What Is a translating program which translates entire source program into machine language program?

Java

High-level source code is converted to native machine code either by an interpreter or a compiler. Interpreted code requires a runtime in order to perform the conversion each time the program is executed whereas compiled code typically produces a standalone machine code program that can be executed without further interpretation. There are some exceptions, most notably Java which is first compiled to Java byte code which is then interpreted by the Java virtual machine on each execution.

An interpreter always converts high-level code to native machine code but, because it must perform the conversion while the code is executing via the interpreter, interpreted code executes much more slowly than native machine code.

Technically, a compiler is simply a program that converts high-level code to a lower-level code. That lower-level code could be another high-level language. For example, a C++ source could be compiled to produce a C source. However, converting the other way around (such as from C to C++) can only be done manually because humans can think in abstract terms far more easily than machines can.

Low-level assembly language source are converted to machine code using an assembler. Machine code can also be disassembled to produce an assembly-like source known as a disassembly, however the resultant code is extremely low-level because there are no named variables (only memory offsets) and all user-comments will have been stripped out during assembly or compilation.