Define non standard data types

Answer 1

In programming, non-standard data types can be anything; a model for a house, a person, a bank account, a galaxy. They all have in common that they are created through composition of basic built-in data types such as integers, strings or floating point data types.

To define non-standard data types, programming languages typically provide three to four language tools:

Arrays can combine multiple entities of the same basic or non-standard type into as larger new type. When an array is viewed as its own data type rather than a simple accumulation of variables, the whole is more than the sum of its parts: in the C Programming language, for example, an array of bytes defines a string or characters.

Structures can combine multiple entities of the same or different data types into one record, from which all record members are accessible at the same time. For example, a student's record might hold the name and date of birth for a single student.

Unions can combine multiple entities of the same or different data types into one record, from which only one record member is accessible at any one time. For example, a union containing the key characteristic of a geometric shape might contain the number of edges of a polygon as an integer number, or a sphere's radius as a floating point variable. The programmer needs an additional method to determine whether this union ought to be interpreted as the edge-count or the radius.

Object-oriented programming languages also support classes. In simple terms, classes are structures which also contain functions, not only data. (Note this is a radical simplification. Classes and object-oriented programming have more to offer than glorified data structures.)