| Commit message (Collapse) | Author | Age | Files | Lines |
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
Distinguishing `hint` and `info` is too confusing; it's enough to have
`error`, `warning`, `todo`, and `note`.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Problem: Sharing highlight queries with upstream tree-sitter and
Helix is difficult.
Solution: Where reasonable, use capture names in tree-sitter's standard
list or Helix's Atom-style hierarchy.
Specifically:
* tree-sitter "standard capture names"
(https://github.com/tree-sitter/tree-sitter/blob/3f44b896852eb7daaa6df4fb778c9bb52c70c815/highlight/src/lib.rs#L20-L72):
- `@parameter` -> `@variable.parameter`
- `@field` -> `@variable.member`
- `@namespace` -> `@module`
- `@float` -> `@number.float`
- `@symbol` -> `@string.special.symbol`
- `@string.regex` -> `@string.regexp`
- `@text.*` -> `@markup.*` (`strong`, `italic`, `link`, `strikethrough`; with exceptions; see below)
- `@text.title` -> `@markup.heading`
- `@text.literal` -> `@markup.raw`
- `@text.reference` -> `@markup.link`
- `@text.uri` -> `@markup.link.url` (in markup links)
- `@string.special` -> `@markup.link.label` (non-url links)
- `@punctuation.special` -> `@markup.list` (markdown lists only; move subitems from `@text.todo`)
* Helix captures
(https://docs.helix-editor.com/master/themes.html#syntax-highlighting):
- `@method` -> `@function.method`
- `@method.call` -> `@function.method.call`
- `@text.{todo,warning,note,danger}` -> `@comment.{error,warning,hint,info,todo}`
- `@text.diff.{add,delete,}` -> `@diff.{plus,minus,delta}`
- `@text.uri` -> `@string.special.url` (outside markup)
- `@preproc` -> `@keyword.directive`
- `@define` -> `@keyword.directive`(`.define`?)
- `@storageclass` -> `@keyword.storage`
- `@conditional` -> `@keyword.conditional`
- `@debug` -> `@keyword.debug`
- `@exception` -> `@keyword.exception`
- `@include` -> `@keyword.import`
- `@repeat` -> `@keyword.repeat`
* cleanup
- remove some redundant `@conceal` (but still allow it for conceal-only patterns)
- remove obsolete `@error` (syntax linting is out of scope for this repo)
- sort, cleanup capture list in `CONTRIBUTING.md`
|
| |
|
|
|
|
|
|
|
| |
Problem:
Comment highlight query produces too many @spell captures.
Solution:
Remove the query. This is fine because comment parser is only used in
injection and the parent language has @spell for comment.
|
| | |
|
| | |
|
| |
|
| |
see also: https://github.com/stsewd/tree-sitter-comment/issues/22
|
| |
|
|
| |
https and https links are recognized now
ref https://github.com/stsewd/tree-sitter-comment/pull/21
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
There are a couple cases where the comment highlights
will match things that aren't necessary a tag.
For example in Rust, when commenting a block that deals with
generics (generics are declared as `T: Type`),
this may be distracting.
So, instead of highlighting all tags, we can restrict
to highlighting the explicit keywords.
Ref https://github.com/stsewd/tree-sitter-comment/issues/14
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
as string.find is much quicker than vim.regex:match*
|
| | |
|
| | |
|
| |
|
|
| |
As requested in https://github.com/nvim-treesitter/nvim-treesitter/issues/236#issuecomment-797878646
|
|
|
Closes #236
|