FP (programming language)

From Wikipedia, the free encyclopedia
  (Redirected from FFP (programming language))
Jump to: navigation, search
FP
Paradigm function-level
Designed by John Backus
First appeared 1977
Dialects
FP84
Influenced by
APL[1]
Influenced
FL, Haskell, J

FP (short for Function Programming) is a programming language created by John Backus to support the function-level programming[2] paradigm. This allows eliminating named variables. The language was introduced in Backus's 1977 Turing Award lecture, "Can Programming Be Liberated from the von Neumann Style?", subtitled "a functional style and its algebra of programs."

Overview[edit]

The values that FP programs map into one another comprise a set which is closed under sequence formation:

if x1,...,xn are values, then the sequencex1,...,xn〉 is also a value

These values can be built from any set of atoms: booleans, integers, reals, characters, etc.:

boolean   : {T, F}
integer   : {0,1,2,...,∞}
character : {'a','b','c',...}
symbol    : {x,y,...}

is the undefined value, or bottom. Sequences are bottom-preserving:

x1,...,,...,xn〉  =  

FP programs are functions f that each map a single value x into another:

f:x represents the value that results from applying the function f 
    to the value x

Functions are either primitive (i.e., provided with the FP environment) or are built from the primitives by program-forming operations (also called functionals).

An example of primitive function is constant, which transforms a value x into the constant-valued function . Functions are strict:

f: = 

Another example of a primitive function is the selector function family, denoted by 1,2,... where:

i:〈x1,...,xn〉  =  xi  if  1 ≤ i ≤ n
              =  ⊥   otherwise

Functionals[edit]

In contrast to primitive functions, functionals operate on other functions. For example, some functions have a unit value, such as 0 for addition and 1 for multiplication. The functional unit produces such a value when applied to a function f that has one:

unit +   =  0
unit ×   =  1
unit foo =  ⊥

These are the core functionals of FP:

composition  fg        where    fg:x = f:(g:x)
construction [f1,...fn] where   [f1,...fn]:x =  〈f1:x,...,fn:x
condition (hf;g)    where   (hf;g):x   =  f:x   if   h:x  =  T
                                             =  g:x   if   h:x  =  F
                                             =      otherwise
apply-to-all  αf       where   αf:〈x1,...,xn〉  = 〈f:x1,...,f:xn
insert-right  /f       where   /f:〈x〉             =  x
                       and     /f:〈x1,x2,...,xn〉  =  f:〈x1,/f:〈x2,...,xn〉〉
                       and     /f:〈 〉             =  unit f
insert-left  \f       where   \f:〈x〉             =  x
                      and     \f:〈x1,x2,...,xn〉  =  f:〈\f:〈x1,...,xn-1〉,xn〉
                      and     \f:〈 〉             =  unit f

Equational functions[edit]

In addition to being constructed from primitives by functionals, a function may be defined recursively by an equation, the simplest kind being:

fEf

where Ef is an expression built from primitives, other defined functions, and the function symbol f itself, using functionals.

See also[edit]

  • FL - Backus' FP successor

References[edit]

  1. ^ The Conception, Evolution, and Application of Functional Programming Languages Paul Hudak, 1989
  2. ^ Backus, J. (1978). "Can programming be liberated from the von Neumann style?: A functional style and its algebra of programs". Communications of the ACM. 21 (8): 613. doi:10.1145/359576.359579.  Backus' 1977 Turing Award lecture