Java Code Examples for org.apache.commons.csv.CSVFormat#newFormat()

@Bean("movieItems")
List<MovieItem> getItems(DataSpace movieDataSpace) throws Exception {
    File movieItemFile = new File("data/ml-100k/u.item");
    List<MovieItem> items = new LinkedList<>();

    QualityAttribute<Integer> itemAttribute = movieDataSpace.getQualityAttribute("item");
    try (InputStream stream = new FileInputStream(movieItemFile); InputStreamReader reader = new InputStreamReader(stream, StringUtility.CHARSET); BufferedReader buffer = new BufferedReader(reader)) {
        try (CSVParser parser = new CSVParser(buffer, CSVFormat.newFormat('|'))) {
            Iterator<CSVRecord> iterator = parser.iterator();
            while (iterator.hasNext()) {
                CSVRecord datas = iterator.next();
                // 物品标识
                int id = Integer.parseInt(datas.get(0));
                // 物品索引
                int index = itemAttribute.convertData(id);
                // 物品标题
                String title = datas.get(1);
                // 物品日期
                LocalDate date = StringUtility.isEmpty(datas.get(2)) ? LocalDate.of(1970, 1, 1) : LocalDate.parse(datas.get(2), formatter);
                MovieItem item = new MovieItem(index, title, date);
                items.add(item);
            }
        }
    }

    items = new ArrayList<>(items);
    return items;
}
 

/**
 * 
 * @param delimiter
 * @param encapsulator
 * @param ignoreSurroundingSpaces
 * @param ignoreEmptyLines
 * @return
 */
public static CSVFormat getFormat(char delimiter,
		Character encapsulator,
		boolean ignoreSurroundingSpaces, 
		boolean ignoreEmptyLines) {
	CSVFormat format = CSVFormat.newFormat(delimiter);
	format = format.withIgnoreEmptyLines(ignoreEmptyLines)
       		.withIgnoreSurroundingSpaces(ignoreSurroundingSpaces)
       		.withAllowMissingColumnNames(true)
       		.withQuote(encapsulator);
	
	return format;
}
 

public Map<String, List<String>> readTextFile(File textFile) throws Exception {
	
	Map<String, List<String>> ret = new HashMap<String, List<String>>();
	
	String delimiter = config.getUserSyncFileSourceDelimiter();
	
	CSVFormat csvFormat = CSVFormat.newFormat(delimiter.charAt(0));
	
	CSVParser csvParser = new CSVParser(new BufferedReader(new FileReader(textFile)), csvFormat);
	
	List<CSVRecord> csvRecordList = csvParser.getRecords();
	
	if ( csvRecordList != null) {
		for(CSVRecord csvRecord : csvRecordList) {
			List<String> groups = new ArrayList<String>();
			String user = csvRecord.get(0);
			
			user = user.replaceAll("^\"|\"$", "");
				
			int i = csvRecord.size();
			
			for (int j = 1; j < i; j ++) {
				String group = csvRecord.get(j);
				if ( group != null && !group.isEmpty()) {
					 group = group.replaceAll("^\"|\"$", "");
					 groups.add(group);
				}
			}
			ret.put(user,groups);
		 }
	}

	csvParser.close();

	return ret;
}
 

/**
 * Reads tab delimited file and returns them as list of list, with [x] =
 * colummn and [x][y] is value in column. Needed for reading counts
 * file, as there the rows are the samples, as opposed to expression and
 * genotype file where the columns are the samples. Needs to be read in
 * memory, so minimal memory requirement is larger than the size of the
 * counts file.
 * 
 * 
 * @param filepath The path to a tab delimited file to read
 * 
 * @return A 2D array with each array being one column from filepath except first column
 * 		   and a 1D array with the first column (without header)
 * 
 * @throws IOException	If file at filepath can not be read
 */
public static Object[] readTabDelimitedColumns(String filepath) throws IOException {
	List<List<String>> allColumns = new ArrayList<List<String>>();
	// parses file on tabs
	CSVParser parser = new CSVParser(new FileReader(filepath), CSVFormat.newFormat('\t'));
	Boolean header = true;
	int rowNumber = 0;
	int columnIndexHeader = 0;
	List<String> firstColumn = new ArrayList<String>();
	for (CSVRecord row : parser) {
		rowNumber++;
		// starts at 1 because 1st element of column is the samplename, unless its the header row
		int columnStart = 1;
		if(header){
			columnStart = 0;
		}
		for (int columnIndex = columnStart; columnIndex < row.size(); columnIndex++) {
			// header can start from 0 if it is R styled, so check if element 0 has a value
			// R style is e.g.
			// colNameA	colNameB
			// rowNameA	AAValue	AAvalue
			// rownameB ABValue BAvalue
			// while csv style has a tab before colNameA
			if(header){
				String columnValue = row.get(columnIndex);
				if(columnValue.length() == 0){
					continue;
				}
				allColumns = addSingleValueTo2DArray(allColumns, columnIndexHeader,columnValue);
				columnIndexHeader++;
				continue;
			}
			else{
				// This changes the allColumns list of list in place, e.g. for example loop -> [[]] -> [[1]] -> [[1,2]] -> [[1,2],[3]] -> etc
				allColumns = addSingleValueTo2DArray(allColumns, columnIndex - 1, row.get(columnIndex));
			}
		}
		if(!header){
			firstColumn.add(row.get(0));
			if(row.size()-1 != columnIndexHeader){
				DeconvolutionLogger.log.info(String.format("Table %s does not have the same number of columns as there are in the header at row %d",filepath,rowNumber));
				DeconvolutionLogger.log.info(String.format("Number of header columns: %d",columnIndexHeader));
				DeconvolutionLogger.log.info(String.format("Number of columns at row %d: %d", rowNumber, row.size()-1));
				DeconvolutionLogger.log.info(row.toString());
				parser.close();
				throw new RuntimeException(String.format("Cellcount percentage table does not have the same number of columns as there are celltypes at row %d",rowNumber));
			}
		}
		if(header){
			header = false;
		}

	}
	parser.close();
	return new Object[] {firstColumn, allColumns};
}