GNU Octave

Implementation of programming language MATLAB

GNU Octave (named after the former professor of lead developer) is a system for numerical computations which uses a high-level programming language compatible with MATLAB.

Development of Octave started in 1988; initially it was intended for use in chemical reactor design course. In 1992 it became a separate software product, and started to be developed as such. In 1997 it became GNU Octave, i.e., it joined GNU Project. Since then it is distributed under GNU GPL license.

GNU Octave is written in C++ with STL. The scripts in this language are interpreted.

Octave syntax repeats the one of MATLAB in most cases, but extends it with some new possibilities:

  • strings can be both single-quoted and double-quoted (MATLAB allows only single quotes).
  • MATLAB allows only % and %{ ... %} comments, Octave adds # и #{ ... #}.
  • Octave adds C-like operators += -= *= /= \= ^= .*= ./= .\= .^= ! != ++ --.
  • functions can be defined both in scripts and in the interpreter itself (MATLAB requires that functions are defined in special .m files).
  • Octave allows and recommends to use specific block ends: endif, endfor, endwhile etc.
  • Octave’s rules of lines continuations are less strict: they require explicit continuations only when the lines are ambiguous. Both \ and ... can be used as continuations.
  • Octave has not only try-catch construct but also unwind_protect inspired by Lisp. This allows to create try-catch-finally structures for error handling.
  • Octave allows indexing of temporary variables. For example, x = 1:10; y = sin(x)(1:3) is a valid sequence of commands in Octave, while MATLAB would require a temporary variable to store sin(x).

Links:

Examples:

Hello, World!:

Example for versions GNU Octave 3.2.3

The first function is identical to the C one. The second one is Octave-specific. 

printf("Hello, World!\n");

disp("Hello, World!");

Factorial:

Example for versions GNU Octave 3.2.3

This example uses built-in function factorial. Note that at this scale of magnitude the results are exact, but in general case Octave is not meant for arbitrary-precision computations, so huge values of factorial will be calculated approximately. 

for i = 0 : 16
  printf("%d! = %d\n", i, factorial(i));
endfor

Factorial:

Example for versions GNU Octave 3.2.3

This example uses iterative factorial definition. Semicolons at the end of lines suppress the automated output of the calculated values (in this case of fact) in interactive mode, so that the formatted output doesn’t get littered. 

fact = 1;
for i = 0 : 16
  printf("%d! = %d\n", i, fact);
  fact *= i+1;
endfor

Factorial:

Example for versions GNU Octave 3.2.3

This example uses recursive factorial definition. 

function f = fact(n)
  if (n <= 1)
    f = 1;
  else
    f = n * fact(n - 1);
  endif
endfunction

for i = 0 : 16
  printf("%d! = %d\n", i, fact(i));
endfor

Fibonacci numbers:

Example for versions GNU Octave 3.2.3

This example uses recursive definition of Fibonacci numbers. 

function f = fib(n)
  if (n <= 1)
    f = n;
  else
    f = fib(n - 1) + fib(n - 2);
  endif
endfunction

for i = 1 : 16
  printf("%d, ", fib(i));
endfor
disp("...");

Quadratic equation:

Example for versions GNU Octave 3.2.3

Octave is suited for numeric computations, so it has built-in methods of solving typical problems, including finding the roots of a polynomial. To do this one can call roots function for a vector of polynomial coefficients listed in order of descending powers (so the coefficient at the highest power is listed first). 

roots([2 -3 1])