brian-team / brian-code-editor Goto Github PK

I think it would be best if we'd follow Python's scopes as much as possible. E.g. we should also use the following names (replacing python by brian each time):

• comment.line.number-sign.python for # comments
• punctuation.parenthesis.begin.python and ...end.python for parentheses (I think we'll need to deal with this for completion, anyway)
• keyword.operator.arithmetic.python for operators such as +, *, or ** (again, can be copy&pasted from Python probably)

On the right-hand side of an equation, some of the names will correspond to variables and some to constants, but we cannot know which is which in a TextMate grammar (this needs "semantic highlighting"). For now, I'd therefore mark all names (e.g. in the example gl, El, v, …) as constants.

Also note that we do not need special handling for names that only have one letter (currently constant.character.brian) – these should be handled in the same way as multi-letter names.

The part following the colon (:) of an equation defines a unit (e.g. volt, but possible also something like 1/second, optionally followed by one or more comma-separated flags in parentheses. E.g.

... : siemens  # no flag
... : volt (constant)  # 1 flag
... : 1/second (shared, constant)  # 2 flags

We should mark the two parts differently

In the example, there are function calls like exp(...) or exprel(...), these should be marked as meta.function-call.brian. This could probably be taken from Python's grammar (or more or less any other programming language)?

GSoC 2023 Project

Currently, highlighting/completion/verification are only done in Equations definitions. Without much additional work, we could also extend this to NeuronGroup and Synapses, when they directly define equations as their model argument.