From 31c966ece24da893b3a1d16f6c1d959ee3bf5ddd Mon Sep 17 00:00:00 2001
From: Max Brunsfeld <maxbrunsfeld@gmail.com>
Date: Wed, 4 Oct 2017 09:24:51 -0700
Subject: [PATCH] Handle multi-line comments with asterisks at the end

---
 corpus/statements.txt | 22 ++++++++++++++++++++++
 grammar.js            | 12 +++++-------
 src/grammar.json      | 18 +++---------------
 src/parser.c          |  4 ++++
 4 files changed, 34 insertions(+), 22 deletions(-)

diff --git a/corpus/statements.txt b/corpus/statements.txt
index c2a1eee..c52025c 100644
--- a/corpus/statements.txt
+++ b/corpus/statements.txt
@@ -109,3 +109,25 @@ recur:
           (identifier)
           (call_expression (field_expression (identifier) (field_identifier)) (argument_list)))))
       (if_statement (identifier) (goto_statement (statement_identifier))))))
+
+============================================
+Comments with asterisks
+============================================
+
+/*************************
+ * odd number of asterisks
+ *************************/
+int a;
+
+/**************************
+ * even number of asterisks
+ **************************/
+int b;
+
+---
+
+(translation_unit
+  (comment)
+  (declaration (type_identifier) (identifier))
+  (comment)
+  (declaration (type_identifier) (identifier)))
diff --git a/grammar.js b/grammar.js
index 905f0e0..abbca8f 100644
--- a/grammar.js
+++ b/grammar.js
@@ -706,17 +706,15 @@ module.exports = grammar({
       ')'
     ),
 
+    // http://stackoverflow.com/questions/13014947/regex-to-match-a-c-style-multiline-comment/36328890#36328890
     comment: $ => token(choice(
-      seq(
-        '//',
-        /.*/
-      ),
+      seq('//', /.*/),
       seq(
         '/*',
-        repeat(choice(/[^\*]/, /\*[^/]/)),
-        '*/'
+        /[^*]*\*+([^/*][^*]*\*+)*/,
+        '/'
       )
-    ))
+    )),
   }
 });
 
diff --git a/src/grammar.json b/src/grammar.json
index eb30b93..8880701 100644
--- a/src/grammar.json
+++ b/src/grammar.json
@@ -3578,24 +3578,12 @@
                 "value": "/*"
               },
               {
-                "type": "REPEAT",
-                "content": {
-                  "type": "CHOICE",
-                  "members": [
-                    {
-                      "type": "PATTERN",
-                      "value": "[^\\*]"
-                    },
-                    {
-                      "type": "PATTERN",
-                      "value": "\\*[^\\/]"
-                    }
-                  ]
-                }
+                "type": "PATTERN",
+                "value": "[^*]*\\*+([^\\/*][^*]*\\*+)*"
               },
               {
                 "type": "STRING",
-                "value": "*/"
+                "value": "/"
               }
             ]
           }
diff --git a/src/parser.c b/src/parser.c
index 5f2bff5..58dcef2 100644
--- a/src/parser.c
+++ b/src/parser.c
@@ -2053,6 +2053,8 @@ static bool ts_lex(TSLexer *lexer, TSStateId state) {
         ADVANCE(94);
       END_STATE();
     case 95:
+      if (lookahead == '*')
+        ADVANCE(95);
       if (lookahead == '/')
         ADVANCE(96);
       if (lookahead != 0)
@@ -3866,6 +3868,8 @@ static bool ts_lex(TSLexer *lexer, TSStateId state) {
       ACCEPT_TOKEN(sym_preproc_arg);
       if (lookahead == '\n')
         ADVANCE(94);
+      if (lookahead == '*')
+        ADVANCE(287);
       if (lookahead == '/')
         ADVANCE(96);
       if (lookahead == '\\')