java类org.antlr.v4.runtime.misc.Triple的实例源码

public static List<Float> getWSErrorRates(LangDescriptor language,
                                              FeatureMetaData[] injectWSFeatures,
                                              FeatureMetaData[] alignmentFeatures)
        throws Exception
    {
        LeaveOneOutValidator validator = new LeaveOneOutValidator(language.corpusDir, language);
        Triple<List<Formatter>,List<Float>,List<Float>> results =
            validator.validateDocuments(injectWSFeatures, alignmentFeatures, false, null);
        List<Formatter> formatters = results.a;
        List<Float> wsErrorRates = new ArrayList<>(); // don't include align errors
        for (Formatter formatter : formatters) {
            ClassificationAnalysis analysis =
                new ClassificationAnalysis(formatter.testDoc, formatter.getAnalysisPerToken());
            wsErrorRates.add(analysis.getWSErrorRate());
        }
//      System.out.println(results.c);
//      System.out.println("vs");
//      System.out.println(wsErrorRates);
        return wsErrorRates;
    }

public static List<Float> getAlignmentErrorRates(LangDescriptor language,
                                                     FeatureMetaData[] injectWSFeatures,
                                                     FeatureMetaData[] alignmentFeatures)
        throws Exception
    {
        LeaveOneOutValidator validator = new LeaveOneOutValidator(language.corpusDir, language);
        Triple<List<Formatter>,List<Float>,List<Float>> results =
            validator.validateDocuments(injectWSFeatures, alignmentFeatures, false, null);
        List<Formatter> formatters = results.a;
        List<Float> alignErrorRates = new ArrayList<>(); // don't include align errors
        for (Formatter formatter : formatters) {
            ClassificationAnalysis analysis =
                new ClassificationAnalysis(formatter.testDoc, formatter.getAnalysisPerToken());
            alignErrorRates.add(analysis.getAlignmentErrorRate());
        }
//      System.out.println(results.c);
//      System.out.println("vs");
//      System.out.println(alignErrorRates);
        return alignErrorRates;
    }

protected Triple<Integer, String, String> exec(String[] cmd, String workingDir) throws IOException, InterruptedException {
    ProcessBuilder pb = new ProcessBuilder();
    pb.command(Arrays.asList(cmd)).directory(new File(workingDir));
    Process process = pb.start();
    int resultCode = process.waitFor();
    String stdout = dump(process.getInputStream());
    String stderr = dump(process.getErrorStream());
    return new Triple<>(resultCode, stdout, stderr);
}

public void checkCExec(String filename) throws Exception {
    URL testFolderURL = TestCGen.class.getClassLoader().getResource(SAMPLES_DIR);
    String testFolder = testFolderURL.getPath();
    String workingDir = getWorkingDir();

    String J_pathToFile = testFolder+"/"+filename;
    String C_filename = basename(filename)+".c";

    JTran jTran = new JTran();
    String C_code = jTran.translate(J_pathToFile, C_filename, false, false);

    Utils.writeFile(workingDir+"/"+C_filename, C_code);

    // compile
    String[] cc = {"cc", "-o", basename(filename), C_filename};
    Triple<Integer, String, String> cc_result = exec(cc, getWorkingDir());
    int execCode = cc_result.a;
    String stdout = cc_result.b;
    String stderr = cc_result.c;

    assertEquals("", stdout);
    assertEquals("", stderr);
    assertEquals(0, execCode);

    // execute
    String[] exec_cmd = {"./"+basename(filename)};
    Triple<Integer, String, String> result = exec(exec_cmd, getWorkingDir());
    execCode = result.a;
    stdout = result.b;
    stderr = result.c;

    String expected_output_filename = basename(filename)+".txt";
    String expected_output = readFile(testFolder+"/"+expected_output_filename);

    assertEquals(expected_output, stdout);
    assertEquals("", stderr);
    assertEquals(0, execCode);
}

/** Return error rate for each document using leave-one-out validation */
public List<Float> scoreDocuments() throws Exception {
    List<String> allFiles = getFilenames(new File(rootDir), language.fileRegex);
    List<InputDocument> documents = Tool.load(allFiles, language);
    List<Float> errors = new ArrayList<>();
    for (int i = 0; i<documents.size(); i++) {
        Triple<Formatter,Float,Float> results =
            validate(language, documents, documents.get(i).fileName, k, null, false, false);
        Float errorRate = results.c;
        errors.add(errorRate);
    }
    return errors;
}

public static List<Float> checkStability(LangDescriptor language) throws Exception {
    List<Float> errorRates = new ArrayList<>();

    // format the corpus into tmp dir
    LeaveOneOutValidator validator0 = new LeaveOneOutValidator(language.corpusDir, language);
    Triple<List<Formatter>, List<Float>, List<Float>> results0 = validator0.validateDocuments(false, "/tmp/stability/1");
    errorRates.add( BuffUtils.median(results0.c) );

    List<Formatter> formatters0 = results0.a;
    // now try formatting it over and over
    for (int i = 1; i<=STAGES; i++) {
        String inputDir  = "/tmp/stability/"+i;
        String outputDir = "/tmp/stability/"+(i+1);
        LeaveOneOutValidator validator = new LeaveOneOutValidator(inputDir, language);
        Triple<List<Formatter>, List<Float>, List<Float>> results =
            validator.validateDocuments(false, outputDir);
        List<Formatter> formatters = results.a;
        List<Float> distances = new ArrayList<>();
        for (int j = 0; j<formatters.size(); j++) {
            Formatter f0 = formatters0.get(j);
            Formatter f = formatters.get(j);
            float editDistance = normalizedLevenshteinDistance(f.getOutput(), f0.getOutput());
            distances.add(editDistance);
        }
        errorRates.add( BuffUtils.median(distances) );
    }

    return errorRates;
}

public static Triple<Formatter,Float,Float> validate(LangDescriptor language,
                                                         List<InputDocument> documents,
                                                         InputDocument testDoc,
                                                         boolean saveOutput,
                                                         boolean computeEditDistance)
        throws Exception
    {
//      kNNClassifier.resetCache();
        Corpus corpus = new Corpus(documents, language);
        corpus.train();
//      System.out.printf("%d feature vectors\n", corpus.featureVectors.size());
        Formatter formatter = new Formatter(corpus, language.indentSize);
        String output = formatter.format(testDoc, false);
        float editDistance = 0;
        if ( computeEditDistance ) {
            editDistance = normalizedLevenshteinDistance(testDoc.content, output);
        }
        ClassificationAnalysis analysis = new ClassificationAnalysis(testDoc, formatter.getAnalysisPerToken());
//      System.out.println(testDoc.fileName+": edit distance = "+editDistance+", error rate = "+analysis.getErrorRate());
        if ( saveOutput ) {
            File dir = new File(outputDir+"/"+language.name);
            if ( saveOutput ) {
                dir = new File(outputDir+"/"+language.name);
                dir.mkdir();
            }
            Utils.writeFile(dir.getPath()+"/"+new File(testDoc.fileName).getName(), output);
        }
        return new Triple<>(formatter, editDistance, analysis.getErrorRate());
    }

public Triple<Formatter,Float,Float> validateOneDocument(String fileToExclude,
                                                         String outputDir,
                                                         boolean collectAnalysis)
    throws Exception
{
    List<String> allFiles = getFilenames(new File(rootDir), language.fileRegex);
    List<InputDocument> documents = Tool.load(allFiles, language);
    return validate(language, documents, fileToExclude,
                    Formatter.DEFAULT_K, outputDir, false, collectAnalysis);
}

public Triple<List<Formatter>,List<Float>,List<Float>> validateDocuments(boolean computeEditDistance,
                                                                         String outputDir)
    throws Exception
{
    return validateDocuments(Trainer.FEATURES_INJECT_WS, Trainer.FEATURES_HPOS,
                             computeEditDistance, outputDir);
}

public Triple<Formatter,Float,Float> validate(LangDescriptor language,
                                              List<InputDocument> documents,
                                              String fileToExclude,
                                              int k,
                                              String outputDir,
                                              boolean computeEditDistance,
                                              boolean collectAnalysis)
    throws Exception
{
    return validate(language, documents, fileToExclude,
                    k, Trainer.FEATURES_INJECT_WS, Trainer.FEATURES_HPOS,
                    outputDir, computeEditDistance, collectAnalysis);
}

/**
 * From {@code (A)?} build either:
 *
 * <pre>
 *  o--A-&gt;o
 *  |     ^
 *  o----&gt;|
 * </pre>
 *
 * or, if {@code A} is a block, just add an empty alt to the end of the
 * block
 */

@Override
public Handle optional(GrammarAST optAST, Handle blk) {
    BlockStartState blkStart = (BlockStartState)blk.left;
    ATNState blkEnd = blk.right;
    preventEpsilonOptionalBlocks.add(new Triple<Rule, ATNState, ATNState>(currentRule, blkStart, blkEnd));

    boolean greedy = ((QuantifierAST)optAST).isGreedy();
    blkStart.nonGreedy = !greedy;
    epsilon(blkStart, blk.right, !greedy);

    optAST.atnState = blk.left;
    return blk;
}

/**
 * From {@code (blk)+} build
 *
 * <pre>
 *   |---------|
 *   v         |
 *  [o-blk-o]-&gt;o-&gt;o
 * </pre>
 *
 * We add a decision for loop back node to the existing one at {@code blk}
 * start.
 */

@Override
public Handle plus(GrammarAST plusAST, Handle blk) {
    PlusBlockStartState blkStart = (PlusBlockStartState)blk.left;
    BlockEndState blkEnd = (BlockEndState)blk.right;
    preventEpsilonClosureBlocks.add(new Triple<Rule, ATNState, ATNState>(currentRule, blkStart, blkEnd));

    PlusLoopbackState loop = newState(PlusLoopbackState.class, plusAST);
    loop.nonGreedy = !((QuantifierAST)plusAST).isGreedy();
    atn.defineDecisionState(loop);
    LoopEndState end = newState(LoopEndState.class, plusAST);
    blkStart.loopBackState = loop;
    end.loopBackState = loop;

    plusAST.atnState = loop;
    epsilon(blkEnd, loop);      // blk can see loop back

    BlockAST blkAST = (BlockAST)plusAST.getChild(0);
    if ( ((QuantifierAST)plusAST).isGreedy() ) {
        if (expectNonGreedy(blkAST)) {
            g.tool.errMgr.grammarError(ErrorType.EXPECTED_NON_GREEDY_WILDCARD_BLOCK, g.fileName, plusAST.getToken(), plusAST.getToken().getText());
        }

        epsilon(loop, blkStart);    // loop back to start
        epsilon(loop, end);         // or exit
    }
    else {
        // if not greedy, priority to exit branch; make it first
        epsilon(loop, end);         // exit
        epsilon(loop, blkStart);    // loop back to start
    }

    return new Handle(blkStart, end);
}

/**
 * From {@code (blk)*} build {@code ( blk+ )?} with *two* decisions, one for
 * entry and one for choosing alts of {@code blk}.
 *
 * <pre>
 *   |-------------|
 *   v             |
 *   o--[o-blk-o]-&gt;o  o
 *   |                ^
 *   -----------------|
 * </pre>
 *
 * Note that the optional bypass must jump outside the loop as
 * {@code (A|B)*} is not the same thing as {@code (A|B|)+}.
 */

@Override
public Handle star(GrammarAST starAST, Handle elem) {
    StarBlockStartState blkStart = (StarBlockStartState)elem.left;
    BlockEndState blkEnd = (BlockEndState)elem.right;
    preventEpsilonClosureBlocks.add(new Triple<Rule, ATNState, ATNState>(currentRule, blkStart, blkEnd));

    StarLoopEntryState entry = newState(StarLoopEntryState.class, starAST);
    entry.nonGreedy = !((QuantifierAST)starAST).isGreedy();
    atn.defineDecisionState(entry);
    LoopEndState end = newState(LoopEndState.class, starAST);
    StarLoopbackState loop = newState(StarLoopbackState.class, starAST);
    entry.loopBackState = loop;
    end.loopBackState = loop;

    BlockAST blkAST = (BlockAST)starAST.getChild(0);
    if ( ((QuantifierAST)starAST).isGreedy() ) {
        if (expectNonGreedy(blkAST)) {
            g.tool.errMgr.grammarError(ErrorType.EXPECTED_NON_GREEDY_WILDCARD_BLOCK, g.fileName, starAST.getToken(), starAST.getToken().getText());
        }

        epsilon(entry, blkStart);   // loop enter edge (alt 1)
        epsilon(entry, end);        // bypass loop edge (alt 2)
    }
    else {
        // if not greedy, priority to exit branch; make it first
        epsilon(entry, end);        // bypass loop edge (alt 1)
        epsilon(entry, blkStart);   // loop enter edge (alt 2)
    }
    epsilon(blkEnd, loop);      // block end hits loop back
    epsilon(loop, entry);       // loop back to entry/exit decision

    starAST.atnState = entry;   // decision is to enter/exit; blk is its own decision
    return new Handle(entry, end);
}

public void checkCGen(String filename) throws Exception {
    URL testFolderURL = TestCGen.class.getClassLoader().getResource(SAMPLES_DIR);
    String testFolder = testFolderURL.getPath();
    String workingDir = getWorkingDir();

    String J_pathToFile = testFolder+"/"+filename;
    String C_filename = basename(filename)+".c";

    JTran jTran = new JTran();
    String C_code = jTran.translate(J_pathToFile, C_filename, false, false);

    Utils.writeFile(workingDir+"/"+C_filename, C_code);

    String[] indent_result_cmd = {
        "indent",
        "-bap", "-bad", "-br", "-nce", "-ncs", "-nprs", "-npcs", "-sai", "-saw",
        "-di1", "-brs", "-blf", "--indent-level4", "-nut", "-sob", "-l200",
        C_filename,
        "-o", C_filename // write on top of itself
    };

    // normalize generated code
    exec(indent_result_cmd, workingDir);

    // format the expected file as well
    String expected_C_CodeFilename = testFolder+"/"+C_filename;
    String[] indent_expected_cmd = {
        "indent",
        "-bap", "-bad", "-br", "-nce", "-ncs", "-nprs", "-npcs", "-sai", "-saw",
        "-di1", "-brs", "-blf", "--indent-level4", "-nut", "-sob", "-l200",
        expected_C_CodeFilename,
        "-o", "expected_"+C_filename
    };
    exec(indent_expected_cmd, workingDir);

    // compare with expected c file
    String[] diff_cmd = {
        "diff", "expected_"+C_filename, C_filename
    };
    Triple<Integer, String, String> result = exec(diff_cmd, workingDir);
    int execCode = result.a;
    String stdout = result.b;
    String stderr = result.c;

    assertEquals("", stdout);
    assertEquals("", stderr);
    assertEquals(0, execCode);
}

public static void main(String[] args) throws Exception {
    LeaveOneOutValidator validator = new LeaveOneOutValidator(JAVA8_DESCR.corpusDir, JAVA8_DESCR);
    Triple<List<Formatter>,List<Float>,List<Float>> results = validator.validateDocuments(false, "output");
    System.out.println(results.b);
    System.out.println(results.c);
}

public static void main(String[] args) throws Exception {
    LeaveOneOutValidator validator = new LeaveOneOutValidator(ANTLR4_DESCR.corpusDir, ANTLR4_DESCR);
    Triple<List<Formatter>,List<Float>,List<Float>> results = validator.validateDocuments(false, "output");
    System.out.println(results.b);
    System.out.println(results.c);
}

public static void main(String[] args) throws Exception {
    LeaveOneOutValidator validator = new LeaveOneOutValidator(JAVA_GUAVA_DESCR.corpusDir, JAVA_GUAVA_DESCR);
    Triple<List<Formatter>,List<Float>,List<Float>> results = validator.validateDocuments(false, "output");
    System.out.println(results.b);
    System.out.println(results.c);
}

public static void main(String[] args) throws Exception {
    LeaveOneOutValidator validator = new LeaveOneOutValidator(SQLITE_CLEAN_DESCR.corpusDir, SQLITE_CLEAN_DESCR);
    Triple<List<Formatter>,List<Float>,List<Float>> results = validator.validateDocuments(false, "output");
    System.out.println(results.b);
    System.out.println(results.c);
}

public static void main(String[] args) throws Exception {
    LeaveOneOutValidator validator = new LeaveOneOutValidator(JAVA_DESCR.corpusDir, JAVA_DESCR);
    Triple<List<Formatter>,List<Float>,List<Float>> results = validator.validateDocuments(false, "output");
    System.out.println(results.b);
    System.out.println(results.c);
}

public static void main(String[] args) throws Exception {
    LeaveOneOutValidator validator = new LeaveOneOutValidator(QUORUM_DESCR.corpusDir, QUORUM_DESCR);
    Triple<List<Formatter>,List<Float>,List<Float>> results = validator.validateDocuments(false, "output");
    System.out.println(results.b);
    System.out.println(results.c);
}

public static float[] getMedianErrorRates(LangDescriptor language, int maxNumFiles, int trials) throws Exception {
        SubsetValidator validator = new SubsetValidator(language.corpusDir, language);
        List<InputDocument> documents = Tool.load(validator.allFiles, language);
        float[] medians = new float[Math.min(documents.size(),maxNumFiles)+1];

        int ncpu = Runtime.getRuntime().availableProcessors();
        if ( FORCE_SINGLE_THREADED ) {
            ncpu = 2;
        }
        ExecutorService pool = Executors.newFixedThreadPool(ncpu-1);
        List<Callable<Void>> jobs = new ArrayList<>();

        for (int i = 1; i<=Math.min(validator.allFiles.size(), maxNumFiles); i++) { // i is corpus subset size
            final int corpusSubsetSize = i;
            Callable<Void> job = () -> {
                try {
                    List<Float> errorRates = new ArrayList<>();
                    for (int trial = 1; trial<=trials; trial++) { // multiple trials per subset size
                        Pair<InputDocument, List<InputDocument>> sample = validator.selectSample(documents, corpusSubsetSize);
                        Triple<Formatter, Float, Float> results = validate(language, sample.b, sample.a, true, false);
//                  System.out.println(sample.a.fileName+" n="+corpusSubsetSize+": error="+results.c);
//              System.out.println("\tcorpus =\n\t\t"+Utils.join(sample.b.iterator(), "\n\t\t"));
                        errorRates.add(results.c);
                    }
                    Collections.sort(errorRates);
                    int n = errorRates.size();
                    float median = errorRates.get(n/2);
                    System.out.println("median "+language.name+" error rate for n="+corpusSubsetSize+" is "+median);
                    medians[corpusSubsetSize] = median;
                }
                catch (Throwable t) {
                    t.printStackTrace(System.err);
                }
                return null;
            };
            jobs.add(job);
        }

        pool.invokeAll(jobs);
        pool.shutdown();
        boolean terminated = pool.awaitTermination(60, TimeUnit.MINUTES);
        return medians;
    }

public Triple<List<Formatter>,List<Float>,List<Float>> validateDocuments(FeatureMetaData[] injectWSFeatures,
                                                                         FeatureMetaData[] alignmentFeatures,
                                                                         boolean computeEditDistance,
                                                                         String outputDir)
    throws Exception
{
    List<Formatter> formatters = Collections.synchronizedList(new ArrayList<>());
    List<Float> distances = Collections.synchronizedList(new ArrayList<>());
    List<Float> errors = Collections.synchronizedList(new ArrayList<>());
    long start = System.nanoTime();
    try {
        List<String> allFiles = getFilenames(new File(rootDir), language.fileRegex);
        final List<InputDocument> documents = Tool.load(allFiles, language);
        final List<InputDocument> parsableDocuments = filter(documents, d -> d.tree!=null);
        long stop = System.nanoTime();
        System.out.printf("Load/parse all docs from %s time %d ms\n",
                          rootDir,
                          (stop-start)/1_000_000);

        int ncpu = Runtime.getRuntime().availableProcessors();
        if ( FORCE_SINGLE_THREADED ) {
            ncpu = 2;
        }
        ExecutorService pool = Executors.newFixedThreadPool(ncpu-1);
        List<Callable<Void>> jobs = new ArrayList<>();

        for (int i = 0; i<parsableDocuments.size(); i++) {
            final String fileName = parsableDocuments.get(i).fileName;
            Callable<Void> job = () -> {
                try {
                    Triple<Formatter, Float, Float> results =
                        validate(language, parsableDocuments, fileName,
                                 Formatter.DEFAULT_K, injectWSFeatures, alignmentFeatures,
                                 outputDir, computeEditDistance, false);
                    formatters.add(results.a);
                    float editDistance = results.b;
                    distances.add(editDistance);
                    Float errorRate = results.c;
                    errors.add(errorRate);
                }
                catch (Throwable t) {
                    t.printStackTrace(System.err);
                }
                return null;
            };
            jobs.add(job);
        }

        pool.invokeAll(jobs);
        pool.shutdown();
        pool.awaitTermination(60, TimeUnit.MINUTES);
    }
    finally {
        long final_stop = System.nanoTime();
        Double medianTrainingTime = median(trainingTimes);
        double medianFormattingPerMS = median(formattingTokensPerMS);
        System.out.printf("Total time %dms\n", (final_stop-start)/1_000_000);
        System.out.printf("Median training time %dms\n",
                          medianTrainingTime.intValue());
        System.out.printf("Median formatting time tokens per ms %5.4fms, min %5.4f max %5.4f\n",
                          medianFormattingPerMS,
                          BuffUtils.min(formattingTokensPerMS),
                          BuffUtils.max(formattingTokensPerMS));
    }
    return new Triple<>(formatters,distances,errors);
}

public Triple<Formatter,Float,Float> validate(LangDescriptor language,
                                                  List<InputDocument> documents,
                                                  String fileToExclude,
                                                  int k,
                                                  FeatureMetaData[] injectWSFeatures,
                                                  FeatureMetaData[] alignmentFeatures,
                                                  String outputDir,
                                                  boolean computeEditDistance,
                                                  boolean collectAnalysis)
        throws Exception
    {
        final String path = new File(fileToExclude).getAbsolutePath();
        List<InputDocument> others = filter(documents, d -> !d.fileName.equals(path));
        List<InputDocument> excluded = filter(documents, d -> d.fileName.equals(path));
        assert others.size() == documents.size() - 1;
//      kNNClassifier.resetCache();
        if ( excluded.size()==0 ) {
            System.err.println("Doc not in corpus: "+path);
            return null;
        }
        InputDocument testDoc = excluded.get(0);
        long start = System.nanoTime();
        Corpus corpus = new Corpus(others, language);
        corpus.train();
        long stop = System.nanoTime();
        Formatter formatter = new Formatter(corpus, language.indentSize, k, injectWSFeatures, alignmentFeatures);
        InputDocument originalDoc = testDoc;
        long format_start = System.nanoTime();
        String output = formatter.format(testDoc, collectAnalysis);
        long format_stop = System.nanoTime();
        float editDistance = 0;
        if ( computeEditDistance ) {
            editDistance = normalizedLevenshteinDistance(testDoc.content, output);
        }
        ClassificationAnalysis analysis = new ClassificationAnalysis(originalDoc, formatter.getAnalysisPerToken());
        System.out.println(testDoc.fileName+": edit distance = "+editDistance+", error rate = "+analysis.getErrorRate());
        if ( outputDir!=null ) {
            File dir = new File(outputDir+"/"+language.name+"/"+Tool.version);
            if ( !dir.exists() ) {
                dir.mkdirs();
            }
            Utils.writeFile(dir.getPath()+"/"+new File(testDoc.fileName).getName(), output);
        }
        long tms = (stop - start) / 1_000_000;
        long fms = (format_stop - format_start) / 1_000_000;
        trainingTimes.add((double)tms);
        float tokensPerMS = testDoc.tokens.size() / (float) fms;
        formattingTokensPerMS.add((double)tokensPerMS);
        System.out.printf("Training time = %d ms, formatting %d ms, %5.3f tokens/ms (%d tokens)\n",
                          tms,
                          fms,
                          tokensPerMS, testDoc.tokens.size());
//      System.out.printf("classify calls %d, hits %d rate %f\n",
//                        kNNClassifier.nClassifyCalls, kNNClassifier.nClassifyCacheHits,
//                        kNNClassifier.nClassifyCacheHits/(float) kNNClassifier.nClassifyCalls);
//      System.out.printf("kNN calls %d, hits %d rate %f\n",
//                        kNNClassifier.nNNCalls, kNNClassifier.nNNCacheHits,
//                        kNNClassifier.nNNCacheHits/(float) kNNClassifier.nNNCalls);
        return new Triple<>(formatter, editDistance, analysis.getErrorRate());
    }

public static void main(String[] args) throws Exception {
    LeaveOneOutValidator validator = new LeaveOneOutValidator(TSQL_CLEAN_DESCR.corpusDir, TSQL_CLEAN_DESCR);
    Triple<List<Formatter>,List<Float>,List<Float>> results = validator.validateDocuments(false, "output");
    System.out.println(results.b);
    System.out.println(results.c);
}