Taxonomy of Programming Languages CSCI 432 Computer Science

Taxonomy of Programming Languages CSCI 432 Computer Science Theory

Misc Notes We have been studying languages all semester. § Languages defined by Finite Automata § Regular Languages § etc

Importance of Types The syntax of a language is governed by the constraints that define its data types. For example: ◦ everything in Java is a class ◦ most scripting languages use weak type checking

High Level v. Low Level Language: § eg, Assembly § every instruction is done by a piece of hardware § faster than compiled or interpreted High Level Language: § easier to code, manage, test, etc.

Compiled v. Interpreted Advantages of Compiled § the machine code in the. exe file runs fast § compiler probably optimizes your code Advantages of Interpreted § portable

Imperative Programming Languages English definition of "Imperative" - essential, authoritative command statements that change the machine's state § this is the majority of programming languages § some imperative languages have declarative properties § two subtypes § procedural § object-oriented

Components of Imperative Langs v Sequence o o how to control program flow from one statement to the next eg blocks of code, procedures, functions, recursion v Selection o o making decisions eg if statements and case statements v Iteration o o repeating instructions eg for loops and while loops

Declarative Programming Languages specify what is to be done, not how to do it § does not specify control flow or order of operations § examples § HTML § SQL

Generations § 2 nd Generation Languages (2 GL) § Assembly § 3 rd Generation Languages (3 GL) § Higher level than 2 GL, like C and C++ § some 3 GLs, like Python, have libraries of routines that give them 4 GL characteristics

Generations § 4 th Generation Language (4 GL) § Operating on large collections of data § for example: SQL, report writers § 5 th Generation Language (5 GL) § Problem solving based on constraints instead of algorithms § Artificial Intelligence programming § area of research for past 20 years, without significant progress § given a set of constraints, finding an algorithm to solve the problem is usually very difficult, thus usually requires a programmer (3 GL or 4 GL)

Domain Specific Languages as opposed to a General Purpose Language Examples: § HTML § statistical modeling languages § Open. GL shading language § software engineering - eg requirements specifications § shells - ls, ps, grep, sort, head, wc, …

Parallel Languages non-parallel a[]=b[]+c[] : for (i=0; i<size; i++) a[i] = b[i] + c[i]; Compiler and OS: forall (i=0; i<size; i++) a[i] = b[i] + c[i]; Programmer: function add (start, stop) { for (i=start; i<stop; i++) a[i] = b[i] + c[i]; } for (processor=0; processor<8; processor++) clone (processor, add(processor*size/8, (processor+1)*size/8));

FORTRAN and COBOL Example FORTRAN Code PROGRAM DEMO PRINT *, 'Enter number? ' READ *, X IF (X. LE. 0) THEN PRINT *, 'That is negative. ' STOP END IF PRINT *, 'That is positive. ' STOP END

LISP q List Processing q Everything is a list ((Jimmy Carter), (Ronald Reagan), (George Bush), ……. ) >(defun factorial (N) (if (= N 1) 1 (* N (factorial (- N 1))))) >(eval factorial(5)) (120)

CLIPS q known facts and rules that work on facts q logic programming language (assert (person (name Steve Dannelly) (occupation Professor) (home Rock Hill))) (facts) (defrule fire-emergency (type fire)) => (assert (response (action sprinkler-on))) )