Erlang (programming language)/Tutorials/Simplify: Difference between revisions

From Citizendium
Jump to navigation Jump to search
imported>Eric Evers
(New page: = Simply Numerics = Lets simplify some numerical values recursively. Humans like to read numbers in simplest form. Rather than 0+1i humans prefer i Rather than 4.000 t...)
 
imported>Tom Morris
 
(3 intermediate revisions by 2 users not shown)
Line 1: Line 1:
{{subpages}}
= Simply Numerics =
= Simply Numerics =
(auto-demotion of numerical types)


Lets simplify some numerical values recursively. Humans like to  
Lets simplify some numerical values recursively. Humans like to  
read numbers in simplest form. Rather than  
read numbers in simplest form. Rather than  
     0+1i  
     0 + 1i  
humans prefer  
humans prefer  
     i
     i
Line 13: Line 15:
the following function: simplify, which is overloaded for many
the following function: simplify, which is overloaded for many
types of arguments, integer, float, imaginary, complex and matrix.  
types of arguments, integer, float, imaginary, complex and matrix.  
We wish to remove extra zeros when possible, because they can be a distraction,
and do not add any thing to calculations. This is an example of auto-demotion
of numerical types. Many programming languages have auto-promotion of numerical
types but almost never have auto-demotion.


  -module(simple).
  -module(simple).

Latest revision as of 06:07, 8 August 2009


Simply Numerics

(auto-demotion of numerical types)

Lets simplify some numerical values recursively. Humans like to read numbers in simplest form. Rather than

    0 + 1i 

humans prefer

    i

Rather than

    4.000

the integer

    4 

looks nicer. In a matrix, we prefer the integers when possible, so we construct the following function: simplify, which is overloaded for many types of arguments, integer, float, imaginary, complex and matrix. We wish to remove extra zeros when possible, because they can be a distraction, and do not add any thing to calculations. This is an example of auto-demotion of numerical types. Many programming languages have auto-promotion of numerical types but almost never have auto-demotion.

-module(simple).
-compile(export_all). 

% Simplifies numbers by removing zeros 
 
start() -> 
	Inputs = [ 
 		3.3, 
 		3.0,
		{4,0,i},
		{3.0,i}, 
		{0,3,i}, 
		{0,3.3,i},
		[ [1.0,2.0], [3.0,4.0] ],
		[ [{0,1,i},{3.0,i}], [4,5.0] ],
		[ [1, 2, 3], [5.0, 4, 3] ]
		],
	Outputs = lists:map(fun simple:simplify/1, Inputs),
	Arrows = lists:duplicate(length(Outputs),"  ->  "),
	Inputs_and_Outputs = zipit(Inputs,Arrows,Outputs),
	Inputs_and_Outputs.

simplify(A) when trunc(A) == A ->          % 1.0 -> 1
	trunc(A);                           % make float integer
	
simplify({B,i}) ->                         % 1.0 i ->  1 i
	{simplify(B),i};                    % simplify imaginary

simplify({A,0,i}) ->                       % 2.2 + 0 i -> 2.2
       simplify(A);                        % make complex real
 	
simplify({0,B,i}) ->                       % 0 + B i -> B i 
       {simplify(B),i};                    % make complex imaginary
	
simplify({A,B,i}) ->                       % 1.0 + 2.0 i -> 1 + 2 i
       {simplify(A),simplify(B),i};        % simplify complex
 	
simplify([[A,B],[C,D]]) ->                 % [[ A.0, B.0 ]   ->  [[A,B]   
       [ [simplify(A), simplify(B)],       %  [ C.0, D.0 ]]       [C,D]] 
         [simplify(C), simplify(D)] ];     % simplify 2x2 matrix

simplify([]) -> [];                        %  [A.0, B.0, C.0 ...] -> [A,B,C...]
simplify([H|T]) ->                         % simplify vectors and matrixes   
       [simplify(H)] ++ simplify(T);       % of any size
	                                        
simplify(A) ->								 
       A.
	
zipit([],[],[]) -> [];                      % [1,2,3],[a,b,c],[do,re,me] ->	
zipit([H1|T1],[H2|T2],[H3|T3]) ->           %   [{1,a,do},{2,b,re},{3,b,me}]
       [{H1,H2,H3}] ++ zipit(T1,T2,T3).     % tripple zip

Outputs

Simplified numbers

1> c(simple).                       % compile
{ok,simple}
2> simple:start().                  % run
[{3.30000,"  ->  ",3.30000},
 {3.00000,"  ->  ",3},
 {{4,0,i},"  ->  ",4},
 {{3.00000,i},"  ->  ",{3,i}},
 {{0,3,i},"  ->  ",{3,i}},
 {{0,3.30000,i},"  ->  ",{3.30000,i}},
 {[[1.00000,2.00000],[3.00000,4.00000]],
  "  ->  ",
  [[1,2],[3,4]]},
 {[[{0,1,i},{3.00000,i}],[4,5.00000]],
  "  ->  ",
  [[{1,i},{3,i}],[4,5]]},
 {[[1,2,3],[5.00000,4,3]],"  ->  ",[[1,2,3],[5,4,3]]}]