GMP_47 GMP_47 - 2 months ago 12
Java Question

Java - How to read Byte by Byte a .dat file

I'm trying to read a .dat using Java with no other Classes.
This is the file's structure:
Header
Serial: Word; //2 bytes
Filename: String[255]; //1 byte
Date: Word; //2 bytes
FieldNumbers: Word; //2 bytes
NumbersOfRecords: Word; //2 bytes

Info about Fields
FieldCode: Word; //2 bytes
FieldName: ShortString; //1 byte

Info in Field
FieldCode: Word; //2 bytes
FieldText: String[255]; //1 byte

DateTime = double


What I must know is how to use BufferedReader to get each Byte, read it as an int, then turn same int to a string and show it on screen.
Can I create different Methods for reading each type of data? Can I make it read 2 bytes at the same time?

UPDATE:

package binarios5;

import java.io.FileNotFoundException;
import java.io.IOException;
import java.nio.ByteBuffer;
import java.nio.ByteOrder;
import java.nio.charset.StandardCharsets;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;

public class Main5
{

public static void main(String[] args) throws FileNotFoundException, IOException
{

try
{
Path path = Paths.get("C:\\\\Dev-Pas\\\\EXAMEN2.dat");
System.out.println("File open");
byte[] bytes = Files.readAllBytes(path);
ByteBuffer buffer = ByteBuffer.wrap(bytes);
buffer.order(ByteOrder.BIG_ENDIAN);
short serial = buffer.getShort();
System.out.println("----[CONTENIDO DEL ARCHIVO]--------------------");
System.out.println("Nro. de Serie: " + serial);
int largoCadena = buffer.get();//Bytes 1 int Longitud de la cadena
//System.out.println("largoCadena: " + largoCadena);//33
byte[] bytesChar = new byte[largoCadena];//CString
buffer.get(bytesChar);
String nombre = new String(bytesChar, StandardCharsets.ISO_8859_1);
System.out.println("Nombre: " + nombre);

short date = buffer.getShort();//FALTA DECODIFICAR FECHA
System.out.println("Fecha sin procesar. "+date);//FALTA DECODIFICAR FECHA

short cantCampos = buffer.getShort(); //cantidad de campos que tienen los registros
System.out.println("Cantidad de Campos Customizados: "+cantCampos);//debe decir 4
int[] codCampo = new int[cantCampos];
String[] nombreCampo = new String[10];


for (int i = 0; i < cantCampos; i++) //leer RegType segun la cantidad de campos
{
codCampo[i] = buffer.getShort();//Bytes 2 codigo del campo
int largoCadena2 = buffer.get();//Bytes 1 int Longitud de la cadena
byte[] bytesChar2 = new byte[largoCadena2];
buffer.get(bytesChar2);
nombreCampo[i] = new String(bytesChar2, StandardCharsets.ISO_8859_1);
}

for (int i = 0; i < cantCampos; i++)//mostrar codigos y campos
{
System.out.println("Campo [codigo: " + codCampo[i] + ", descripcion: " + nombreCampo[i] + "]");
}

short cantRegistros = buffer.getShort();//cantidad de registros total
System.out.println("Cantidad de Registros: "+cantRegistros);
System.out.println("-----------------------");//OK

String[] contenidoCampo = new String[10];
for (int i = 0; i < cantRegistros; i++) //leyendo RegData 5 veces
{
short cantCamposCompletos = buffer.getShort();

for (int j = 0; j < cantCamposCompletos; j++)
{
short codCampoInterno = buffer.getShort();
int largoCadena3 = buffer.get();
byte[] bytesChar3 = new byte[largoCadena3];
buffer.get(bytesChar3);
contenidoCampo[j] = new String(bytesChar3, StandardCharsets.ISO_8859_1);
System.out.println(nombreCampo[j]+": "+contenidoCampo[j]);
}
System.out.println("-----------------------");
}

System.out.println("----[FIN CONTENIDO DEL ARCHIVO]-----------------");
}
catch (IOException e)
{
System.out.println("File I/O error!");
}

}


}

Answer

In java Reader and Writer are for Unicode text, String, 2-bytes char.

For binary data, byte[] one needs an InputStream, OutputStream.

One can use an InputStream:

BufferedInputStream in = new BufferedInputStream(new FileInputStream(...));

In your case you want to read short and such. For that you could wrap it around a DataInputStream.

However using a ByteBuffer is easiest to begin with. It can be read from a file (FileChannel), but the simple case is:

Path path = Paths.get("C:/xxx/yyy.dat");
byte[] bytes = Files.readAllBytes(path);
ByteBuffer buffer = ByteBuffer.wrap(bytes);
//buffer.order(ByteOrder.LITTLE_ENDIAN); // So short is read as LSB,MSB

Worked out:

// Header
short serial = buffer.getShort();
byte[] fileNameB = new byte[255];
buffer.get(fileNameB);
// If 0 terminated asciz string:
int len = fileNameB.length;
for (int i = 0; i < fileNameB.length; ++i) {
    if (fileNameB[i] == 0) {
        len = i;
        break;
    }
}
String fileName = new String(fileNameB, 0, len, StandardCharsets.ISO_8859_1);

short date = buffer.getShort();
short fieldNumbers = buffer.getShort();
short numbersOfRecords = buffer.getShort();

for (int fieldI = 0; fieldI < fieldNumber; ++fieldI) {
    // Info about Fields 
    short fieldCode = buffer.getShort();
    //byte fieldName: ShortString;   //1 byte
}

Info in Field FieldCode: Word; //2 bytes FieldText: String[255]; //1 byte

DateTime = double

String getPascalString(ByteBuffer buffer) {
    int length = buffer.get() & 0xFF;
    byte[] bytes = new byte[length];
    buffer.get(bytes);
    return new String(bytes, StandardCharsets.ISO_8859_1);
}

Will deliver: d:/documentos/te...

short packedDate = buffer.getShort();
int year = packedDate & 0x7F; // + 1900?
int month = (packedDate >> 7) & 0xF:
int day = (packedDate >> 11) & 0x1F;