Merge pull request #251 from susliko/nested-multiline-comments

Nested block comments

corpus/comments.txt

@@ -0,0 +1,33 @@
+================================================================================
+Single line comments
+================================================================================
+
+// comment 1
+// comment 2
+
+--------------------------------------------------------------------------------
+
+(compilation_unit
+  (comment)
+  (comment))
+
+================================================================================
+Block comments
+================================================================================
+/**/
+/** comment 1
+ * /* comment 2
+ *  /* / * * /comment 3 */
+ // comment 4
+ * @param
+ *  */
+*/
+
+--------------------------------------------------------------------------------
+
+(compilation_unit
+  (block_comment)
+  (block_comment
+    (block_comment
+      (block_comment)
+      (comment))))

corpus/expressions.txt

@@ -76,7 +76,7 @@ class C {
                   (identifier)
                   (operator_identifier)
                   (integer_literal)))))
-          (comment)
+          (block_comment)
           (call_expression
             (field_expression
               (identifier)
@@ -813,7 +813,7 @@ class C {
               (identifier)
               (identifier))
             (comment)
-            (comment)
+            (block_comment)
             (identifier)))))))
 
 ================================================================================
@@ -895,6 +895,7 @@ class C {
   def main() {
     a = b max c
     d + e + f
+    a ** b
   }
 }
 
@@ -920,6 +921,10 @@ class C {
               (operator_identifier)
               (identifier))
             (operator_identifier)
+            (identifier))
+          (infix_expression
+            (identifier)
+            (operator_identifier)
             (identifier)))))))
 
 ================================================================================

grammar.js

@@ -27,7 +27,8 @@ module.exports = grammar({
 
   extras: $ => [
     /\s/,
-    $.comment
+    $.comment,
+    $.block_comment,
   ],
 
   supertypes: $ => [
@@ -1293,7 +1294,7 @@ module.exports = grammar({
     wildcard: $ => '_',
 
     /**
-     * Regex patterns created to avoid matching // comments.
+     * Regex patterns created to avoid matching // comments and /* comment starts.
      * This could technically match illeagal tokens such as val ?// = 1
      */
     operator_identifier: $ => token(choice(
@@ -1308,8 +1309,8 @@ module.exports = grammar({
       seq(
         // opchar
         /[\-!#%&*+\/\\:<=>?@\u005e\u007c~\p{Sm}\p{So}]/,
-        // opchar minus slash
-        /[\-!#%&*+\\:<=>?@\u005e\u007c~\p{Sm}\p{So}]/,
+        // opchar minus slash and asterisk
+        /[\-!#%&+\\:<=>?@\u005e\u007c~\p{Sm}\p{So}]/,
         // opchar*
         repeat(/[\-!#%&*+\/\\:<=>?@\u005e\u007c~\p{Sm}\p{So}]/),
       ),
@@ -1519,14 +1520,14 @@ module.exports = grammar({
       repeat1($.guard),
     ),
 
-    comment: $ => token(choice(
-      seq('//', /.*/),
-      seq(
-        '/*',
-        /[^*]*\*+([^/*][^*]*\*+)*/,
-        '/'
-      )
-    ))
+    comment: $ => token(seq('//', /.*/)),
+
+    block_comment: $ => seq(
+      token('/*'),
+      repeat(token(/./)),
+      token('*/')
+    ),
+    
   }
 })
 

script/smoke_test.sh

@@ -4,7 +4,7 @@
 
 SCALA_SCALA_LIBRARY_EXPECTED=94
 SCALA_SCALA_COMPILER_EXPECTED=84
-DOTTY_COMPILER_EXPECTED=80
+DOTTY_COMPILER_EXPECTED=81
 SYNTAX_COMPLEXITY_CEILING=2300
 
 if [ ! -d "$SCALA_SCALA_DIR" ]; then