Selecting second column (FileReader)

Question

I'm looping csv. I have two question:

1) I'm selecting second column by name like

if(tab[1].equals("Col2")

I don't want to put the name of column. I want to select just second column.

2) how to skip first line (header)

Here is sample of code to looping csv:

String csvFile = "C:\\test.csv";
BufferedReader br = null;
String line = "";
String cvsSplitBy = ";";

try{
    br = new BufferedReader(new FileReader(csvFile));

    while ((line = br.readLine()) != null) {
        String[] tab=line.split(cvsSplitBy);      

        int tmp;
        if(tab[1].equals("Col2")){                

            tmp = Integer.parseInt(tab[2]);

                for(int i=0;i<tmp;i++){  
                   // TO DO
                }
        }
    }
}

Answer 1

Better to make use of CSVReader for this, which provides lot of APIs for processing your csv files. Here is a complete working code, ofcourse, without exception handling.

String csvFile = "C:\\test.csv";
CSVReader reader;
String[] nextRow;
char cvsSplitBy = ';';

try {

    //Last argument will determine how many lines to skip. 1 means skip header
    reader = new CSVReader(new FileReader(csvFile), cvsSplitBy, CSVParser.DEFAULT_QUOTE_CHARACTER, 1);

    while ((nextRow = reader.readNext()) != null) {

        if(nextRow.length > 2){
            //nextRow[1] will always give second column value
            int tmp = Integer.parseInt(nextRow[1]);
            for (int i = 0; i < tmp; i++) {
                // TO DO
            }
        }
    }
} catch (FileNotFoundException e) {
    e.printStackTrace();
} catch (IOException e) {
    e.printStackTrace();
}

Answer 2

Here is an example using Apache Commons CSV, and its CSVParser.

The first line is considered to be the header and is skipped (withFirstRecordAsHeader()) , the "columns" of each record can be accessed with their index (get(int)) .The indexes are 0-based .

Just adapt the charset and CSVFormat to your needs.

CSVParser parser = null;

try {

    parser = CSVParser.parse(new File(csvFile), Charset.forName("UTF-8"),
            CSVFormat.RFC4180.withFirstRecordAsHeader());

    List<CSVRecord> records = parser.getRecords();

    for (CSVRecord record : records) {

        int tmp = Integer.parseInt(record.get(1));

        for (int i = 0; i < tmp; i++) {
            // TO DO
        }

    }

} catch (IOException e) {

    e.printStackTrace();

} finally {
    try {
        parser.close();
    } catch (IOException e) {

    }

}

Answer 3

With univocity-parsers this becomes a piece of cake:

CsvParserSettings parserSettings = new CsvParserSettings(); //many options here, check the tutorial.
parserSettings.setHeaderExtractionEnabled(true); //header is extracted and not part of the result
parserSettings.selectIndexes(1); //select 2nd column (indexes are 0-based)
CsvParser parser = new CsvParser(parserSettings);
List<String[]> allRows = parser.parseAll(csvFile);

Note that this will work even if some of the rows are empty of have just one column, whereas all other solutions posted here will fail unless you handle such situations by yourself.

Not only this involves WAY less code (and complexity), the parser is also ~4 times faster than Commons CSV, and ~3 times faster than OpenCSV.

Disclaimer: I'm the author of this library, it's open-source and free (Apache v2.0 License)