feat(markdown)!: switch to split parser (#3048)

* switch to split markdown parser with separate block and inline parsers to improve performance
* add exclude_children! directive (useful for something like Injected markdown incorrectly highlights indented docstrings #2212)
* split markdown queries into block and inline ones and add the injection for inline into block grammar
* add include_dir option to parser configs (needed because the two grammars don't live in the repos root directory)

BREAKING CHANGE: downstream queries need to be adapted to new parser

2 files changed, 39 insertions, 1 deletions

diff --git a/lua/nvim-treesitter/parsers.lua b/lua/nvim-treesitter/parsers.lua
index dc02c9322..b24b6ce0c 100644
--- a/lua/nvim-treesitter/parsers.lua
+++ b/lua/nvim-treesitter/parsers.lua
@@ -528,9 +528,22 @@ list.hcl = {
 list.markdown = {
   install_info = {
     url = "https://github.com/MDeiml/tree-sitter-markdown",
+    location = "tree-sitter-markdown",
     files = { "src/parser.c", "src/scanner.cc" },
-    branch = "main",
+    branch = "split_parser",
+  },
+  maintainers = { "@MDeiml" },
+  experimental = true,
+}
+
+list.markdown_inline = {
+  install_info = {
+    url = "https://github.com/MDeiml/tree-sitter-markdown",
+    location = "tree-sitter-markdown-inline",
+    files = { "src/parser.c", "src/scanner.cc" },
+    branch = "split_parser",
   },
+  maintainers = { "@MDeiml" },
   experimental = true,
 }
 
diff --git a/lua/nvim-treesitter/query_predicates.lua b/lua/nvim-treesitter/query_predicates.lua
index be713304c..75f0efc04 100644
--- a/lua/nvim-treesitter/query_predicates.lua
+++ b/lua/nvim-treesitter/query_predicates.lua
@@ -128,3 +128,28 @@ query.add_directive("downcase!", function(match, _, bufnr, pred, metadata)
     metadata[key] = string.lower(text)
   end
 end)
+
+query.add_directive("exclude_children!", function(match, _pattern, _bufnr, pred, metadata)
+  local capture_id = pred[2]
+  local node = match[capture_id]
+  local start_row, start_col, end_row, end_col = node:range()
+  local ranges = {}
+  for i = 0, node:named_child_count() - 1 do
+    local child = node:named_child(i)
+    local child_start_row, child_start_col, child_end_row, child_end_col = child:range()
+    if child_start_row > start_row or child_start_col > start_col then
+      table.insert(ranges, {
+        start_row,
+        start_col,
+        child_start_row,
+        child_start_col,
+      })
+    end
+    start_row = child_end_row
+    start_col = child_end_col
+  end
+  if end_row > start_row or end_col > start_col then
+    table.insert(ranges, { start_row, start_col, end_row, end_col })
+  end
+  metadata.content = ranges
+end)