Coding conventions

General recommendations

Syntax is chosen as much as possible from the user point of view, to reflect the concepts as directly as possible. Ideally, a Brian script should be readable by someone who doesn’t know Python or Brian, although this isn’t always possible. Function, class and keyword argument names should be explicit rather than abbreviated and consistent across Brian. See Romain’s paper On the design of script languages for neural simulators for a discussion.

Code style

We use the PEP-8 coding conventions for our code, and use the black formatting tool to enforce a consistent code style. To make sure your code is formatted in the same way, you can either integrate black with your editor/IDE, or install the pre-commit tool (pre-commit documentation), and install it as a “git hook” with pre-commit install. This will automatically run black (and in the future, additional linting tools) before each commit. In case that run changes the code formatting, it will reformat the relevant files, and you will have to git add these changes before doing the commit. For pull requests, this check will also be run automatically on the GitHub CI infrastructure.

Note

In rare cases like manually aligned tables of related values, you can use fmt: skip (for single lines), or fmt: off and fmt: on (for code blocks), to exclude code from black’s formatting. Also note that the code formatting is only enforced for files in the brian2 package itself, code examples in the documentation or examples and tutorials do not have to follow black’s style.

The code style includes the following conventions in particular:

  • Use 4 spaces instead of tabs per indentation level

  • Use spaces after commas and around the following binary operators: assignment (=), augmented assignment (+=, -= etc.), comparisons (==, <, >, !=, <>, <=, >=, in, not in, is, is not), Booleans (and, or, not).

  • Do not use spaces around the equals sign in keyword arguments or when specifying default values. Neither put spaces immediately inside parentheses, brackets or braces, immediately before the open parenthesis that starts the argument list of a function call, or immediately before the open parenthesis that starts an indexing or slicing.

  • Avoid using a backslash for continuing lines whenever possible, instead use Python’s implicit line joining inside parentheses, brackets and braces.

Imports

Imports should be on different lines (e.g. do not use import sys, os) and should be grouped in the following order, using blank lines between each group:

  1. standard library imports

  2. third-party library imports (e.g. numpy, scipy, sympy, …)

  3. brian imports

This rule is enforced by using the isort tool, which is integrated with pre-commit in the same way as black, described above.

Note

In rare cases, where logical grouping makes more sense that isort’s ordering, or when the order of imports matters, you can skip sorting imports in a file by including the comment # isort:skip_file.

Additional rules for imports:

  • Use absolute imports for everything outside of “your” package, e.g. if you are working in brian2.equations, import functions from the stringtools modules via from brian2.utils.stringtools import .... Use the full path when importing, e.g. do from brian2.units.fundamentalunits import seconds instead of from brian2 import seconds.

  • Use “new-style” relative imports for everything in “your” package, e.g. in brian2.codegen.functions.py import the Function class as from .specifiers import Function.

  • Do not use wildcard imports (from brian2 import *), instead import only the identifiers you need, e.g. from brian2 import NeuronGroup, Synapses. For packages like numpy that are used a lot, use import numpy as np. But note that the user should still be able to do something like from brian2 import * (and this style can also be freely used in examples and tests, for example). Modules always have to use the __all__ mechanism to specify what is being made available with a wildcard import. As an exception from this rule, the main brian2/__init__.py may use wildcard imports.

String formatting

In general, we use Python f-strings instead of the .format method or the % operator to format strings. For example, rather use:

raise KeyError(f"Unknown variable '{var}'")  # ✔

instead of:

raise KeyError("Unknown variable '{}'".format(var))  #  ❌
raise KeyError("Unknown variable %s" % var)  #  ❌

There are some corner cases where it still makes sense to use either of these, though. The format method can be useful when processing several strings instead of single literals:

formatted = []
for s in strings:
    formatted.append(s.format(**values))

The % operator, or string concatenation, can be used when dealing with strings that contain curly braces, which would become difficult to read as an f-string:

latex_code = r'\begin{equation}%s\end{equation}' % equation  # OK
latex_code = r'\begin{equation}' + equation + r'\end{equation}' # OK

Python does not make a difference between single quotation marks and double quotation marks. For consistency, we use black’s style that double quotes (i.e. "...") are used everywhere, except when this would lead to escaping of double quotation marks within the string itself (e.g. for include = f'#include "{header_file}"'). When you need to display text to the user (e.g. in exception messages), use single quotation marks to highlight words to avoid this situation (e.g. "The 'item' value should not be 0.").

Commits only changing the style

We have sometimes made big commits updating the style in our code, which can make using tools like git blame more difficult, since many lines are affected by such commits. We add the references to such commits to a file .git-blame-ignore-revs in the main directory, and you can tell git blame to ignore these commits with:

git config blame.ignoreRevsFile .git-blame-ignore-revs