Java CSVFormat.TDF属性代码示例

public DebateHTMLParser()
        throws IOException
{
    urlStances = new HashMap<>();

    InputStream inputStream = this.getClass().getClassLoader()
            .getResourceAsStream("rfd-controversies/rfd-manual-cleaning-controversies.tsv");

    CSVParser parser = new CSVParser(new InputStreamReader(inputStream), CSVFormat.TDF);
    for (CSVRecord csvRecord : parser) {
        boolean keepRFDDebate = "".equals(csvRecord.get(0));

        if (keepRFDDebate) {
            String url = csvRecord.get(5);

            SortedSet<String> stances = new TreeSet<>();
            // yes/no stance?
            if ("y".equals(csvRecord.get(3))) {
                stances.addAll(Arrays.asList("Yes", "No"));
            }
            else {
                stances.addAll(Arrays.asList(csvRecord.get(3), csvRecord.get(4)));
            }

            if (stances.size() != 2) {
                throw new IllegalStateException(
                        "Expected 2 stances but got " + stances + "; " + csvRecord.get(1));
            }

            urlStances.put(url, stances);
        }
    }
}
 

@Test
public void evaluateSpeakerAssignments() throws Exception {
	JCasIterator iter = SimplePipeline
			.iteratePipeline(
					CollectionReaderFactory.createReaderDescription(XmiReader.class,
							XmiReader.PARAM_SOURCE_LOCATION, "src/test/resources/SpeakerIdentifier/tx4z.0.xmi",
							XmiReader.PARAM_LENIENT, true),
					createEngineDescription(FigureReferenceAnnotator.class),
					createEngineDescription(SpeakerIdentifier.class, SpeakerIdentifier.PARAM_CREATE_SPEAKER_FIGURE,
							false),
					createEngineDescription(XmiWriter.class, XmiWriter.PARAM_TARGET_LOCATION, "target/doc"))
			.iterator();
	StringBuilder b = new StringBuilder();
	CSVPrinter writer = new CSVPrinter(b, CSVFormat.TDF);
	SummaryStatistics types = new SummaryStatistics();
	SummaryStatistics tokens = new SummaryStatistics();
	while (iter.hasNext()) {
		JCas jcas = iter.next();
		int s = 0;
		int all = 0;
		Counter<String> unassigned = new Counter<String>();
		for (Speaker speaker : JCasUtil.select(jcas, Speaker.class)) {
			if (speaker.getFigure() == null) {
				unassigned.add(speaker.getCoveredText());
				s++;
			}
			all++;
		}
		types.addValue(unassigned.size());
		tokens.addValue(s);
		writer.printRecord(JCasUtil.selectSingle(jcas, Drama.class).getDocumentId(), s, unassigned.size(), all);
	}
	writer.printRecord("mean", tokens.getMean(), types.getMean());
	writer.printRecord("min", tokens.getMin(), types.getMin());
	writer.printRecord("max", tokens.getMax(), types.getMax());
	writer.close();
	System.out.println(b.toString());
}
 

/**
 * Returns a CSVFormat object given the CSV format as a string.
 *
 * @param format
 * @return
 */
private CSVFormat getCsvFormat(String format) {
    CSVFormat csvFormat = null;

    switch (format.trim().toLowerCase()) {
        case "default":
            csvFormat = CSVFormat.DEFAULT;
            break;
        case "excel":
            csvFormat = CSVFormat.EXCEL;
            break;
        case "informixunload":
        case "informix-unload":
        case "informix_unload":
            csvFormat = CSVFormat.INFORMIX_UNLOAD;
            break;
        case "informixunloadcsv":
        case "informix-unload-csv":
        case "informix_unload_csv":
            csvFormat = CSVFormat.INFORMIX_UNLOAD_CSV;
            break;
        case "mysql":
            csvFormat = CSVFormat.MYSQL;
            break;
        case "postgres":
        case "postgresql-csv":
        case "postgresql_csv":
            csvFormat = CSVFormat.POSTGRESQL_CSV;
            break;
        case "postgresql-text":
        case "postgresql_text":
            csvFormat = CSVFormat.POSTGRESQL_TEXT;
            break;
        case "rfc4180":
            csvFormat = CSVFormat.RFC4180;
        case "tdf":
            csvFormat = CSVFormat.TDF;
        default:
            throw new RuntimeException(String.format("CSV format \"%s\" is not among the supported formats"));
    }

    return csvFormat;
}
 

public TSVWriter() {
	super(CSVFormat.TDF);
}
 

public static DataTable loadCSV(String fileName, String formatType, VariableType[] colTypesOverride, String[] colNamesOverride, boolean hasHeaderRow) {
	try {
		// use apache commons io + csv to load but convert to list of String[]
		// byte-order markers are handled if present at start of file.
		FileInputStream fis = new FileInputStream(fileName);
		final Reader reader = new InputStreamReader(new BOMInputStream(fis), "UTF-8");
		CSVFormat format;
		if ( formatType==null ) {
			format = hasHeaderRow ? CSVFormat.RFC4180.withHeader() : CSVFormat.RFC4180;
		}
		else {
			switch ( formatType.toLowerCase() ) {
				case "tsv":
					format = hasHeaderRow ? CSVFormat.TDF.withHeader() : CSVFormat.TDF;
					break;
				case "mysql":
					format = hasHeaderRow ? CSVFormat.MYSQL.withHeader() : CSVFormat.MYSQL;
					break;
				case "excel":
					format = hasHeaderRow ? CSVFormat.EXCEL.withHeader() : CSVFormat.EXCEL;
					break;
				case "rfc4180":
				default:
					format = hasHeaderRow ? CSVFormat.RFC4180.withHeader() : CSVFormat.RFC4180;
					break;
			}
		}
		final CSVParser parser = new CSVParser(reader, format);
		List<String[]> rows = new ArrayList<>();
		int numHeaderNames = parser.getHeaderMap().size();
		try {
			for (final CSVRecord record : parser) {
				String[] row = new String[record.size()];
				for (int j = 0; j<record.size(); j++) {
					row[j] = record.get(j);
				}
				rows.add(row);
			}
		}
		finally {
			parser.close();
			reader.close();
		}

		VariableType[] actualTypes = computeColTypes(rows, numHeaderNames);

		Set<String> colNameSet = parser.getHeaderMap().keySet();
		String[] colNames = colNameSet.toArray(new String[colNameSet.size()]);
		if ( colNamesOverride!=null ) {
			colNames = colNamesOverride;
		}
		if ( colTypesOverride!=null ) {
			actualTypes = colTypesOverride;
		}
		return fromStrings(rows, actualTypes, colNames, false);
	}
	catch (Exception e) {
		throw new IllegalArgumentException("Can't open and/or read "+fileName, e);
	}
}