JAVA Read file conversion character Set __java

Scene: Need to read a TXT file content, but the TXT is written using UTF-8, the development environment default character set is GBK.

Core code:

 Public arraylist<string[]> ReadData (String srcpath) {//TODO auto-generated method stub arraylist& Lt
        string[]> DataList = new arraylist<string[]> ();
        String errorpath = null;
        String backuppath = null;
        String receivedpath = null;

        File srcfile = null;
            try {backuppath = Adjustpath (Srcpath, "Backup");
            Errorpath = Adjustpath (Srcpath, "error");

            Receivedpath = Adjustpath (Srcpath, "received");
            string[] Infilenames = new File (receivedpath). List (); if (null!= infilenames && infilenames.length > 0) {for (String filename:infilenames) {//AS
                    The fruit folder has multiple files read String FilePath = receivedpath + file.separator + filename;
                    Srcfile = new File (FilePath);
                    System.out.println (Srcfile.length ());
                    InputStream fis = new FileInputStream (srcfile); Unicodereader ur = new Unicodereader (FIS, "UTF-8");
                    BufferedReader br = new BufferedReader (UR);
                    string[] Spliteddata = null;
                    String str = NULL; while (str = Br.readline ())!= null && str.length () > 0) {spliteddata = Splitstr (str
                        );
                    Datalist.add (Spliteddata);
                    } br.close ();
                    Ur.close ();

                Fis.close ();
            (FileNotFoundException e) {logger.error ("exception occurred" + E.getmessage ());
        E.printstacktrace ();
            catch (IOException IoE) {ioe.printstacktrace ();
        Logger.error ("Off Stream exception:" + Ioe.getmessage ());
    return DataList; }

The point is,
while ((str = br.readline ())!= null && str.length () > 0)

In this decision condition, Br.readline () is first assigned to STR
The character is read as Br.readline () executes this method
So it can't be written
while (Br.readline ())!= null && br.readline ()) > 0) {str = Br.readline ()}
In this case str reads out forever as null;

At the same time UTF-8 encoded format TXT file, beginning with the BOM, through this way to parse, the first line of the first word identifier.
You need to create a new reader subclass to refer to
http://blog.csdn.net/jackpk/article/details/5702964/