Translators

Made with Ada Comp. Sci. and P.M.T.

Translator

Piece of software that converts a computer program into a form that can be executed by the processor

Types of translator

• Interpreter

• Compiler

• Assembler

Interpreters and their translation process

• Translates and executess each line of source code one by one. For each line:

  • Checks syntax, and if theres an error it is reported and the program halts

  • If no errors found, the line of code is converted to its machine code equivelent and is executed

  • If a runtime error occurs, the program crashes

Advantages of interpreters

• Line-by-line approach allows quick identification and correction of each errors as it occurs, without having to translate the whole program each time.

  • This makes interpreters very useful during developement of a program

• More portable - the same source code (or byte code) can be executed on a range of platforms as long as the right interpreter is available.

  • This makes them useful when the platform on which a program will run is unkown (e.g. for websites using JS)

Disadvantages of interpreters

• Every time the program is run it must be translated again. This means it will run more slowly than an executable file produced by compilation.

• Requires interpreter to be installed on the computer for the program to run

Compilers

• Translates code all at once

• Produce an executable file, but does not execute the code

• Executable file produced is machine code suitable for a specific instruction set architecture

• A list of errors reported at end of compilation process (if there are any)

Advantages of compilers

• Once successfully compiled, the executable file wont need to be tranlated again (unless the code is changed)

  • Code can be run without a translator being present

  • Executes more quickly than interpreted code once compiled

• Provides protection of source code - useful for closed source software where you don’t want the source code to be publicly accessible

Disadvantages of compilers

• Errors are reported at the end of the compilation process

  • Any erros require the program to be corrected and recompiled until no errors are present

• Produces machine code for a specific instruction set architecture

Stages of compilation (first to last)

• Lexical analysis

• Syntax analysis

• Code generation

• Code optimisation

What happens in lexical analysis

• Non-program elements (e.g. comments & whitespace) are removed

• Reserved words / keywords are tokenized

• Identifier names are added to a symbol table - this allows the compiler to keep track of all the identifiers that have been declared in the program

What happens in syntax analysis

• Checks code tokens follow the grammatical rules of the language

• An abstract syntax tree is created

  • Can be used to check no code tokens are missing

• Checks code tokens are in correct order

• If a syntax error is found it is added to a list of erros and will be reported at the end of compilation. Error diagnostics will be given with the list.

• Adds detail to the symbol table - e.g. data type or scope of variable

• Semantic analysis occurs - highlights certain types of logic errors in the program (e.g. list index is out of range)

Abstract syntax tree

• Maps the structure of the program

• Used to check syntax is correct

• Dependant on syntax so is different for each programming language

Code generation

The abstract syntax tree is used to create object code

Object code

• Binary represeentation of source code

• Executable file before linked libraries have been included

Code optimisation

• Occurs throughout compilation process, but especially in the code generation stage

• May identify redundant or repeated code

• Can remove and rearrange code as necessary

Purpose of code optimisation

• To make the program run faster/ code is more efficient

• To make the program use fewer resources/less memory

Assemblers and their translation process

• Translate assembly code into executable code

• Assembly code is platform specifc and as such the assemblers are as well

• Code is translated on an almost one to one basis - each line of assembly code is equivelent to almost one line of machine code

• Comments in the assembly program are removed

• Symbolic references are replaced with actual addresses

Linkers

Piece of code the links programs to software libraries to create a single executable file

Static linker

• Combines code and libraries directly into one file. This increases the size of the file

• External updates to libraries wont affect the program - this allows specific versions of a library to be used

Dynamic linker

• Adds the addresses to libraries (and modules) where they are used in the executable file

• Keeps the file size small and allows external updates to libraries to feed through to the main file

Why are linkers needed?

• The user running the program will not necessarily have the library installed on their machine

• Therefore the relevant code needs to be included within the final executable file

• It is the job of the linker to combine this code.

Loaders

• Provided by the operating system

• When the program is executed, the loader:

  • Copies the executable code from secondary storage into main memory

  • Loads the required software libraries

(Software) Libraries

• Sections of code written by other authors containing useful routines

• Are often precompiled

Advantages of using libraries

• Saves time as there is no need to rewrite code

• Allows you to use the expertise of others to…

  • … complete tasks that require specialist knowledge

  • … abstract away complexity

• If it’s from a reliable source, it has usually been thoroughly tested and therefore is efficient and reliable. This:

  • Makes debugging easier

  • Saves time

Disadvantages of libraries

• May (significantly) increase size of compiled file if the library contains many routines that aren't being used.

• Not written by the programmer so introduces uncertainty.

  • Programmer needs to spend time familiarising themselves with it

  • May require further testing

What is bytecode / intermediate code

• Source code that has been compiled into low-level code designed to be run on a virtual machine (the interpreter).

• It is partially compiled code

How is byte code / intermdediate code used

• The source code is compiled into low-level ____

• This can then be executed by any machine that has the suitable virtual machine (interpreter) installed on it

• As such the ____ can be distributed as a ready to run program

Advantages of bytecode / intermediate code

• Speeds up translation process

• Executable on a wide range of platforms as long as suitble virtual machine (interpreter) is installed on the machine

Disadvantages of bytecode / intermediate code

• Slower execution than fully compiled code

• Less protection of source code than a fully compiled language

What are the differences between a compiler and an interpreter?

• Compiler translates code all at once, whereas interpreter translates code line by line

• Compiler produces executable file for reuse so doesn’t need to be translated everytime it is run. An interpreter needs to re-translate every time the program is run.

• Compiler lists all errors at the end of compilation. Interpreter stops at the first error.

• Compiled programs have the source code hidden. Interpreted programs have the source code visible.