An expression has many characteristics that determine what sort of operations can be performed.

• Scalar: A scalar type is a simple-type value, ranging from 1 to 8 bytes in size, that can be added, subtracted, multiplied, etc.
• Structured: A structure type is a reference to a memory location (but not strictly a pointer to this memory location) that contains a non-simple type data structure.  A structure’s size can be any positive number of bytes.
• Pointer: A pointer type is a 16-byte portion of data that refers to existing data.
• Variable Name: The name of a global variable, member variable, parameter variable, or local variable.  In many cases, it is not possible to distinguish any additional characteristics about a variable until the expression is further evaluated against other expressions.  For example, the “sizeof” operator or structure member dereference operator respect the current variable scope when evaluating the expression.
• Variable Type: A simple type name (void, char, short, long, longlong, float, or double) or the name of a data structure.  Variable type is only used in special cases, such as “sizeof,” “new,” “newchild,” and in typecasts.
• Integer: The scalar types char, short, long, and longlong are integer types.
• Floating Point: The scalar types float and double are floating-point types.
• Pointer-To-Void: A pointer with no certainty as to the pointed-to type.
• Pointer-To-Scalar: A pointer that points to a scalar type.
• Pointer-To-Structure: A pointer that points to a structured type.
• Signed/Unsigned Integer: An integer scalar expression can be either signed or unsigned.  In some contexts, such as addition, an integer is assumed to be either signed or unsigned regardless of the sign associated with the expression.  In other contexts, such as multiplication, the sign of the expression determines how the expression must be evaluated.
• Constant Value: A constant value is a scalar, structure, or pointer value that represents transitionary data that can be read from, but cannot be written to.
• L-Value: An L-value is a scalar, structure, or pointer value that can be both read from and written to (that is, it can serve as a valid left-hand-side value of an assignment operation).

  See also:  [Expression characteristics] [Identifier interpretations] [Built-in functions] [Special load considerations]
[Additional rules for expressions] [Final expression type result] [L-value status gain and loss]
[Overly complex expressions] [Limitations on pointer usage] [Structure member dereference:  bit scan blocks]