Programming Language Translators
Contents
Overview
CraigNDave
https://www.youtube.com/watch?v=qHM4rtpYuxE&list=PLCiOXwirraUDJYD-VxmztyhsDP0QM2BHW&index=3
Computer Science Tutor
https://www.youtube.com/watch?v=eXu0c5vZnhU&list=PL04uZ7242_M6O_6ITD6ncf7EonVHyBeCm&index=16
TRC Video
https://www.youtube.com/watch?v=Z01WWoqO-Kw
Assembler
An assembler translates assembly language into machine code. One line of assembler will create a single line of machine code, because the assembler mnemonic is converted into the binary instruction.
Compiler
A compiler translates code written in a high level programming language into object code, the machine code of a particular machine. It 'compiles' the entire source code before translating it. Remember this compiles the whole program at the same time, and any syntax error within the program will cause the compiler to fail. Also remember that once compiled you only need to recompile if the program is changed.
Interpreter
An interpreter also translates code written in a programming language but unlike a compiler it does this one line of code at a time, and it doesn't create any object code which means the program is interpreted every time it runs. This makes it slower to run than a compiler, but interpreters are useful for partially testing or debugging programs. Especially because the interpreter will only find errors if it makes it to the syntax error.
One key issue with interpreters, is iteration (loops). Each run through the loop will be interpreted again even though its the same code as before, a compiler just compiles once.
Compiler vs Interpreter
Compiler over Interpreter
- Object code is saved, so you only need to compile once so no need to re-compile. Obviously if the code is changed it will need to be re-compiled.
- Executes faster than interpreted code, because interpreted code is interpreted and execute one at a time. Compiled code is simply executed.
- Object code can be distributed (without the compiler).
- Secure because the source code is hidden, the compiler generated object code is not easily reverse engineered.
- Compilers can be more efficient, for example an interpreter must interpret and then execute each time it passes through a loop. Compiled code can just execute each time through the loop.
Interpreter over Compiler
- Easier to partially test or debug programs, it will run until an error is reached. This will therefore allow you to test code that you haven't completely finished.
Types of Code
Source Code
The code written by the programmer.
Object Code
The code generated by compiling your source code. This object code can then be executed without the need to recompile.
Byte Code
Compilers & Interpreters would suggest there are clear divides between the languages which are interpreted and the languages that are compiled. This is not the case, some languages use both. The source code is first compiled into byte code, and then when run the byte code is then interpreted.
Languages such as Java work in this way. Java code is compiled into byte code which can then be distributed onto any platform. This is because the byte code is an instruction set than can be executed in a virtual machine. The virtual machine can then emulate the architecture of a computer allowing the byte code to be executed on any platform.
https://www.youtube.com/watch?v=gWKoW_Z5Liw&list=PLCiOXwirraUDJYD-VxmztyhsDP0QM2BHW&index=4