Fontify doxygen support to 'c-ts-mode', 'c++-ts-mode' and 'java-ts-mode'

Add doxygen support to 'c-ts-mode', 'c++-ts-mode' and
'java-ts-mode' using tree-sitter-doxygen from
github.com/tree-sitter-grammars.
* lisp/progmodes/c-ts-common.el
(c-ts-mode-doxygen-comment-font-lock-settings): Add font locking
rules for doxygen comment.

* lisp/progmodes/c-ts-mode.el (c-ts-mode--feature-list): Add
'document' feature.
(c-ts-mode--doxygen-comment-regex): New regular expression for
doxygen comments.
(c-ts-mode, c++-ts-mode): Add support for doxygen parser.
* lisp/progmodes/java-ts-mode.el (java-ts-mode): Add support for
doxygen parser.  (Bug#71874)

3 files changed, 193 insertions, 114 deletions

diff --git a/lisp/progmodes/c-ts-common.el b/lisp/progmodes/c-ts-common.el
index 3882a697c48..a1f257ee09a 100644
--- a/lisp/progmodes/c-ts-common.el
+++ b/lisp/progmodes/c-ts-common.el
@@ -348,6 +348,28 @@ and /* */ comments.  SOFT works the same as in
       (delete-region (line-beginning-position) (point))
       (insert whitespaces)))))
 
+;; Font locking using doxygen parser
+(defvar c-ts-mode-doxygen-comment-font-lock-settings
+  (treesit-font-lock-rules
+   :language 'doxygen
+   :feature 'document
+   :override t
+   '((document) @font-lock-doc-face)
+
+   :language 'doxygen
+   :override t
+   :feature 'keyword
+   '((tag_name) @font-lock-constant-face
+     (storageclass) @font-lock-constant-face)
+
+   :language 'doxygen
+   :override t
+   :feature 'definition
+   '((tag (identifier) @font-lock-variable-name-face)
+     (function (identifier) @font-lock-function-name-face)
+     (function_link) @font-lock-function-name-face))
+  "Tree-sitter font lock rules for doxygen like comment styles.")
+
 ;;; Statement indent
 
 (defvar c-ts-common-indent-offset nil
diff --git a/lisp/progmodes/c-ts-mode.el b/lisp/progmodes/c-ts-mode.el
index e7f74fc53f2..a0f2bc33963 100644
--- a/lisp/progmodes/c-ts-mode.el
+++ b/lisp/progmodes/c-ts-mode.el
@@ -63,6 +63,9 @@
 ;; will set up Emacs to use the C/C++ modes defined here for other
 ;; files, provided that you have the corresponding parser grammar
 ;; libraries installed.
+;;
+;; If the tree-sitter doxygen grammar is available, then the comment
+;; blocks will be highlighted according to this grammar.
 
 ;;; Code:
 
@@ -539,7 +542,7 @@ NODE should be a labeled_statement.  PARENT is its parent."
 ;;; Font-lock
 
 (defvar c-ts-mode--feature-list
-  '(( comment definition)
+  '(( comment document definition)
     ( keyword preprocessor string type)
     ( assignment constant escape-sequence label literal)
     ( bracket delimiter error function operator property variable))
@@ -591,6 +594,10 @@ MODE is either `c' or `cpp'."
                       "LIVE_BUFFER" "FRAME"))
   "A regexp matching all the variants of the FOR_EACH_* macro.")
 
+(defvar c-ts-mode--doxygen-comment-regex
+  (rx (| "/**" "/*!" "//!" "///"))
+  "A regexp that matches all doxygen comment styles.")
+
 (defun c-ts-mode--font-lock-settings (mode)
   "Tree-sitter font-lock settings.
 MODE is either `c' or `cpp'."
@@ -1317,30 +1324,47 @@ in your init files."
     (when c-ts-mode-emacs-sources-support
       (treesit-parser-create 'c nil nil 'for-each))
 
-    (treesit-parser-create 'c)
-    ;; Comments.
-    (setq-local comment-start "/* ")
-    (setq-local comment-end " */")
-    ;; Indent.
-    (setq-local treesit-simple-indent-rules
-                (c-ts-mode--get-indent-style 'c))
-    ;; Font-lock.
-    (setq-local treesit-font-lock-settings (c-ts-mode--font-lock-settings 'c))
-    ;; Navigation.
-    (setq-local treesit-defun-tactic 'top-level)
-    (treesit-major-mode-setup)
-
-    ;; Emacs source support: handle DEFUN and FOR_EACH_* gracefully.
-    (when c-ts-mode-emacs-sources-support
-      (setq-local add-log-current-defun-function
-                  #'c-ts-mode--emacs-current-defun-name)
-
-      (setq-local treesit-range-settings
-                  (treesit-range-rules 'c-ts-mode--emacs-set-ranges))
-
-      (setq-local treesit-language-at-point-function
-                  (lambda (_pos) 'c))
-      (treesit-font-lock-recompute-features '(emacs-devel)))))
+    (let ((primary-parser (treesit-parser-create 'c)))
+      ;; Comments.
+      (setq-local comment-start "/* ")
+      (setq-local comment-end " */")
+      ;; Indent.
+      (setq-local treesit-simple-indent-rules
+                  (c-ts-mode--get-indent-style 'c))
+      ;; Font-lock.
+      (setq-local treesit-font-lock-settings
+                  (c-ts-mode--font-lock-settings 'c))
+      ;; Navigation.
+      (setq-local treesit-defun-tactic 'top-level)
+      (treesit-major-mode-setup)
+
+      ;; Emacs source support: handle DEFUN and FOR_EACH_* gracefully.
+      (when c-ts-mode-emacs-sources-support
+        (setq-local add-log-current-defun-function
+                    #'c-ts-mode--emacs-current-defun-name)
+
+        (setq-local treesit-range-settings
+                    (treesit-range-rules 'c-ts-mode--emacs-set-ranges))
+
+        (setq-local treesit-language-at-point-function
+                    (lambda (_pos) 'c))
+        (treesit-font-lock-recompute-features '(emacs-devel)))
+
+      ;; Inject doxygen parser for comment.
+      (when (treesit-ready-p 'doxygen)
+        (setq-local treesit-primary-parser primary-parser)
+        (setq-local treesit-font-lock-settings
+                    (append
+                     treesit-font-lock-settings
+                     c-ts-mode-doxygen-comment-font-lock-settings))
+        (setq-local treesit-range-settings
+                    (treesit-range-rules
+                     :embed 'doxygen
+                     :host 'c
+                     :local t
+                     `(((comment) @cap
+                        (:match
+                         ,c-ts-mode--doxygen-comment-regex @cap)))))))))
 
 (derived-mode-add-parents 'c-ts-mode '(c-mode))
 
@@ -1368,24 +1392,40 @@ recommended to enable `electric-pair-mode' with this mode."
   :after-hook (c-ts-mode-set-modeline)
 
   (when (treesit-ready-p 'cpp)
-
-    (treesit-parser-create 'cpp)
-
-    ;; Syntax.
-    (setq-local syntax-propertize-function
-                #'c-ts-mode--syntax-propertize)
-
-    ;; Indent.
-    (setq-local treesit-simple-indent-rules
-                (c-ts-mode--get-indent-style 'cpp))
-
-    ;; Font-lock.
-    (setq-local treesit-font-lock-settings (c-ts-mode--font-lock-settings 'cpp))
-    (treesit-major-mode-setup)
-
-    (when c-ts-mode-emacs-sources-support
-      (setq-local add-log-current-defun-function
-                  #'c-ts-mode--emacs-current-defun-name))))
+    (let ((primary-parser (treesit-parser-create 'cpp)))
+
+      ;; Syntax.
+      (setq-local syntax-propertize-function
+                  #'c-ts-mode--syntax-propertize)
+
+      ;; Indent.
+      (setq-local treesit-simple-indent-rules
+                  (c-ts-mode--get-indent-style 'cpp))
+
+      ;; Font-lock.
+      (setq-local treesit-font-lock-settings
+                  (c-ts-mode--font-lock-settings 'cpp))
+      (treesit-major-mode-setup)
+
+      (when c-ts-mode-emacs-sources-support
+        (setq-local add-log-current-defun-function
+                    #'c-ts-mode--emacs-current-defun-name))
+
+      ;; Inject doxygen parser for comment.
+      (when (treesit-ready-p 'doxygen)
+        (setq-local treesit-primary-parser primary-parser)
+        (setq-local treesit-font-lock-settings
+                    (append
+                     treesit-font-lock-settings
+                     c-ts-mode-doxygen-comment-font-lock-settings))
+        (setq-local treesit-range-settings
+                    (treesit-range-rules
+                     :embed 'doxygen
+                     :host 'cpp
+                     :local t
+                     `(((comment) @cap
+                        (:match
+                         ,c-ts-mode--doxygen-comment-regex @cap)))))))))
 
 (derived-mode-add-parents 'c++-ts-mode '(c++-mode))
 
diff --git a/lisp/progmodes/java-ts-mode.el b/lisp/progmodes/java-ts-mode.el
index 4ceb211ade1..68ead567632 100644
--- a/lisp/progmodes/java-ts-mode.el
+++ b/lisp/progmodes/java-ts-mode.el
@@ -24,6 +24,8 @@
 
 ;;; Commentary:
 ;;
+;; If the tree-sitter doxygen grammar is available, then the comment
+;; blocks will be highlighted according to this grammar.
 
 ;;; Code:
 
@@ -312,7 +314,7 @@ Return nil if there is no name or if NODE is not a defun node."
 
 
 (defvar java-ts-mode--feature-list
-  '(( comment definition )
+  '(( comment document definition )
     ( constant keyword string type)
     ( annotation expression literal)
     ( bracket delimiter operator)))
@@ -326,76 +328,91 @@ Return nil if there is no name or if NODE is not a defun node."
   (unless (treesit-ready-p 'java)
     (error "Tree-sitter for Java isn't available"))
 
-  (treesit-parser-create 'java)
-
-  ;; Comments.
-  (c-ts-common-comment-setup)
-
-  ;; Indent.
-  (setq-local c-ts-common-indent-type-regexp-alist
-              `((block . ,(rx (or "class_body"
-                                  "array_initializer"
-                                  "constructor_body"
-                                  "annotation_type_body"
-                                  "interface_body"
-                                  "lambda_expression"
-                                  "enum_body"
-                                  "switch_block"
-                                  "record_declaration_body"
-                                  "block")))
-                (close-bracket . "}")
-                (if . "if_statement")
-                (else . ("if_statement" . "alternative"))
-                (for . "for_statement")
-                (while . "while_statement")
-                (do . "do_statement")))
-  (setq-local c-ts-common-indent-offset 'java-ts-mode-indent-offset)
-  (setq-local treesit-simple-indent-rules java-ts-mode--indent-rules)
-
-  ;; Electric
-  (setq-local electric-indent-chars
-              (append "{}():;," electric-indent-chars))
-
-  ;; Navigation.
-  (setq-local treesit-defun-type-regexp
-              (regexp-opt '("method_declaration"
-                            "class_declaration"
-                            "record_declaration"
-                            "interface_declaration"
-                            "enum_declaration"
-                            "import_declaration"
-                            "package_declaration"
-                            "module_declaration"
-                            "constructor_declaration")))
-  (setq-local treesit-defun-name-function #'java-ts-mode--defun-name)
-
-  (setq-local treesit-thing-settings
-              `((java
-                (sexp ,(rx (or "annotation"
-                               "parenthesized_expression"
-                               "argument_list"
-                               "identifier"
-                               "modifiers"
-                               "block"
-                               "body"
-                               "literal"
-                               "access"
-                               "reference"
-                               "_type"
-                               "true"
-                               "false")))
-                (sentence ,(rx (or "statement"
-                                   "local_variable_declaration"
-                                   "field_declaration"
-                                   "module_declaration"
-                                   "package_declaration"
-                                   "import_declaration")))
-                (text ,(regexp-opt '("line_comment"
-                                     "block_comment"
-                                     "text_block"))))))
-
-  ;; Font-lock.
-  (setq-local treesit-font-lock-settings java-ts-mode--font-lock-settings)
+  (let ((primary-parser (treesit-parser-create 'java)))
+
+    ;; Comments.
+    (c-ts-common-comment-setup)
+
+    ;; Indent.
+    (setq-local c-ts-common-indent-type-regexp-alist
+                `((block . ,(rx (or "class_body"
+                                    "array_initializer"
+                                    "constructor_body"
+                                    "annotation_type_body"
+                                    "interface_body"
+                                    "lambda_expression"
+                                    "enum_body"
+                                    "switch_block"
+                                    "record_declaration_body"
+                                    "block")))
+                  (close-bracket . "}")
+                  (if . "if_statement")
+                  (else . ("if_statement" . "alternative"))
+                  (for . "for_statement")
+                  (while . "while_statement")
+                  (do . "do_statement")))
+    (setq-local c-ts-common-indent-offset 'java-ts-mode-indent-offset)
+    (setq-local treesit-simple-indent-rules java-ts-mode--indent-rules)
+
+    ;; Electric
+    (setq-local electric-indent-chars
+                (append "{}():;," electric-indent-chars))
+
+    ;; Navigation.
+    (setq-local treesit-defun-type-regexp
+                (regexp-opt '("method_declaration"
+                              "class_declaration"
+                              "record_declaration"
+                              "interface_declaration"
+                              "enum_declaration"
+                              "import_declaration"
+                              "package_declaration"
+                              "module_declaration"
+                              "constructor_declaration")))
+    (setq-local treesit-defun-name-function #'java-ts-mode--defun-name)
+
+    (setq-local treesit-thing-settings
+                `((java
+                   (sexp ,(rx (or "annotation"
+                                  "parenthesized_expression"
+                                  "argument_list"
+                                  "identifier"
+                                  "modifiers"
+                                  "block"
+                                  "body"
+                                  "literal"
+                                  "access"
+                                  "reference"
+                                  "_type"
+                                  "true"
+                                  "false")))
+                   (sentence ,(rx (or "statement"
+                                      "local_variable_declaration"
+                                      "field_declaration"
+                                      "module_declaration"
+                                      "package_declaration"
+                                      "import_declaration")))
+                   (text ,(regexp-opt '("line_comment"
+                                        "block_comment"
+                                        "text_block"))))))
+
+    ;; Font-lock.
+    (setq-local treesit-font-lock-settings
+                java-ts-mode--font-lock-settings)
+
+    ;; Inject doxygen parser for comment.
+    (when (treesit-ready-p 'doxygen)
+      (setq-local treesit-primary-parser primary-parser)
+      (setq-local treesit-font-lock-settings
+                  (append treesit-font-lock-settings
+                          c-ts-mode-doxygen-comment-font-lock-settings))
+      (setq-local treesit-range-settings
+                  (treesit-range-rules
+                   :embed 'doxygen
+                   :host 'java
+                   :local t
+                   `(((block_comment) @cap (:match "/\\*\\*" @cap)))))))
+
   (setq-local treesit-font-lock-feature-list java-ts-mode--feature-list)
 
   ;; Imenu.