High- and low-level languages and their translators

A programming language that is easier for the processor to interpret but harder for humans to understand. Two low-level languages: • machine language/object code • assembly language

• lowest level of language • consists of binary and/or hexadecimal • raw instructions that the CPU carries out (processor reads machine code directly, no need to translate it) • each CPU family has a different set of machine code

• uses abbreviations or mnemonics for each operation code • easier to read and write than machine language • must be translated into machine code by an assembler before it can be run by a processor

• translation program that translates processor-specific assembly code (source code) into machine code (object code) for a specific processor • translates an entire program but does not run it • it is rapid and reliable

• complete freedom of choice of instructions • direct control over the processor’s communication with its input, output and storage devices • efficient programs can be written that: • fir into limited storage space • require limited RAM • run rapidly

• lack of relation to a human language makes it difficult for programmers to read • machine-oriented nature makes it hard for a programmer to learn, as each operation code must be memorised or looked up • processor-specific nature of the instruction set means that programs are not transferrable or portable between families of processor • programs are relatively hard to test, debug and maintain

A programming language that looks more similar to a human language or mathematical notation and is therefore easier for programmers to use. However, a program must be used to translate the language for the processor. They are intended to help programmers solve problems rather than to micro-manage the computer’s hardware.

• a program that translates all the high-level source code by compilation • compilation must be completed before the program can be run, i.e. the entire program must be translated before the processor can execute the instructions • if there are errors in the source code, the compiler produces an error report, the programmer corrects the errors and compiles the whole program again • therefore, the debugging of a large program before it can be run can be a slow, iterative process

• program that translates high-level language statements and executes them on a specific processor and operating system • does not create machine code • analyses one source code statement at a time and runs appropiate subroutines to execute it • if the interpreter finds an error in the source code, it produces an error report and stops the execution • the process of translation slow the execution, but debugging a large program is faster as the program does not need to be re-compiled after every edit

• compiler produces object code; interpreter doesn’t produce object code • compiler translates whole program in one go; interpreter translates and executes line at a time • compiler produces list of all errors; interpreter produces error message each time an error is encountered • compiler produces “stand alone code”; interpreter doesnt produce “stand alone code” • compilation process is slow but resultant code runs very quickly; interpreted code runs slowly