Octave style guide

From Octave
Jump to: navigation, search

A lot of GNU Octave is written in the Octave language itself. This document details the Octave style used by the GNU Octave project.

Being part of the GNU project, Octave inherits the GNU coding standards. However, those were written with C in mind and can't always apply to Octave code.

See also the GNU Octave C++ style guide.

Formatting[edit]

Line Length[edit]

Keep the length of source lines to 79 characters or less, for maximum readability in the widest range of environments. This is inherited from the GNU Coding Standards.

Indentation[edit]

Use only spaces, and indent 2 spaces at a time.

We use spaces for indentation. Absolutely do not use tabs in your code. You should probably set your editor to emit spaces when you hit the tab key.

Whitespace[edit]

When calling functions, put spaces after commas and before the calling parentheses, like this:

x = max (sin (y+3), 2);

An exception are matrix or cell constructors:

[sin(x), cos(x)]
{sin(x), cos(x)}

Here, putting spaces after sin, cos would result in a parse error.

For indexing expressions, do not put a space after the identifier (this differentiates indexing and function calls nicely). The space after a comma is not necessary if index expressions are simple, i.e., you may write

A(:,i,j)

but

A([1:i-1;i+1:n], XI(:,2:n-1))

When constructing matrices, prefer using the comma rather than the space to distinguish between columns.

  M = [1, 2, 3
       4, 5, 6];

However, if the matrix is large or the indentation makes it clear the comma may be dropped.

  prices = [ 1.01  2.02  3.03
            44.04 55.05  6.06];

Error messages[edit]

When you encounter an error condition, call the function error (or print_usage). The error and print_usage functions do not return. It is customary to prefix the error message with the name of the function that generated it. For example:

error ("my_cool_function: input A must be a matrix");

Octave has several functions that produce error messages according to the Octave guidelines. Consider using inputParser and validateattributes.

Naming[edit]

Use lowercase names if possible. Uppercase is acceptable for variable names consisting of 1-2 letters. Do not use mixed case names. Function names must be lowercase. Function names are global, so choose them wisely.

General naming functions[edit]

Function names[edit]

For most public functions we are limited by Matlab compatibility. Use whatever name Matlab choose.

For functions that are not present in Matlab favour the use of underscores. For example, base64_decode, common_size, or compare_versions. There are exceptions to this:

Matching C functions
If the function exists elsewhere with a common name, use it. For example, dup2, waitpid, printf, argv, or getopt.
Matching similar functions
If there are similarly named functions, consider using same style. For example, fftconvn and histthresh, match the naming of fftconv2 and graythresh.

Variable names[edit]

Quoted Strings[edit]

Always use double quotes for strings and characters rather than the Matlab single quote convention. Both quote types are accepted by Octave, but double quoted strings are interpreted slightly differently (see Strings in the manual for details).

Do:

a = "Hello, world";
b = "x";
disp ("This \"string\" contains a\nnewline");

Don't:

s = 'Hello, world';
if (x(1) == 'c')
  disp ('Don''t quote one character this way, even if you''re a C programmer');
endif

There are a few edge cases where single quoted strings may be preferable, and are permitted as exceptions under this style guide.

String containing double quotes
A string that contains many double quote characters itself, where escaping all of them with backslashes becomes inconvenient, may be easier with single quotes.
String containing backslashes
A string that contains literal backslashes, in particular a regular expression pattern, where doubly escaping certain character sequences is both inconvenient and harder to read, is usually better done with single quotes.
Argument interpreted differently
A string argument to the regexp family of functions may be interpreted differently depending on whether it is a double quoted or single quoted string. Certain escape sequences are interpreted only in a single quoted string for Matlab compatibility.

ending blocks[edit]

Always use a specific end-of-block statement (like endif, endswitch) rather than the generic end.

Enclose the condition of an if, while, until, or switch statement in parentheses, as in C:

if (isvector (a))
  s = sum (a);
endif

Do not do this, however, with the iteration counter portion of a for statement. Write:

for i = 1:n
  b(i) = sum (a(:,i));
endfor

! operator[edit]

The Octave operator ! should be used for logical negation, rather than ~. The negation operator is written with a space between the operator and its target, e.g., ! A.


Comments[edit]

# or %[edit]

Always use # to write comments.

Absolutely do not use %# or mix % and # in the same file.

Block and Inline comment[edit]

Use a single # for inline comments. Use double ## for block comments.

Comments that start with a single sharp-sign, #, are used to explain the code on the same line as the comment itself. These comments should all be aligned to the same column to the right of the source code. In the Emacs mode for Octave, the M-; (@code{indent-for-comment}) command automatically inserts such a # in the right place, or aligns such a comment if it is already present. Example:

C = 2 * pi * r;    # formula for circumference of a circle

Comments that start with a double sharp-sign, ##, are stand-alone comments that occupy an entire line. These comments should be aligned to the same level of indentation as the code. Such comments usually describe the purpose of the following lines or the state of the program at that point. Example:

## Calculate area and volume of a sphere
A = 4 * pi * r^2;
V = 4/3 * pi * r^3;

Commenting out code[edit]

Do not comment code out. If the code is no longer used, remove it. We use version control, we can always bring it back.

%! for test and demo blocks[edit]

Any demos or Built-In Self Tests (BIST) using the %!demo or %!test syntax should begin two lines after the endfunction keyword. Demo blocks should be listed before test blocks.

See the section Writing tests on the Tests page.

FIXME notes[edit]

The preferred comment mark for places that may need further attention is with FIXME: comments.