Abstract: A method for inferring the shape and dimension of arrays for high-level, array-based languages such as MATLAB is presented. The method uses the algebraic properties that underlie MATLAB's shape semantics and infers the shape that the program expression assumes. In one embodiment, a shape-tuple of the result of a program expression is inferred by creating a shape-tuple expression comprising the shape-tuples of the operands and the shape-tuple operator.

Abstract: An improved method and process for array shape inferencing for high-level array-based languages such as MATLAB and APL. The process is based on a framework that algebraically describes the shape of an expression at compile time. The method leverages on algebraic properties that underlie MATLAB's shape semantics and exactly captures the shape that the expression assumes at run time. Other highlights of this method are its generality and the uniformity of its approach. Compared with the traditional shadow variable scheme, the algebraic view permits powerful shape-related assertions and optimizations not possible in the conventional approach.