intbitcopy and uintbitcopy

Exact bit copy of variable into signed int__ or unsigned uint__ of same element size

Description

Two c-mex functions are supplied. intbitcopy copies a variable into an integer equivalent of the same data element size using an exact bit copy. E.g., doubles are copied to int64, singles are copied into int32, char is copied into int16, etc. Similar to typecast except the result is based on the data element size rather than fixed by input selection. intbitcopy will handle any number of input arguments as long as there are matching output arguments to store the results. For structures and cells, intbitcopy will convert each individual component or cell, keeping the overall structure of the variable the same (same field names, same cell layout, etc.). For non-convertible types (e.g., vpa, function handles, etc.), intbitcopy will return a deep copy. The 2nd mex function uintbitcopy behaves exactly the same as intbitcopy except it returns unsigned integer types.

• MATLAB release: MATLAB 7.2 (R2006a)
• Other requirements: Any C compiler, such as the built in lcc compiler.