How to remove ^M characters from a text file independent of platform using Java?

I am trying to figureout how to delete ^M characters from a text file that is generated from Java code with following code.

   public StringBuilder toCsv(Table table) {
        StringBuilder stringBuilder = new StringBuilder();
        String csv = new String();
        for (Column cName : table.getColumns()){
            csv += QUOT;
            csv += cName.getName();
            csv += QUOT;
            csv += CSV_SEPERATOR;
        }

        csv += "\n";
        stringBuilder.append(csv);

        for (Row row : table) {
            Collection<Object> values = row.values();
            String csvString = "";
            if (values.size() == 10) {
                String ep = QUOT + CSV_SEPERATOR + QUOT ;
                csvString = StringUtils.join(row.values(),ep );
                csvString.replaceAll("\'", "");
                csvString = QUOT + csvString + QUOT;
                logger.info("line ++++ " + csvString);
            }
            stringBuilder.append(csvString);
            stringBuilder.append("\n");
        }

        return stringBuilder;
    }

then I am using following method to write the data to file

 public void writeCsv(String data, String path, String fileName) throws IOException {
    String completePath = path + "/" + fileName;
    Writer out = new BufferedWriter(new OutputStreamWriter(
            new FileOutputStream(completePath)));
    try {
        out.write(data);
    } finally {
        out.close();
    }
}

Context

I am generating CSV files using http://jackcess.sourceforge.net/ from a Microsoft access (.mdb) file. When I generate csv and open using vim, I see lots of ^M in the middle of lines. NOTE: I am on MacOS

I have tried following to remove ^M (which I believe is a MS Windows CARRIAGE_RETURN) before writing to csv

 csvLine.replaceAll("\n\r", "");

AND

 csvLine.replaceAll("\r\n", "");

AND

csvLine.replaceAll("\\r", "");

Generated CSV

'10773.0';'';'';'';'Thu Jul 14 00:00:00 CEST 2016';'By Cash';'';'10000.0';'';'2102.0'
    '10001.0';'';'';'';'Thu Jul 14 00:00:00 CEST 2016';'Pet Soup cash';'087470^M
        ^M
        ^M
087470';'-45000.0';'';'2102.0'
'10360.0';'';'';'';'Thu Jul 14 00:00:00 CEST 2016';'By Cash';'';'37000.0';'';'2101.0'
'10444.0';'';'';'';'Thu Jul 14 00:00:00 CEST 2016';'By Cash';'';'2000.0';'';'2101.0'

As you can see one line in above CSV is broken with ^M which is not desired. I need to programmatically remove such characters from the file.

Expected output after removing ^M and proceeding line

'10001.0';'';'';'';'Thu Jul 14 00:00:00 CEST 2016';'Pet Soup cash';'087470087470';'-45000.0';'';'2102.0'

Any help will be appreciated.

Solution

Strings are immutable, so the .replaceAll method does not change the value of the existing String; it performs the replacement and returns a new String value. So,

String csvString = "123,foo,234";
csvString.replaceAll("foo", "");
System.out.println(csvString);

prints

123,foo,234

showing that the string is unchanged. What you want to do is

String csvString = "123,foo,234";
csvString = csvString.replaceAll("foo", "");  // save the new value
System.out.println(csvString);

which prints

123,,234

In your particular case, it looks like you want to do

csvString = csvString.replaceAll("\r\n", "");  // save the new value

since you want to remove both the carriage_return (which appears as ^M) and the new_line (which starts a new line in the text file).