Description
Is your request related to a specific problem you're having?
See #3261. Generally speaking meta currently is used in many places for "this might be important, give it special emphasis" - it's a bit of a "catch all"... which is counter to our own policy of semantically tagging content vs visually tagging it. Related: #2500 and the higher fidelity goals.
Some examples of how meta is currently used across grammars:
julia>- REPL promptsbob@localhost>- shell prompts<!DOCTYPE html>HTML doctypes#include "blah.h"- C preprocessor directives
The solution you'd prefer / feature you'd like to see added...
I haven't put a ton of thought into this yet, so I'll just put ideas into the following section.
Any alternative solutions you considered...
- Better scope(s) could be found for some of these
- the
metanamespace could be expanded - or both
Examples of expanding the namespace:
meta.prompt- shell and repo prompts (enh) add meta.prompt scope for REPL prompts #3490meta.preprecessor- preprocessor directives
Not sure about HTML doctype... and honestly we'd need to take a look at ALL the grammars where we use meta and come up with an exhaustive list. If we're going to make a sweeping change here I want to do it eyes wide open and get it right the first time if at all possible.
Additional context...
Long-term I really like how TextMate grammar scopes use meta for "block scopes" that by themselves are typically unhighlighted but rather provide additional context for themes to more properly highlighting things WITHIN those scope. It seems nice, but I'm not sure how we get there from here without either coming up with another name entirely... or touching all the uses of meta at once and moving them to another top-level scope.
I mention this here because properly solving this likely requires touching/renaming meta everywhere to begin with so it's worth taking the time to think about if we're preparing to embark on a PR that is going to need to touch meta everywhere anyways.