Numeric values
Prolog supports traditional integer and floating point numeric expressions. It provides a variety of functions to test if a variable or atom has a specific numeric subtype, e.g. number(X), integer(X), float(X), etc.

Functions and operators
Numerous conventional binary and unary operators are supported:

Functions/operators that return numeric values
These can be used within a larger math expression, and actually return a value that can be used with "is", e.g.
X is (-Y * sqrt(Z))

Note: when using "is", the item on the left must be an uninstantiated variable (i.e. a variable with no value yet) and the item on the right must be a value or an expression that can be evaluated to a value (i.e. no uninstantiated variables on the right hand side).
This means that expressions like X is (X + 1) cannot possibly work, since if X has a value it is invalid on the left side, and if it doesn't have a value then it is invalid on the right side.

• unary operators (e.g. Y is -X.) include: + -
• binary operators (e.g. Z is X-Y.) include: + - * / // ^ >> mod rem xor
• math functions (e.g. F is abs(X).) include: abs(X), max(X,Y), sign(X), min(X,Y), random(X), round(X), truncate(X), ceiling(X), floor(X), sqrt(X), sin(X), cos(X), tan(X), asin(X), acos(X), atan(X), ln(X), log(X), log10(X), exp(X)

  Comparison and type-checking functions

• =:= and =/= for numeric equality/inequality, e.g. X =:= Y.
• < =< > >= for other numeric comparisons, e.g. X >= Y.
• /\ for bitwise and, \/ for bitwise or, \ for complement, xor for exclusive-or, << and >> for shift left/right, e.g. X is Y << 2.
• type checking functions include: integer(X), float(X), number(X)

  Functions that take both the data and result as parameters
  Many of these are variants on the math functions discussed above, e.g. sqrt(Num,Result), floor(Num,Result), etc.
  Note that the return value of these is a typical prolog true/false.