+			text.removeAllElements();
+			try {
+				/* file encoding = "UTF-8" */
+				FileInputStream fileInputStream = new FileInputStream(new File(dir, files[i]));
+				InputStreamReader inputStreamReader = new InputStreamReader(fileInputStream, "UTF-8");
+				BufferedReader in = new BufferedReader(inputStreamReader);
+				while ((help = in.readLine()) != null) {
+					help = help.replaceAll("&", "&amp;");
+					help = help.replaceAll("<", "&lt;");
+					help = help.replaceAll(">", "&gt;");
+					help = help.replaceAll("\"", "&quot;");
+					text.addElement(help);
+				}
+				in.close();
+				inputStreamReader.close();
+				fileInputStream.close();
+			} catch (FileNotFoundException e) {
+				System.out.println("File not found: " + ((new File(dir, files[i])).toString()));
+			} catch (IOException e) {
+
+			}
+			result[i] = new String[text.size()];
+			text.copyInto(result[i]);
+		}
+		return result;
+	}
+
+	/*
+	 * Used by the "Report" class. All source files are returned as an array of
+	 * an array of chars.
+	 */
+	public char[][] readFilesChar(String[] files) {
+		char[][] result = new char[files.length][];
+
+		for (int i = 0; i < files.length; i++) {
+			try {
+				File file = new File(dir, files[i]);
+				int size = (int) file.length();
+				char[] buffer = new char[size];
+
+				FileReader fis = new FileReader(file);
+
+				if (size != fis.read(buffer)) {
+					System.out.println("Not right size read from the file, " + "but I will still continue...");
+				}
+
+				result[i] = buffer;
+				fis.close();
+			} catch (FileNotFoundException e) {
+				// TODO: Should an ExitException be thrown here?
+				System.out.println("File not found: " + ((new File(dir, files[i])).toString()));
+			} catch (IOException e) {
+
+			}
+		}
+		return result;
+	}
+
+	public int compareTo(Submission other) {
+		return name.compareTo(other.name);
+	}
+
+	public String toString() {
+		return name;
+	}
+}

jplag/src/test/resources/samples/PartialPlagiarism/C/Token.java

@@ -0,0 +1,47 @@
+import java.io.Serializable;
+
+abstract public class Token implements TokenConstants, Serializable {
+	private static final long serialVersionUID = 862935679966383302L;
+	public int type;
+	public String file;
+	
+	protected boolean marked;
+	protected boolean basecode = false;
+	protected int hash = -1;// hash-value. set and used by main algorithm (GSTiling)
+	
+	public Token(int type, String file, int line) {
+		this(type, file, line, -1, -1);
+	}
+	
+	public Token(int type, String file, int line, int column, int length) {
+		this.type = type;
+		this.file = file;
+		setLine(line>0 ? line : 1);
+		setColumn(column);
+		setLength(length);
+	}
+	
+	// abstract members
+	abstract public int getLine();
+	abstract public int getColumn();
+	abstract public int getLength();
+	protected void setLine(int line) {}
+	protected void setColumn(int line) {}
+	protected void setLength(int line) {}
+	
+	// this is made to distinguish the character front end.
+	// maybe other front ends can use it too?
+	protected int getIndex() { return -1; }
+	
+	public static String type2string(int type) {
+		return "<abstract>";
+	}
+	
+	public String toString() {
+		return type2string(type);
+	}
+	
+	public static int numberOfTokens() { 
+		return 1;
+	} 
+}

jplag/src/test/resources/samples/PartialPlagiarism/C/TokenConstants.java

@@ -0,0 +1,8 @@
+public interface TokenConstants {
+  final static int FILE_END = 0;
+	/**
+	 * Used to optionally separate methods from each other with an always marked
+	 * token
+	 */
+  final static int SEPARATOR_TOKEN = 1;
+}

jplag/src/test/resources/samples/PartialPlagiarism/D/Match.java

@@ -0,0 +1,54 @@
+package samples.PartialPlagiarism.D;
+
+public class Match {
+  public int startA;
+  public int startB;
+  public int length;
+
+  public Match(int startA, int startB, int length) {
+    this.startA = startA;
+    this.startB = startB;
+    this.length = length;
+  }
+  public Match() {}
+
+  public void set(int startA, int startB, int length) {
+    this.startA = startA;
+    this.startB = startB;
+    this.length = length;
+  }
+
+  public final boolean contains(int index, int sub) {
+    int start = (sub==0 ? startA : startB);
+    return (start <= index && index < (start+length));
+  }
+
+  public final boolean overlap(Match other) {
+    if (startA < other.startA) {
+      if ((other.startA - startA) < length) return true;
+    } else {
+      if ((startA - other.startA) < other.length) return true;
+    }
+    if (startB < other.startB) {
+      if ((other.startB - startB) < length) return true;
+    } else {
+      if ((startB - other.startB) < other.length) return true;
+    }
+    return false;
+  }
+
+  public final boolean overlap(int oStartA, int oStartB, int oLength) {
+    if (startA < oStartA) {
+      if ((oStartA - startA) < length) return true;
+    } else {
+      if ((startA - oStartA) < oLength) return true;
+    }
+    if (startB < oStartB) {
+      if ((oStartB - startB) < length) return true;
+    } else {
+      if ((startB - oStartB) < oLength) return true;
+    }
+    return false;
+  }
+}
+

jplag/src/test/resources/samples/PartialPlagiarism/D/Matches.java

@@ -0,0 +1,59 @@
+package samples.PartialPlagiarism.D;
+
+/* Minimal class that stores "Match" objects.
+ */
+public class Matches {
+  public Match[] matches;
+
+  private int anzahl;
+  private final int increment = 20;
+  
+  public Matches() {
+    matches = new Match[10];
+    for (int i=0; i<10; i++) matches[i] = new Match();
+    anzahl = 0;
+  }
+
+  public final int size() {
+    return anzahl;
+  }
+
+  public final void ensureCapacity(int minCapacity) {
+    int oldCapacity = matches.length;
+    if (minCapacity > oldCapacity) {
+      Match[] oldMatches = matches;
+      int newCapacity = (oldCapacity + increment);
+      if (newCapacity < minCapacity) {
+	newCapacity = minCapacity;
+      }
+      matches = new Match[newCapacity];
+      System.arraycopy(oldMatches, 0, matches, 0, oldCapacity);
+      for (int i=oldCapacity; i<newCapacity; i++)
+	matches[i] = new Match();
+    }
+  }
+  /*
+  public final void addMatch(Match match) {
+    for (int i=0; i<anzahl; i++)
+      if (match.overlap(matches[i])) return;  // do not allow overlaps
+    ensureCapacity(anzahl + 1);
+    matches[anzahl++] = match;
+  }
+  */
+  public final void addMatch(int startA, int startB, int length) {
+    for (int i=anzahl-1; i>=0; i--) { // starting at the end is better(?)
+      if (matches[i].overlap(startA,startB,length)) return;
+                                      // no overlaps!
+    }
+    ensureCapacity(anzahl + 1);
+    //if (matches[anzahl] != null)  // object recycling...
+    matches[anzahl].set(startA,startB,length);
+    //else
+    //matches[anzahl ] = new Match(startA,startB,length);
+    anzahl++;
+  }
+
+  public final void clear() {
+    anzahl=0;
+  }
+}

jplag/src/test/resources/samples/PartialPlagiarism/D/Structure.java

@@ -0,0 +1,120 @@
+package samples.PartialPlagiarism.D;
+
+import java.io.*;
+
+/** The tokenlist */
+public class Structure implements TokenConstants {
+	public Token[] tokens = new Token[0];
+	Table table = null;
+	int hash_length = -1;
+
+	int files; // number of END_FILE tokens
+	private int anzahl;
+
+	//private final int increment = 300;
+
+	public Structure() {
+		tokens = new Token[400];
+		files = anzahl = 0;
+	}
+
+	public final int size() {
+		return anzahl;
+	}
+
+	public final void ensureCapacity(int minCapacity) {
+		int oldCapacity = tokens.length;
+		if (minCapacity > oldCapacity) {
+			Token[] oldTokens = tokens;
+			int newCapacity = (2 * oldCapacity);
+			if (newCapacity < minCapacity) {
+				newCapacity = minCapacity;
+			}
+			tokens = new Token[newCapacity];
+			System.arraycopy(oldTokens, 0, tokens, 0, anzahl);
+		}
+	}
+
+	public final void addToken(Token token) {
+		ensureCapacity(anzahl + 1);
+		if (anzahl > 0 && tokens[anzahl - 1].file.equals(token.file))
+			token.file = tokens[anzahl - 1].file; // To save memory ...
+		if ((anzahl > 0) && (token.getLine() < tokens[anzahl - 1].getLine()) && (token.file.equals(tokens[anzahl - 1].file)))
+			token.setLine(tokens[anzahl - 1].getLine());
+		// just to make sure
+
+		tokens[anzahl++] = token;
+		if (token.type == FILE_END)
+			files++;
+	}
+
+	public final String toString() {
+		StringBuffer buf = new StringBuffer();
+
+		try {
+			for (int i = 0; i < anzahl; i++) {
+				String s = tokens[i].toString();
+				buf.append(i);
+				buf.append("\t");
+				buf.append(s);
+				if (i < anzahl - 1) {
+					buf.append("\n");
+				}
+			}
+		} catch (OutOfMemoryError e) {
+			return "Tokenlist to large for output: " + (anzahl) + " Tokens";
+		}
+		return buf.toString();
+	}
+
+	public void save(File file) {
+
+		try {
+			ObjectOutputStream p = new ObjectOutputStream(/* new GZIPOutputStream */(new FileOutputStream(file)));
+
+			p.writeInt(anzahl);
+			p.writeInt(hash_length);
+			p.writeInt(files);
+
+			for (int i = 0; i < anzahl; i++)
+				p.writeObject(tokens[i]);
+			p.flush();
+			p.close();
+		} catch (IOException e) {
+			System.out.println("Error writing file: " + file.toString());
+		}
+	}
+
+	/* returns "true" when successful */
+	public boolean load(File file) {
+		try {
+			ObjectInputStream p = new ObjectInputStream(/* new GZIPInputStream */(new FileInputStream(file)));
+
+			int newAnzahl = p.readInt();
+			hash_length = p.readInt();
+			files = p.readInt();
+			ensureCapacity(newAnzahl);
+			anzahl = newAnzahl;
+			for (int i = 0; i < anzahl; i++) {
+				tokens[i] = (Token) p.readObject();
+				// special case for text tokens:
+				// 	if (tokens[i] instanceof jplag.text.TextToken) {
+				// 	  jplag.text.TextToken token = (jplag.text.TextToken)tokens[i];
+				// 	  jplag.text.TextToken.put(token.getText(), token.type);
+				// 	}
+			}
+			p.close();
+			table = null;
+		} catch (FileNotFoundException e) {
+			System.out.println("File not found: " + file.toString());
+			return false;
+		} catch (IOException e) {
+			System.out.println("Error reading file: " + file.toString() + " (" + e + ")");
+			return false;
+		} catch (ClassNotFoundException e) {
+			System.out.println("Class not found in file: " + file.toString());
+			return false;
+		}
+		return true;
+	}
+}

jplag/src/test/resources/samples/PartialPlagiarism/D/Submission.java

@@ -0,0 +1,206 @@
+package samples.PartialPlagiarism.D;
+
+import java.io.*;
+import java.net.URL;
+import java.text.DecimalFormat;
+import java.util.Vector;
+
+/*
+ * Everything about a single submission is stored in this object. (directory,
+ * files, ...)
+ */
+public class Submission implements Comparable<Submission> {
+	public String name;
+
+	private boolean readSubDirs;
+
+	public File dir;
+
+	public String[] files = null; // = new String[0];
+
+	public Structure structure;
+
+	public int structSize = 0;
+
+	// public long structMem;
+	boolean exact_match = false; // used for fallback
+
+	public boolean errors = false;
+
+	public DecimalFormat format = new DecimalFormat("0000");
+
+	public Submission(String name, File dir, boolean readSubDirs) {
+		this.dir = dir;
+		this.name = name;
+		this.readSubDirs = readSubDirs;
+		try {
+			lookupDir(dir, "");
+		} catch (Throwable b) {
+		}
+	}
+
+	public Submission(String name, File dir) {
+		this.dir = dir;
+		this.name = name;
+		this.readSubDirs = false;
+
+		files = new String[1];
+		files[0] = name;
+
+	}
+
+	// recursively read in all the files
+	private void lookupDir(File dir, String subDir) throws Throwable {
+		File aktDir = new File(dir, subDir);
+		if (!aktDir.isDirectory())
+			return;
+		if (readSubDirs) {
+			String[] dirs = aktDir.list();
+			if (!subDir.equals(""))
+				for (int i = 0; i < dirs.length; i++)
+					lookupDir(dir, subDir + File.separator + dirs[i]);
+			else
+				for (int i = 0; i < dirs.length; i++)
+					lookupDir(dir, dirs[i]);
+		}
+		String[] newFiles = aktDir.list(new FilenameFilter() {
+			public boolean accept(File dir, String name) {
+					return false;
+			}
+		});
+		if (files != null) {
+			String[] oldFiles = files;
+			files = new String[oldFiles.length + newFiles.length];
+			if (subDir != "")
+				for (int i = 0; i < newFiles.length; i++)
+					files[i] = subDir + File.separator + newFiles[i];
+			else
+				System.arraycopy(newFiles, 0, files, 0, newFiles.length);
+			System.arraycopy(oldFiles, 0, files, newFiles.length, oldFiles.length);
+		} else {
+			if (subDir != "") {
+				files = new String[newFiles.length];
+				for (int i = 0; i < newFiles.length; i++)
+					files[i] = subDir + File.separator + newFiles[i];
+			} else
+				files = newFiles;
+		}
+	}
+
+	/* parse all the files... */
+	public boolean parse(){
+		return false;
+	}
+
+	/*
+	 * This method is used to copy files that can not be parsed to a special
+	 * folder: jplag/errors/java old_java scheme cpp /001/(...files...)
+	 * /002/(...files...)
+	 */
+	private void doSmth() {
+		File failureDir = null;
+		try {
+			URL url = Submission.class.getResource("errors");
+			failureDir = new File(url.getFile());
+		} catch (NullPointerException e) {
+			return;
+		}
+
+		if (!failureDir.exists())
+			failureDir.mkdir();
+		int i = 0;
+	}
+
+	private void copyFile(File in, File out) {
+		byte[] buffer = new byte[10000];
+		try {
+			FileInputStream des = new FileInputStream(in);
+			FileOutputStream foo = new FileOutputStream(out);
+			int count;
+			do {
+				count = des.read(buffer);
+				if (count != -1)
+					foo.write(buffer, 0, count);
+			} while (count != -1);
+			des.close();
+			foo.close();
+		} catch (IOException e) {
+			System.out.println("Oh no!");
+		}
+	}
+
+	public int size() {
+		if (structure != null)
+			return structSize = structure.size();
+		return structSize;
+	}
+
+	/*
+	 * Used by the "Report" class. All source files are returned as an array of
+	 * an array of strings.
+	 */
+	public String[][] readFiles(String[] files) {
+		String[][] newName = new String[files.length][];
+		String help;
+
+		Vector<String> text = new Vector<String>();
+		for (int i = 0; i < files.length; i++) {
+			text.removeAllElements();
+			try {
+				/* file encoding = "UTF-8" */
+				FileInputStream fileInputStream = new FileInputStream(new File(dir, files[i]));
+				InputStreamReader inputStreamReader = new InputStreamReader(fileInputStream, "UTF-8");
+				BufferedReader in = new BufferedReader(inputStreamReader);
+				while ((help = in.readLine()) != null) {
+					help = help.replaceAll("&", "&amp;");
+					text.addElement(help);
+				}
+				in.close();
+				inputStreamReader.close();
+				fileInputStream.close();
+			} catch (FileNotFoundException e) {
+				System.out.println("File not found: " + ((new File(dir, files[i])).toString()));
+			} catch (IOException e) {
+				System.out.println("Oh no!");
+			}
+			newName[i] = new String[text.size()];
+			text.copyInto(newName[i]);
+		}
+		return newName;
+	}
+
+	/*
+	 * Used by the "Report" class. All source files are returned as an array of
+	 * an array of chars.
+	 */
+	public char[][] readFilesChar(String[] files) {
+		char[][] variabl = new char[files.length][];
+
+		for (int i = 0; i < files.length; i++) {
+			try {
+				File file = new File(dir, files[i]);
+				int size = (int) file.length();
+				char[] buffer = new char[size];
+
+				FileReader fis = new FileReader(file);
+
+				if (size != fis.read(buffer)) {
+					System.out.println("Not right size read from the file, " + "but I will still continue...");
+				}
+
+				variabl[i] = buffer;
+				fis.close();
+			} catch (FileNotFoundException e) {
+				System.out.println("Oh no");
+			} catch (IOException e) {
+
+			}
+		}
+		return variabl;
+	}
+
+	@Override
+	public int compareTo(Submission o) {
+		return 0;
+	}
+}

jplag/src/test/resources/samples/PartialPlagiarism/D/Token.java

@@ -0,0 +1,44 @@
+package samples.PartialPlagiarism.D;
+
+import java.io.Serializable;
+
+abstract public class Token implements TokenConstants, Serializable {
+	private static final long serialVersionUID = 862335679966383302L;
+	public int type;
+	public String file;
+	
+	protected boolean marked;
+	protected boolean basecode = false;
+	protected int hash = -1;// hash-value. set and used by main algorithm (GSTiling)
+	
+	public Token(int type, String file, int line, int column, int length) {
+		this.type = type;
+		this.file = file;
+		setLine(line>0 ? line : 1);
+		setColumn(column);
+		setLength(length);
+	}
+	
+	// abstract members
+	abstract public int getLine();
+	abstract public int getColumn();
+	abstract public int getLength();
+	protected void setLine(int line) {}
+	protected void setColumn(int line) {}
+	protected void setLength(int line) {}
+	
+	// This is original work not copied
+	protected int getIndex() { return -1; }
+	
+	public static String type2string(int type) {
+		return "<abstract>";
+	}
+	
+	public String toString() {
+		return type2string(type);
+	}
+	
+	public static int numberOfTokens() { 
+		return 1;
+	} 
+}

jplag/src/test/resources/samples/PartialPlagiarism/D/TokenConstants.java

@@ -0,0 +1,12 @@
+package samples.PartialPlagiarism.D;
+
+public interface TokenConstants {
+  final static int FILE_END = 0;
+	/**
+	 * Used to optionally separate methods from each other with an always marked
+	 * token
+	 */
+  final static int SEPARATOR_TOKEN = 1;
+
+  final static int ANOTHER_TOKEN = 42;
+}

jplag/src/test/resources/samples/PartialPlagiarism/E/WhatAmIDoingHere.java

@@ -0,0 +1,5 @@
+public class WhatAmIDoingHere {
+    public static void main(String[] args) {
+        System.out.println("Hello JPlag!");
+    }
+}