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

7 files changed, 121 insertions, 61 deletions

diff --git a/lockfile.json b/lockfile.json
index c92ed250a..33e468610 100644
--- a/lockfile.json
+++ b/lockfile.json
@@ -192,7 +192,10 @@
     "revision": "a4b9187417d6be349ee5fd4b6e77b4172c6827dd"
   },
   "markdown": {
-    "revision": "be3e08acfd85bd87d85f41fde74fdcec25f76dbe"
+    "revision": "acb097808683eea7c6d6d469644a275b9f6a64f5"
+  },
+  "markdown_inline": {
+    "revision": "acb097808683eea7c6d6d469644a275b9f6a64f5"
   },
   "ninja": {
     "revision": "0a95cfdc0745b6ae82f60d3a339b37f19b7b9267"
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)
diff --git a/queries/markdown/highlights.scm b/queries/markdown/highlights.scm
index 9d6959b22..4cc81f9e7 100644
--- a/queries/markdown/highlights.scm
+++ b/queries/markdown/highlights.scm
@@ -1,6 +1,6 @@
-;; From MDeiml/tree-sitter-markdown
-(atx_heading (heading_content) @text.title)
-(setext_heading (heading_content) @text.title)
+;From MDeiml/tree-sitter-markdown
+(atx_heading (inline) @text.title)
+(setext_heading (paragraph) @text.title)
 
 [
   (atx_h1_marker)
@@ -14,33 +14,23 @@
 ] @punctuation.special
 
 [
-  (code_span)
   (link_title)
   (indented_code_block)
   (fenced_code_block)
 ] @text.literal
 
 [
-  (emphasis_delimiter)
-  (code_span_delimiter)
   (fenced_code_block_delimiter)
 ] @punctuation.delimiter
 
 (code_fence_content) @none
 
-(emphasis) @text.emphasis
-
-(strong_emphasis) @text.strong
-
 [
   (link_destination)
-  (uri_autolink)
 ] @text.uri
 
 [
   (link_label)
-  (link_text)
-  (image_description)
 ] @text.reference
 
 [
@@ -52,56 +42,17 @@
   (thematic_break)
 ] @punctuation.special
 
-(block_quote_marker) @punctuation.special
+[
+  (block_continuation)
+  (block_quote_marker)
+] @punctuation.special
 
 [
   (backslash_escape)
-  (hard_line_break)
 ] @string.escape
 
-(image "!" @punctuation.delimiter)
-(image "[" @punctuation.delimiter)
-(image "]" @punctuation.delimiter)
-(image "(" @punctuation.delimiter)
-; (image ")" @punctuation.delimiter)
-
-(inline_link "[" @punctuation.delimiter)
-(inline_link "]" @punctuation.delimiter)
-(inline_link "(" @punctuation.delimiter)
-; (inline_link ")" @punctuation.delimiter)
-
-(shortcut_link "[" @punctuation.delimiter)
-(shortcut_link "]" @punctuation.delimiter)
-
 ([
   (info_string)
   (fenced_code_block_delimiter)
-  (code_span_delimiter)
-  (emphasis_delimiter)
 ] @conceal
 (#set! conceal ""))
-
-
-; Conceal inline links
-(inline_link
-  [
-    "["
-    "]"
-    "("
-   (link_destination)
-    ")"
-] @conceal
-(#set! conceal ""))
-
-
-; Conceal image links
-(image
-  [
-    "!"
-    "["
-    "]"
-    "("
-   (link_destination)
-    ")"
-] @conceal
-  (#set! conceal ""))
diff --git a/queries/markdown/injections.scm b/queries/markdown/injections.scm
index 6fa047ac9..c5aca6c54 100644
--- a/queries/markdown/injections.scm
+++ b/queries/markdown/injections.scm
@@ -1,9 +1,9 @@
 (fenced_code_block
   (info_string
     (language) @language)
-  (code_fence_content) @content)
+  (code_fence_content) @content (#exclude_children! @content))
 
 ((html_block) @html)
-((html_tag) @html)
 
-(document . (thematic_break) (_) @yaml @combined (thematic_break))
+(document . (section . (thematic_break) (_) @yaml @combined (thematic_break)))
+((inline) @markdown_inline (#exclude_children! @markdown_inline))
diff --git a/queries/markdown_inline/highlights.scm b/queries/markdown_inline/highlights.scm
new file mode 100644
index 000000000..495e6f1ab
--- /dev/null
+++ b/queries/markdown_inline/highlights.scm
@@ -0,0 +1,67 @@
+;; From MDeiml/tree-sitter-markdown
+[
+  (code_span)
+  (link_title)
+] @text.literal
+
+[
+  (emphasis_delimiter)
+  (code_span_delimiter)
+] @punctuation.delimiter
+
+(emphasis) @text.emphasis
+
+(strong_emphasis) @text.strong
+
+[
+  (link_destination)
+  (uri_autolink)
+] @text.uri
+
+[
+  (link_label)
+  (link_text)
+  (image_description)
+] @text.reference
+
+[
+  (backslash_escape)
+  (hard_line_break)
+] @string.escape
+
+; "(" not part of query because of
+; https://github.com/nvim-treesitter/nvim-treesitter/issues/2206
+; TODO: Find better fix for this
+(image ["!" "[" "]" "("] @punctuation.delimiter)
+(inline_link ["[" "]" "("] @punctuation.delimiter)
+(shortcut_link ["[" "]"] @punctuation.delimiter)
+
+([
+  (code_span_delimiter)
+  (emphasis_delimiter)
+] @conceal
+(#set! conceal ""))
+
+; Conceal inline links
+(inline_link
+  [
+    "["
+    "]"
+    "("
+   (link_destination)
+    ")"
+] @conceal
+(#set! conceal ""))
+
+
+; Conceal image links
+(image
+  [
+    "!"
+    "["
+    "]"
+    "("
+   (link_destination)
+    ")"
+] @conceal
+  (#set! conceal ""))
diff --git a/queries/markdown_inline/injections.scm b/queries/markdown_inline/injections.scm
new file mode 100644
index 000000000..5cf45dac2
--- /dev/null
+++ b/queries/markdown_inline/injections.scm
@@ -0,0 +1 @@
+((html_tag) @html)