[#944] Add originality threshold flag (#2122)

The existing setup employs a static originality threshold of 0.51.
However, this threshold is tailored for codes, such as Java or Markdown,
and might not be suitable for other programming languages. Additionally,
it doesn't offer flexibility for users who may want a stricter threshold
but are willing to endure longer processing times, or those who prefer a
more lenient threshold but prioritize faster analysis speeds.

Let's enable users to input their preferred originality threshold.

src/main/java/reposense/RepoSense.java

@@ -80,7 +80,7 @@ public static void main(String[] args) {
  cliArguments.isSinceDateProvided(), cliArguments.isUntilDateProvided(),
  cliArguments.getNumCloningThreads(), cliArguments.getNumAnalysisThreads(),
  TimeUtil::getElapsedTime, cliArguments.getZoneId(), cliArguments.isFreshClonePerformed(),
- cliArguments.isAuthorshipAnalyzed());
+ cliArguments.isAuthorshipAnalyzed(), cliArguments.getOriginalityThreshold());
 
  FileUtil.zipFoldersAndFiles(reportFoldersAndFiles, cliArguments.getOutputFilePath().toAbsolutePath(),
  ".json");

src/main/java/reposense/authorship/AuthorshipReporter.java

@@ -30,9 +30,11 @@ public class AuthorshipReporter {
 
  /**
  * Generates and returns the authorship summary for each repo in {@code config}.
- * Further analyzes the authorship of each line in the commit if {@code shouldAnalyzeAuthorship} is true.
+ * Further analyzes the authorship of each line in the commit if {@code shouldAnalyzeAuthorship} is true, based on
+ * {code originalityThreshold}.
  */
- public AuthorshipSummary generateAuthorshipSummary(RepoConfiguration config, boolean shouldAnalyzeAuthorship) {
+ public AuthorshipSummary generateAuthorshipSummary(RepoConfiguration config, boolean shouldAnalyzeAuthorship,
+ double originalityThreshold) {
  List<FileInfo> textFileInfos = fileInfoExtractor.extractTextFileInfos(config);
 
  int numFiles = textFileInfos.size();
@@ -45,7 +47,8 @@ public AuthorshipSummary generateAuthorshipSummary(RepoConfiguration config, boo
  }
 
  List<FileResult> fileResults = textFileInfos.stream()
- .map(fileInfo -> fileInfoAnalyzer.analyzeTextFile(config, fileInfo, shouldAnalyzeAuthorship))
+ .map(fileInfo -> fileInfoAnalyzer.analyzeTextFile(config, fileInfo, shouldAnalyzeAuthorship,
+ originalityThreshold))
  .filter(Objects::nonNull)
  .collect(Collectors.toList());
 

src/main/java/reposense/authorship/FileInfoAnalyzer.java

@@ -1,5 +1,7 @@
 package reposense.authorship;
 
+import static reposense.parser.ArgsParser.DEFAULT_ORIGINALITY_THRESHOLD;
+
 import java.nio.file.Files;
 import java.nio.file.Path;
 import java.nio.file.Paths;
@@ -45,11 +47,13 @@ public class FileInfoAnalyzer {
  /**
  * Analyzes the lines of the file, given in the {@code fileInfo}, that has changed in the time period provided
  * by {@code config}.
- * Further analyzes the authorship of each line in the commit if {@code shouldAnalyzeAuthorship} is true.
+ * Further analyzes the authorship of each line in the commit if {@code shouldAnalyzeAuthorship} is true, based on
+ * {@code originalityThreshold}.
  * Returns null if the file is missing from the local system, or none of the
  * {@link Author} specified in {@code config} contributed to the file in {@code fileInfo}.
  */
- public FileResult analyzeTextFile(RepoConfiguration config, FileInfo fileInfo, boolean shouldAnalyzeAuthorship) {
+ public FileResult analyzeTextFile(RepoConfiguration config, FileInfo fileInfo, boolean shouldAnalyzeAuthorship,
+ double originalityThreshold) {
  String relativePath = fileInfo.getPath();
 
  if (Files.notExists(Paths.get(config.getRepoRoot(), relativePath))) {
@@ -61,7 +65,7 @@ public FileResult analyzeTextFile(RepoConfiguration config, FileInfo fileInfo, b
  return null;
  }
 
- aggregateBlameAuthorModifiedAndDateInfo(config, fileInfo, shouldAnalyzeAuthorship);
+ aggregateBlameAuthorModifiedAndDateInfo(config, fileInfo, shouldAnalyzeAuthorship, originalityThreshold);
  fileInfo.setFileType(config.getFileType(fileInfo.getPath()));
 
  AnnotatorAnalyzer.aggregateAnnotationAuthorInfo(fileInfo, config.getAuthorConfig(), shouldAnalyzeAuthorship);
@@ -83,7 +87,7 @@ public FileResult analyzeTextFile(RepoConfiguration config, FileInfo fileInfo, b
  * {@link Author} specified in {@code config} contributed to the file in {@code fileInfo}.
  */
  public FileResult analyzeTextFile(RepoConfiguration config, FileInfo fileInfo) {
- return analyzeTextFile(config, fileInfo, false);
+ return analyzeTextFile(config, fileInfo, false, DEFAULT_ORIGINALITY_THRESHOLD);
  }
 
  /**
@@ -153,10 +157,11 @@ private FileResult generateBinaryFileResult(RepoConfiguration config, FileInfo f
  * The {@code config} is used to obtain the root directory for running git blame as well as other parameters used
  * in determining which author to assign to each line and whether to set the last modified date for a
  * {@code lineInfo}.
- * Further analyzes the authorship of each line in the commit if {@code shouldAnalyzeAuthorship} is true.
+ * Further analyzes the authorship of each line in the commit if {@code shouldAnalyzeAuthorship} is true, based on
+ * {@code originalityThreshold}.
  */
  private void aggregateBlameAuthorModifiedAndDateInfo(RepoConfiguration config, FileInfo fileInfo,
- boolean shouldAnalyzeAuthorship) {
+ boolean shouldAnalyzeAuthorship, double originalityThreshold) {
  String blameResults;
 
  if (!config.isFindingPreviousAuthorsPerformed()) {
@@ -199,7 +204,7 @@ private void aggregateBlameAuthorModifiedAndDateInfo(RepoConfiguration config, F
  if (shouldAnalyzeAuthorship && !author.equals(Author.UNKNOWN_AUTHOR)) {
  String lineContent = fileInfo.getLine(lineCount / 5 + 1).getContent();
  boolean isFullCredit = AuthorshipAnalyzer.analyzeAuthorship(config, fileInfo.getPath(), lineContent,
- commitHash, author);
+ commitHash, author, originalityThreshold);
  fileInfo.setIsFullCredit(lineCount / 5, isFullCredit);
  }
  }

src/main/java/reposense/authorship/analyzer/AuthorshipAnalyzer.java

@@ -22,9 +22,6 @@
  */
 public class AuthorshipAnalyzer {
  private static final Logger logger = LogsManager.getLogger(AuthorshipAnalyzer.class);
-
- private static final double ORIGINALITY_THRESHOLD = 0.51;
-
  private static final String DIFF_FILE_CHUNK_SEPARATOR = "\ndiff --git a/.*\n";
  private static final Pattern FILE_CHANGED_PATTERN =
  Pattern.compile("\n(-){3} a?/(?<preImageFilePath>.*)\n(\\+){3} b?/(?<postImageFilePath>.*)\n");
@@ -45,11 +42,11 @@ public class AuthorshipAnalyzer {
  private static final String DELETED_LINE_SYMBOL = "-";
 
  /**
- * Analyzes the authorship of {@code lineContent} in {@code filePath}.
+ * Analyzes the authorship of {@code lineContent} in {@code filePath} based on {@code originalityThreshold}.
  * Returns {@code true} if {@code currentAuthor} should be assigned full credit, {@code false} otherwise.
  */
  public static boolean analyzeAuthorship(RepoConfiguration config, String filePath, String lineContent,
- String commitHash, Author currentAuthor) {
+ String commitHash, Author currentAuthor, double originalityThreshold) {
  // Empty lines are ignored and given full credit
  if (lineContent.isEmpty()) {
  return true;
@@ -58,7 +55,7 @@ public static boolean analyzeAuthorship(RepoConfiguration config, String filePat
  CandidateLine deletedLine = getDeletedLineWithLowestOriginality(config, filePath, lineContent, commitHash);
 
  // Give full credit if there are no deleted lines found or deleted line is more than originality threshold
- if (deletedLine == null || deletedLine.getOriginalityScore() > ORIGINALITY_THRESHOLD) {
+ if (deletedLine == null || deletedLine.getOriginalityScore() > originalityThreshold) {
  return true;
  }
 
@@ -80,7 +77,7 @@ public static boolean analyzeAuthorship(RepoConfiguration config, String filePat
 
  // Check the previous version as currentAuthor is the same as author of the previous version
  return analyzeAuthorship(config, deletedLine.getFilePath(), deletedLine.getLineContent(),
- deletedLineInfo.getCommitHash(), deletedLineInfo.getAuthor());
+ deletedLineInfo.getCommitHash(), deletedLineInfo.getAuthor(), originalityThreshold);
  }
 
  /**

src/main/java/reposense/model/CliArguments.java

@@ -36,6 +36,7 @@ public class CliArguments {
  private final ZoneId zoneId;
  private final boolean isFindingPreviousAuthorsPerformed;
  private final boolean isAuthorshipAnalyzed;
+ private final double originalityThreshold;
 
  private boolean isTestMode = ArgsParser.DEFAULT_IS_TEST_MODE;
  private boolean isFreshClonePerformed = ArgsParser.DEFAULT_SHOULD_FRESH_CLONE;
@@ -81,6 +82,7 @@ private CliArguments(Builder builder) {
  this.reportConfigFilePath = builder.reportConfigFilePath;
  this.reportConfiguration = builder.reportConfiguration;
  this.isAuthorshipAnalyzed = builder.isAuthorshipAnalyzed;
+ this.originalityThreshold = builder.originalityThreshold;
  }
 
  public ZoneId getZoneId() {
@@ -195,6 +197,10 @@ public boolean isAuthorshipAnalyzed() {
  return isAuthorshipAnalyzed;
  }
 
+ public double getOriginalityThreshold() {
+ return originalityThreshold;
+ }
+
  @Override
  public boolean equals(Object other) {
  // short circuit if same object
@@ -233,7 +239,8 @@ public boolean equals(Object other) {
  && Objects.equals(this.authorConfigFilePath, otherCliArguments.authorConfigFilePath)
  && Objects.equals(this.groupConfigFilePath, otherCliArguments.groupConfigFilePath)
  && Objects.equals(this.reportConfigFilePath, otherCliArguments.reportConfigFilePath)
- && this.isAuthorshipAnalyzed == otherCliArguments.isAuthorshipAnalyzed;
+ && this.isAuthorshipAnalyzed == otherCliArguments.isAuthorshipAnalyzed
+ && Objects.equals(this.originalityThreshold, otherCliArguments.originalityThreshold);
  }
 
  /**
@@ -268,6 +275,7 @@ public static final class Builder {
  private Path reportConfigFilePath;
  private ReportConfiguration reportConfiguration;
  private boolean isAuthorshipAnalyzed;
+ private double originalityThreshold;
 
  public Builder() {
  }
@@ -520,6 +528,16 @@ public Builder isAuthorshipAnalyzed(boolean isAuthorshipAnalyzed) {
  return this;
  }
 
+ /**
+ * Adds the {@code originalityThreshold} to CliArguments.
+ *
+ * @param originalityThreshold the originality threshold.
+ */
+ public Builder originalityThreshold(double originalityThreshold) {
+ this.originalityThreshold = originalityThreshold;
+ return this;
+ }
+
  /**
  * Builds CliArguments.
  *

src/main/java/reposense/parser/ArgsParser.java

@@ -40,6 +40,7 @@ public class ArgsParser {
  public static final int DEFAULT_NUM_ANALYSIS_THREADS = Runtime.getRuntime().availableProcessors();
  public static final boolean DEFAULT_IS_TEST_MODE = false;
  public static final boolean DEFAULT_SHOULD_FRESH_CLONE = false;
+ public static final double DEFAULT_ORIGINALITY_THRESHOLD = 0.51;
 
  public static final String[] HELP_FLAGS = new String[] {"--help", "-h"};
  public static final String[] CONFIG_FLAGS = new String[] {"--config", "-c"};
@@ -63,6 +64,7 @@ public class ArgsParser {
  public static final String[] TEST_MODE_FLAG = new String[] {"--test-mode"};
  public static final String[] FRESH_CLONING_FLAG = new String[] {"--fresh-cloning"};
  public static final String[] ANALYZE_AUTHORSHIP_FLAGS = new String[] {"--analyze-authorship", "-A"};
+ public static final String[] ORIGINALITY_THRESHOLD_FLAGS = new String[] {"--originality-threshold", "-ot"};
 
  private static final Logger logger = LogsManager.getLogger(ArgsParser.class);
 
@@ -201,6 +203,13 @@ private static ArgumentParser getArgumentParser() {
  .action(Arguments.storeTrue())
  .help("A flag to perform analysis of code authorship.");
 
+ parser.addArgument(ORIGINALITY_THRESHOLD_FLAGS)
+ .dest(ORIGINALITY_THRESHOLD_FLAGS[0])
+ .metavar("(0.0 ~ 1.0)")
+ .type(new OriginalityThresholdArgumentType())
+ .setDefault(DEFAULT_ORIGINALITY_THRESHOLD)
+ .help("The originality threshold for analysis of code authorship.");
+
  // Mutex flags - these will always be the last parameters in help message.
  mutexParser.addArgument(CONFIG_FLAGS)
  .dest(CONFIG_FLAGS[0])
@@ -280,6 +289,7 @@ public static CliArguments parse(String[] args) throws HelpScreenException, Pars
  boolean shouldFindPreviousAuthors = results.get(FIND_PREVIOUS_AUTHORS_FLAGS[0]);
  boolean isTestMode = results.get(TEST_MODE_FLAG[0]);
  boolean isAuthorshipAnalyzed = results.get(ANALYZE_AUTHORSHIP_FLAGS[0]);
+ double originalityThreshold = results.get(ORIGINALITY_THRESHOLD_FLAGS[0]);
  int numCloningThreads = results.get(CLONING_THREADS_FLAG[0]);
  int numAnalysisThreads = results.get(ANALYSIS_THREADS_FLAG[0]);
 
@@ -299,7 +309,8 @@ public static CliArguments parse(String[] args) throws HelpScreenException, Pars
  .numCloningThreads(numCloningThreads)
  .numAnalysisThreads(numAnalysisThreads)
  .isTestMode(isTestMode)
- .isAuthorshipAnalyzed(isAuthorshipAnalyzed);
+ .isAuthorshipAnalyzed(isAuthorshipAnalyzed)
+ .originalityThreshold(originalityThreshold);
 
  LogsManager.setLogFolderLocation(outputFolderPath);
 

src/main/java/reposense/parser/OriginalityThresholdArgumentType.java

@@ -0,0 +1,25 @@
+package reposense.parser;
+
+import net.sourceforge.argparse4j.inf.Argument;
+import net.sourceforge.argparse4j.inf.ArgumentParser;
+import net.sourceforge.argparse4j.inf.ArgumentParserException;
+import net.sourceforge.argparse4j.inf.ArgumentType;
+
+/**
+ * Verifies and parses a string-formatted double, between 0.0 and 1.0, to an {@link Double} object.
+ */
+public class OriginalityThresholdArgumentType implements ArgumentType<Double> {
+ private static final String PARSE_EXCEPTION_MESSAGE_THRESHOLD_OUT_OF_BOUND =
+ "Invalid threshold. It must be a number between 0.0 and 1.0.";
+
+ @Override
+ public Double convert(ArgumentParser parser, Argument arg, String value) throws ArgumentParserException {
+ double threshold = Double.parseDouble(value);
+
+ if (Double.compare(threshold, 0.0) < 0 || Double.compare(threshold, 1.0) > 0) {
+ throw new ArgumentParserException(PARSE_EXCEPTION_MESSAGE_THRESHOLD_OUT_OF_BOUND, parser);
+ }
+
+ return threshold;
+ }
+}