Type Systems of Scripting Languages

Type systems play an important role in the design of programming languages, they prevent programs from having non-correct behavior. Typically, type systems are expressed in two ways, type-checking and type inference. The type information presented on local variables and method arguments makes the difference in choosing one technique rather than the other. Type checking is a derivation type inference. In literature, the time when type-checking occurs, compile or run time, classifies them in static and dynamic. We will take a look on the type systems of some scripting languages. In this paper, we will focus on JavaScript, Python and Ruby. Most scripting languages use dynamic typing. However, we will see some recent work on providing them with static type-checking, soft-typing and type inference. We will discuss about their dynamic and lightweight properties and also about the trade off between the advantages of using static type-checking and the preservation of their properties. We will talk about the object-oriented features presented in scripting languages looking at Smalltalk [Ing78] as the first object-oriented programming language. We will look also at recent works on pluggable type systems [Bra04, ANM06] and on PyPy[RP06], a translation tool from Python to C.