nvim-treesitter/queries/python/injections.scm

((call
  function: (attribute
	  object: (identifier) @_re)
  arguments: (argument_list (string) @regex))
 (#eq? @_re "re")
 (#lua-match? @regex "^r.*"))

(comment) @comment
Python highlights: Add regex injections: re.match(r"...") 2020-12-28 18:20:34 +01:00			`((call`
			`function: (attribute`
			`object: (identifier) @_re)`
			`arguments: (argument_list (string) @regex))`
			`(#eq? @_re "re")`
Prefer lua-match over match as string.find is much quicker than vim.regex:match* 2021-11-18 09:51:03 +00:00			`(#lua-match? @regex "^r.*"))`
Python: inject rst in docstrings (#917) Since hasn't been much discussion around https://github.com/nvim-treesitter/nvim-treesitter/issues/806. I'm just porting the injection queries. I've been using this for a while now. Things I've noticed: - Due that rst uses indentation for its syntax, everything is treated as an block quote (but it looks good). This can be solved by having a predicate like `#dedent!`. - Looks like there is a bug in how the injected content is extracted ``` def foo(): """Foo bar""" ``` That would be parsed as a section title for some reason, but it's a paragraph. In rst it would be a title if the content was: ``` """ Foo bar """ ``` If the content is ``` """Foo bar""" ``` That's just a paragraph. I'll try to debug that from the neovim side next week or so. 2021-02-05 10:01:50 -05:00
Revert "Comment: use `@combined` to create just one tree per buffer (#1252)" This reverts commit 4a7a713c527f57c355bc817f16fc6d955b8cce9b. 2021-04-30 13:22:12 -05:00			`(comment) @comment`