Object reading and writing

Question

i have to read a very large files that consist of header(5000 bytes)main file(10,000 bytes), footer(4000 bytes) this is a binary file.The current code is in c++, the logic implemented is like as:- there are three classes and the data structure defined for these classes for these three classes is same as the information stored in binary.Header data structure is 5000 bytes footer is 4000 bytes and then main.So when ever the large file is read the process invokes read method and pass the object of header and a pointer for reading 5000 bytes.As structure is same all the values read are assigned accordingly to header in single shot.Here is example
class header
char name array[2000];
char Address array[2000];
in reaqd methods i pass object of header. binary file name and a pointer to read first 4000 bytes
so it aitoamtically assign first 2000 to name rest to address
same funtionality i have to do in java.I am not able to do it.
is it posilbe ot i have to read byte array and then fire setmethods for each data to do this or can i simulate this c++ funtionality in java .
pl help

Answer 1

there are
three classes and the data structure defined for these
classes for these three classes is same as the
information stored in binary.So, if I understand what you want to do, you need to read C++ binary data into a Java class, but don't want to go through the process of translating that data and calling individual getters and setters?
First thing: be aware that the byte order of C++ data will differ, depending on the machine that wrote the file. If you're reading files from different machines, you may not get back what you expect (which is one reason that XML exists, to store everything in text rather than binary).
Second: do you have a Java class that exactly mimics the C++ class? You need to be aware of field size and type (eg, a 16-bit C++ int corresponds to a Java short), and recognize that the same types may have different sizes depending on the C++ implementation.
If you have considered these two points, then look at java.lang.reflect.
1) Define a static array containing the names of the fields in your object, in the order they appear in the data. This is necessary because Class.getFields() does not guarantee order of the fields returned.
2) Call getClass() on the passed object.
3) For every field in your list from step 1, call getField() on the Class object that you got from step 2.
4) Read the appropriate amount of data from the file (see below), and use the appropriate set method on the Field object that you got in step 3.
How do you know how much data to read from the file? There are a couple of ways to do this. You can use the getType() method on the Field object (from step 3), then use this to find an appropriate "read and convert" routine. Or, if you want to drive from the C++ code, the table that you define in step 1 could identify the C++ type.
This is, of course, only a skeleton description of what you need to do. But as outlined, it should take only a day or two to implement. The real problem is if your C++ structures change frequently, or you'll be processing data from multiple sources (so byte order and field size may change). In that case, you may (1) find it best to implement a parser to build your field table, and (2) drive everything off the field table, which would then have to contain additional information about the appropriate conversion routine.
Also, you could check SourceForge or other open-source sites, to see if someone has already done this.